[2018] 12 S.C.R. 495 495 RAJASTHAN CYLINDERS AND CONTAINERS LIMITED A v. UNION OF INDIA AND ANOTHER (Civil Appeal No. 3546 of 2014) OCTOBER 01, 2018 B [A. K. SIKRI AND ASHOK BHUSHAN, JJ.] Competition Act, 2002: ss. 3, 19 and 27 – Cartelisation, bid-rigging and collusive bidding – Suo-motu proceedings initiated by Competition Commission C of India – Against bidders/manufacturers of LPG Gas Cylinders – Investigating into complaint about unfair conditions in the tender floated by Indian Oil Corporation Ltd. (IOCL) for supply of LPG Gas Cylinders – Competition Commission held that there was collusive bidding – Competition Appellate Tribunal affirmed the findings of D Competition Commission – On appeal, held: It is duty of the Commission to ensure that the conditions which have tendency to kill the competition are to be curbed – There may not be direct evidence on the basis of which cartelisation or such agreement between the parties can be proved – The standard of proof for such agreement is one of probability – There is a presumption that four E types of agreements mentioned in s. 3(3) will have an appreciable effect on competition – However, the presumption is rebuttable as these agreements are not conclusive proof of the fact that it would result in appreciable adverse effects on competition – If evidence is led which rebuts the presumption, the Commission shall take into F

consideration the factors mentioned in s. 19 – If the evidence collected by the Commission leads to one or more or all the factors mentioned in s. 19(3), it would again be treated as an agreement which may cause or likely to cause an appreciable adverse effect on competition – In the present case, inferences drawn by the Commission, on the basis of the evidence collected by it, have been G duly rebutted by the appellants/manufacturers – They have been able to discharge the onus that shifted upon them – However, at that stage, the Commission failed to carry the matter further – Thus, there is no sufficient evidence to hold that there was any agreement between the appellants for bid rigging – Appeals by the manufactures H 495 496 SUPREME COURT REPORTS [2018] 12 S.C.R. A are allowed – Since no penalty is payable, appeals of the Commission are dismissed as infructuous – Evidence. Words and Phrases: “bid rigging” and “collusive rigging” – Meaning of, in the context of Competition Act, 2002. B Allowing the appeals filed by the manufactures and dismissing the appeals filed by the Competition Commission, the Court

HELD: 1. On the one hand the economic policy of the nation has ushered in the era of liberalisation and globalisation thereby C giving freeplay to the private sector in the manner of conducting business, at the same time, in public interest and in the interest of consumers, a regime of regulators has also been brought to ensure certain checks and balances. Since competition among the enterprises or businessmen is treated as service for a public D purpose and, therefore, there is a need to curb anti-competitive practices. The Competition Commission of India (CCI) is given the task (as a regulator) to ensure that no such anti-competitive practices are undertaken. In fact, Section 18 of the Act casts a specific and positive obligation on CCI to ‘eliminate’ anti-competitive practices and promote competition, interest of the E consmuer and free trade. [Para 72] [536-G-H; 537-A-B] Competition Commission of India vs. Steel Authority of India Limited and Another (2010) 10 SCC 744 : [2010] 11 SCR 112 – relied on. F 2. One of the anti-competitive practices is cartelisation, the essential postulate whereof is agreement between enterprises or association of enterprises or persons or associations of persons in respect of production, supply, distribution, storage, acquisition or control of goods or provisions of service, which causes or is likely to cause an appreciable adverse effect on competition within G

India. Such an agreement is treated as void. The types of agreement which may fall foul of Section 3 are mentioned in sub-section (3) thereof. These include sharing the market by way of allocation of geographical areas of market [clause (c)] and the agreements which result in bid-rigging or collusive bidding H whether directly or indirectly [clause (d)]. There is a presumption RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 497 UNION OF INDIA AND ANOTHER that four types of agreements mentioned in sub-section (3) will A have an appreciable adverse effect on competition. [Para 73] [537-F-H]

3. Section 19(3) of the Act mentions the factors which are to be examined by the CCI while determining whether an agreement has an appreciable adverse effect on competition under B Section 3. However, this inquiry would be needed in those cases which are not covered by clauses (a) to (d) of sub-section (3) of Section 3. Reason is simple. The agreeements of nature mentioned in sub-section (3) are presumed to have an appreciable effect and, therefore, no further exercise is needed by the CCI Conce a finding is arrived at, that a particular agreement fell in any of the aforesaid four categories. Agreements mentioned in Section 3(3) raise a presumption that such agreements shall have an appreciable adverse effect on competition. It follows, as a fortiorari, that the presumption is rebuttable as these agreements are not treated as conclusive proof of the fact that it would result D in appreciable adverse effect on competition. What follows is that once the CCI finds that case is covered by one or more of the clauses mentioned in sub-section (3) of Section 3, it need not undertake any further enquiry and burden would shift upon such enterprises or persons etc. to rebut the said presumption by Eleading adequate evidence. In case such an evidence is led, which dispels the presumption, then the CCI shall take into consideration the factors mentioned in Section 19 of the Act and to see as to whether all or any of these factors are established. If the evidence collected by the CCI leads to one or more or all factors mentioned in Section 19(3), it would again be treated as F an agreement which may cause or is likely to cause an appreciable adverse effect of competition, thereby compelling the CCI to take further remedial action in this behalf as provided under the Act. That, is the broad scheme when Sections 3 and 19 are to be read in conjuction. [Para 74] [538-A-F]

G 4. Explanation to Section 3, assigns meaning to ‘bid rigging’. The necessary ingredients of bid rigging, are: (a) agreement between the parties; (b) these parties are engaged in idential or similar production or trading of goods or provisions of services; and (c) the agreement has the effect of eliminating or reducing competition of bids or adversely affect or manipulating H 498 SUPREME COURT REPORTS [2018] 12 S.C.R. A the process for bidding. Though the expression ‘collusive bidding’ is not defined in the Act, it appears that both ‘bid rigging’ and ‘collusive bidding’ are overlapping concepts. [Paras 75, 77] [538-G; 539-B-C]

5. Having regard to the scheme of the Act, it cannot be said that there is no possibility of a competition in the present B cases and, therefore, CCI had no jurisdiction to carry out any such investigation. Section 3 prohibits anti-competitive agreements and brings about the prime objective of the Competition Act. The purpose of the Act is not only to illuminate practices having adverse effect on the competition but also to C promote and sustain competition in the market. Enforcement provides remedies to avoid situation that will lead to decrease competition in the market. Therefore, effective enforcement is important not only to sanction anti-competitive conduct but also to deter future competitive practices. In the present case itself, there are sixty suppliers of the product for which there are three D buyers. After all, each supplier would like to be L-1 or L-2 so that it is able to get order for larger quantities than the other. In this sense, there would be a competition among them. Further, it would also be in the interest of the buyers like IOCL etc. that the elements of healthy competition persists in the market. [Paras E 78 and 79] [543-D; 544-C-E] 6. It is the duty of the CCI to ensure that the conditions which have tendency to kill the competition are to be curbed. It is also the function of the CCI to ensure that there is a competition so that benefits of such competition are reaped by F the consumers. [Para 79] [544-E]

7. The CCI and Competition Appellate Tribunal (COMPAT) have held that there was collusive bidding in the present case in view of the fact that there is an active trade association of the suppliers; a meeting took place couple of days before the date of bidding; common changes were pointed out by these appellants G who submitted bids on their behalf; and bids were of identical amounts despite varying cost, which were repetitive in nature. The respondents may be right in their submission that there may not be a direct evidence on the basis of which cartelisation or such agreement between the parties can be proved as these H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 499 UNION OF INDIA AND ANOTHER agreements are normally entered into in closed doors. The A standard of proof which is required is one of probability. [Para 80] [544-G-H; 545-A] Technip SA v. SMS Holding (P) Ltd. & Ors. (2205) 5 SCC 465 : [2005] 1 Suppl. SCR 223; Commissioner of Income Tax, Bombay City I, Bombay v. Jubilee Mills B Ltd., Bombay (1963) 48 ITR 9 – relied on. 8. Even in the absence of proof of concluded formal agreement, when there are indicators that there was practical co-operation between the parties which knowingly substitute the risk of competition, that would amount to anti-competitive C practices.[Para 83] [549-D]

9. The manner in which tendering process takes place would show that in such a competitive scenario, the bid which the different bidder would be submitting becomes obvious. It has come on record that just a few days before the tender in D question, another tender was floated by BPCL and on opening of the said tender the rates of L-1, L-2 etc. came to be known. In a scenario like this, that obviously becomes a guiding factor for the bidders to submit their bids. Keeping in mind the aforesaid fact situation, the very factors on the basis of which the CCI has Ecome to the conclusion that there was cartelisation, in fact, become valid explanations to the indicators pointed out by the CCI. [Paras 90 and 91] [555-F-G]

10. Identical products along with market conditions for which there would be only three buyers, in fact, would go in favour of the appellants. The factor of repetitive bidding, though appears F to be a factor against the appellants, was also possible in the aforesaid scneario. The prevailing conditions in fact rule out the possibility of much price variations and all the manufacturers are virtually forced to submit their bid with a price that is quite close to each other. Therefore, it became necessary to sustain G themselves in the market. Hence, the factor that these suppliers are from different region having different cost of manufacture would lose its significance. It is a situation where prime condition is to quote the price at which a particular manufacturer can bag H 500 SUPREME COURT REPORTS [2018] 12 S.C.R. A an order even when its manufacturing cost is more than the manufacturing cost of others. The main purpose for such a manufacuring would be to remain in the fray and not to lose out. Therefore, it would be ready to accept lesser margin. This would answer why there were near identical bids despite varying cost. [Para 91] [556-A-D] B

11. Insofar as meeting of bidders in Mumbai just before the date of submission of tender is concerned, some aspects pointed out by the appellants are not considered by the CCI or the COMPAT at all. No doubt, the meeting took place a couple of days before the date of tender. No doubt, the absence of agenda C coming on record would not make much difference. However, only 19 appellants had attended that meeting. Many others were not even members or did not attend the meeting. In spite thereof, even they quoted almost same rates as the one who attended the meeting. This would lead the Court to the inference that reason D for quoting similar price was not the meeting but something else. [Para 92] [556-D-E] 12. Monopsony consists of a market with a single buyer. When there are only few buyers the market is described as an oligopsony. In such a situation, a manufacturer with no buyers E will have to exit from the trade. Therefore, first condition of oligopsony stands fulfilled. The other condition for the existence of oligopsony is whether the buyers have some influence over the price of their inputs. It is also to be seen as to whether the seller has any ability to raise prices or it stood reduced/eliminated by the aforesaid buyers. [Para 95] [558-G-H; 559-A] F

13. On a hollistic view of the matter, it is clear that the appellants have been able to discharge the onus by referring to various indicators which go on to show that parallel behaviour was not the result of any concerted practice. [Para 96] [559-A-B] 14. Parallel behaviour does not, by itself, amount to a G concerted practice, though it may provide a strong evidence of such a practice. Nevertheless, it is a strong evidence of such a practice. However, before such an inference is drawn it has to be seen that this parallel behaviour has led to conditions of competition which do not correspond to the normal conditions of H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 501 UNION OF INDIA AND ANOTHER the market, having regard to the nature of the products, size and A volume of the undertaking of the said market. [Para 97] [559-B-C] 15. Whenever there is a situation of oligopsony, parallel pricing simplicitor would not lead to the conclusion that there was a concerted practice there has to be other credible and corroborative evidence to show that in an oligopoly a reduction B in price would swiftly attract the customers of the other two or three rivals, the effect upon whom would be so devastating that they would have to react by matching the cut. [Para 97] [559-D-E]

16. The inferences drawn by the CCI on the basis of evidence collected by it are duly rebutted by the appellants and C the appellants have been able to discharge the onus that shifted upon them on the basis of factors pointed out by the CCI. However, at that stage, the CCI failed to carry the matter further by having required and necessary inquiry that was needed in the instant case. [Para 102] [563-F-G] D 17. In such a watertight tender policy of IOCL which gave IOCL full control over the tendering process, it was necessary to summon IOCL. This would have cleared many aspects which are shrouded in mystery and the dust has not been cleared. [Para 103] [563-G-H] E 18. Thus, there is no sufficient evidence to hold that there was any agreement between the appellants for bid rigging. As a consequence, since no penalty is payable, appeals of the CCI are rendered infructuous and dismissed as such. [Para 104] [564-A-B] Union of India vs. Hindustan Development Corporation F (1993) 3 SCC 499 : [1993] 3 SCR 128; Ashoka Smokeless Coal India (P) Ltd. v. Union of India (2007) 2 SCC 640 : [2006] 9 Suppl. SCR 954; Excel Crop Care Limited v. Competition Commission of India & Anr. (2017) 8 SCC 47 : [2017] 5 SCR 901; Union of India v. Hindustan Development Corporation (1993) 1 SCC G 467 : [1993] 3 SCR 108; Union of India v. Hindustan Development Coproration (1993) 3 SCC 499 : [1993] 3 SCR 128; Punjab Land Developement & Reclamation Corporation Ltd. v. Presiding Officer, Labour Court H

502 SUPREME COURT REPORTS [2018] 12 S.C.R. A (1990) 3 SCC 682 : [1990] 3 SCR 111; S. Sundaram Pillai v. V.R. Pattabiraman (1985) 1 SCC 591 : [1985] 2 SCR 643; CCI v. Artistes & Technicians of W.B. Film & Television (2017) 5 SCC 17 : [2017] 5 SCR 1; Union of India vs. Hindustan Development Corporation (1993) 3 SCC 499 : [1993] 3 SCR 128 – referred to. B Ahlstrom Osakeyhtio v. Commission 31.3.1993, ECJ (“Woodpulp”); Theatre Enterprises v. Paramount Films 346 US 357; Monsanto Co. v. Spray-Rite Service Corp. 346 US 357; Monsanto Co. v. Spray-Rite Service Corp. 465 U.S. 752, 104 S.Ct. 1464, 79 L.Ed. 2Nd 775 (1984); C Matsushita v. Zenith Ratio Corp. 475 U.S.574 (1986) Bell Atlantic Corp v. Twombly 550 U.S. 544 – referred to. Case Law Reference [1993] 3 SCR 128 referred to Para 14 D [2006] 9 Suppl. SCR 954 referred to Para 27 [2017] 5 SCR 901 referred to Para 28 [1993] 3 SCR 108 referred to Para 32 E [1993] 3 SCR 128 referred to Para 33 [1990] 3 SCR 111 referred to Para 43 [1985] 2 SCR 643 referred to Para 45 [2017] 5 SCR 1 referred to Para 48 F [1993] 3 SCR 128 referred to Para 52 [2010] 11 SCR 112 referred to Para 72 [2005] 1 Suppl. SCR 223 relied on Para 80 (1963) 48 ITR 9 relied on Para 81 G CIVIL APPELLATE JURISDICTION: Civil Appeal No. 3546 of 2014 From the Judgment and Order dated 20.12.2013 of the Competition Appellate Tribunal in Appeal No. 59 of 2012. H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 503 UNION OF INDIA AND ANOTHER WITH A

Civil Appeal Nos. 4280/2014, 4346/2014, 4649/2014, 4342/2014, 4879/2014, 4868/2014, 6033/2014, 5771/2014, 5772/2014, 5035/2014, 5773/2014, 5649/2014, 5650/2014, 5651/2014, 4972/2014, 6661/2014, 7214/2014, 7102/2014, 6365/2014, 6025/2014, 5993-5994/2014, 6868/2014, 5774/2014, 5775/2014, 5776/2014, 5832-5833/2014, 5777/2014, 5778/2014, B 6371/2014, 8953/2014, 6372/2014, 6373/2014, 6366/2014, 6367/2014, 6374/2014, 6368/2014, 6364/2014, 6369/2014, 6370/2014, 10579/2014, 1724/2015, 5277-5278/2016, 5281-5315/2016, 7359/2016.

Salman Khurshid, Prashanto Sen, Sr. Advs., Manan Verma, Ms. Nidhi Khanna, Mrs. Sarla Chandra, Ms. Diksha Rai, Arjun Krishnan, O. C P. Gaggar, Aditya Gaggar, Sudipto Sircar, Annam D. N. Rao, Annam Venkatesh, Rahul Mishra, Amol Sinha, Ms. Anshum Jain, Rahul Kochar, Naveen Kumar, Deepak Anand, N. S. Nandakumar, N. Krishn Kumar, K. Krishna Kumar, Sarad Kumar Singhania, Ms. Ruchi Kohli, Ms. Pallavi Langar, Jaiveer Shergill, Amit Goyal, Rajesh Sharma, Ms. Nidhi Singh DDubey, Shafiq Khan, Ms. Shalu Sharma, R. Satish Kumar, V. Susheetha, P. V. Yogeswaran, Rameshwar Prasad Goyal, S. Nanda Kumar, R. Sathis Kumar, M. S. Saron Kumar, Ms. N. Deepika Nanda Kumar, Naresh Kumar, Praveen Mahajan, Arjun Krishnan, Ankur Singh, Sumit Srivastava, Kaustav Som, Ms. Gitanjali Kapur, Antony Julian, Arpit Shukla, Udayan Verma, Sarvesh Mishra, Amit Sharma, Raj Bahadur, Ms. Anil Katiyar, E Kedar Nath Tripathy, Girija Ballav Das, B. B. Pradhan, Gaurav Agrawal, Manan Verma, Ms. Diksha Rai, Ms. Palak Mahajan, Kuljeet Rawal, Ashwani Kumar, Ms. Iti Sharma, Pradeep Aggarwal, Karan Khanna, Aniket Bhattacharya, Arjun Agggarwal, Advs. for the appearing parties. The Judgment of the Court was delivered by F A. K. SIKRI, J. 1. All these appeals are filed against the orders dated 20th December, 2013 passed by the Competition Appellate Tribunal (hereinafter referred to as ‘COMPAT’). The COMPAT by the said judgment has upheld the findings of the Competition Commission of India (for short, ‘CCI’) that the appellants/suppliers of Liquefied Petroleum G Gas (LPG) Cylinders to the Indian Oil Corporation Ltd. (for short, ‘IOCL’) had indulged in cartilisation, thereby influencing and rigging the prices, thus, violating the provisions of Section 3(3)(d) of the Competition Act, 2002 (for short, the ‘Act’). The CCI, as a result, imposed severe penalties in the form of fines under Section 27 of the Act. While maintaining the H

504 SUPREME COURT REPORTS [2018] 12 S.C.R. A order of the CCI insofar as it found the appellants guilty of contravention of Section 3(3)(d) and also under Section 3(3)(a) of the Act, the COMPAT has reduced the amount of penalty. These suppliers have filed the instant appeals on the ground that there was no cartilisation and they have not contravened the provisions of the Act. On the other hand, CCI has also come up in appeal challenging latter part of the order B whereby penalties inflicted on the suppliers stand reduced. For the sake of convenience these suppliers will be referred to as the appellants hereinafter.

2. We may point out at the outset that all these appellants are manufacturing gas cylinders of a particular specification having capacity C of 14.2 kg which are needed for use by the three oil companies in India, namely, IOCL, Bharat Petroleum Corporation Ltd. (BPCL) and Hindustan Petroleum Corporation Ltd. (HPCL) [all public sector companies]. It is also a matter of record that apart from the aforesaid three companies there are no other buyers for these cylinders D manufactured by the appellants. Insofar as IOCL is concerned, it is a leading market player in LPG as its market share is 48%. Thus, in case a particular manufacturer is not able to supply its cylinders to the aforesaid three companies, there is no other market for these cylinders and it may force that company to exit from its operations. We may also point out at this stage itself that inquiry was started against 47 companies. The CCI E exonerated two companies and found that 45 companies had entered into an arrangement/agreement insofar as statements of bids pursuant to tenders issued by IOCL are concerned. Out of these 45 companies one did not challenge the orders before the COMPAT and other 44 had filed appeals which have been decided by the COMPAT.

F 3. The manner in which the inquiry was undertaken by the CCI, culminating into the finding of guilt and imposition of penalty, is succintly and sequally recorded by the COMPAT in its impugned order. As there is no dispute about the said factual narration, it would be convenient to borrow the said discussion as recorded by the COMPAT. G 4. The suo-motu proceedings were started by the CCI on the basis of the information received by it in Case No. 10 of 2010 titled M/ s. Pankaj Gas Cylinders Ltd. Vs. Indian Oil Corporation Ltd. in that case a complaint was made by M/s. Pankaj Gas Cylinders before the CCI complaining about unfair conditions in the tender floated by IOCL H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 505 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] for the supply of 105 lakh 14.2 Kg. capacity LPG Cylinders with SC A valves in the year 2010-11, the tender No. being LPG-O/M/PT-03/09- 10. While considering the Director General’s investigation report in Case No. 10 of 2010, the CCI in pursuance of its duties under Section 18 felt that investigation was necessary in the case of all bidders who were the suppliers of 14.2 kg. LPG cylinders in that tender. In the investigation B

report in the said case, the Director General had noted that out of 63 bidders who participated in the tender, 50 bidders were qualified for opening of price bids, while 12 bidders were qualified as new vendors who were not required to submit price bids and one bidder was not qualified for the opening of the price bid. The technical bid of the subject tender was opened on 3.3.2010 and the price bids of 50 qualified bidders C were opened on 23.3.2010. According to the Director General, there was a similar pattern in the bids by all the 50 bidders who submitted price bids for various States. The bids of a large number of parties were exactly identical or near to identical for different States. The Director General had observed that there were strong indications of some sort of D agreement and understanding amongst the bidders to manipulate the process of bidding. It was on this basis the CCI directed further investigation in the matter. The Director General after careful consideration submitted a detailed investigation report to the CCI. After the CCI considered the Efreshly ordered investigation report, it directed that a copy of the report be sent to the parties seeking their objections. In all, 44 opposite parties submitted their objections. After giving them the opportunity to be heard, the CCI passed the order in question. As per the Director General’s report, the process of bidding Ffollowed by the IOCL in the tender was as under :-

i) The bidders would submit their quotations with the bid documents. ii) The existing bidders, who were existing suppliers, were required to submit the price bids and technical bids. G iii) The bidders were to quote for supplies in different States of India in keeping with their installed capacity. iv) After price bids were opened the bidders were arranged according to the rates in the categories of L-1, L-2 and L-3. H 506 SUPREME COURT REPORTS [2018] 12 S.C.R. A v) The rates for the supplies in different States were approved after negotiations with L-1 bidder. In case the L-1 bidder could not supply a required number of cylinders in a particular State, the orders of supplies went to L-2 and also L-3 bidder or likewise depending upon the requirement in that State as per fixed formula provided in the bid documents. B vi) Certain bidders were called new parties. They were required to submit only technical bids and to supply as per L-1 rates determined after the negotiations. vii) One bidder could quote for maximum eight States.

C 7. The Director General after analyzing the bids came to the conclusion that there was not only a similarity of pattern in the price bids submitted by the 50 bidders for making supply to the IOCL but the bids of large number of parties were exactly identical or near to identical in different States. It was also found that bidders, who belonged to same group, might have submitted identical rates. It was found that not only D there was identical pricing in case of group concerns but the rates of other entities not belonging to the group were also found to be identical. The D.G. painstakingly noted the names of group companies as well as non-group companies. He came to the conclusion that in all 37 entities could not be said to be belonging to any single group and were E independently controlled. The Director General found it unusual that unrelated firms had quoted identical rates in different States. The D.G. had analyzed the bidding pattern for the various parties for all the 25 States. He found that :- a. The orders were placed on all the 50 successful bidders. F b. The contracts were awarded to the sets of bidders who had quoted identical rates or near to identical rates in a particular pattern in almost all the States. c. There was a common pattern for quotation depending upon the State. In case of North East the rates were highest, quoted at Rs. G 1240 whereas in case of others rates were Rs.1100, Rs.1127 and Rs. 1151.

d. It was found that only for Andaman and Nicobar Islands there was a single party who had quoted the L-1 rate and got the formal H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 507 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] contract. In other States the contracts were bagged in a group on A the basis of identical or near to identical rates.

e. The similarity of the rates was found even in case of bidders whose factories and offices were not located at one and the same place in the States and where they were required to supply was far off from their factories located in different place. B 8.The D.G. had found further that though the factors like market conditions and small number of companies were different, there was a large scale collusion amongst the bidding parties. He also arrived at a finding to the effect that the LPG Cylinder Manufacturers had formed an Association in the name of Indian LPG Cylinders Manufacturers C Association and the members were interacting through this Association and were using the same as a platform. The date for submitting the bids in the case of the concerned tender was 3.3.2010 and just two days prior to it, two meetings were held on 1st and 2nd March, 2010 in Hotel Sahara Star in Mumbai. As many as 19 parties took part and discussed Dthe tender and, in all probability, prices were fixed there in collusion with each other. The D.G. reported that the bidders had agreed for allocation of territories, e.g., the bidders who quoted the bids for Western India had not generally quoted for Eastern India and that largely the bidders who quoted the lowest in the group in Northern India, had not quoted generally in Southern India. The D.G. also concluded that this behavior E created entry barrier and that there was no accrual of benefits of consumers nor were there any plus factors like improved production or distribution of the goods or the provision of services.

9. Ultimately, the D.G. came to the conclusion that there was a cartel like behavior on the part of the bidders and that the factors F necessary for the formation of cartel existed in the instant case. It was also found that there was certainly a ground to hold concerted action on the part of the bidders. The D.G. had also noted that the rates quoted for the year 2009-10 and in years previous to that were also identical in some cases. Thus, he came to the conclusion that the bids for the year G2010-11 had been manipulated by 50 participating bidders. It was thereafter that the CCI decided to supply the D.G.’s investigation report to the concerned parties and invite their objections. 10. A common reply came to be filed as also the individual replies. After considering the same, the CCI formulated the following issue for determination:- H 508 SUPREME COURT REPORTS [2018] 12 S.C.R. A “Whether there was any collusive agreement between the participating bidders which directly or indirectly resulted in bid rigging of the tender floated by IOCL in March 2010 for procurement of 14.2 kg. LPG cylinders in contravention of Section 3(3)(d) read with Section 3(1) of the Act?”

B 11. After considering the oral as well as written submissions, the CCI answered the issue against the Cylinders Manufacturers and inflicted the penalties against the present appellants. In its impugned order, while determining the issue, the CCI, in the first instance, considered the common replies to the DG’s report filed by as many as 44 opposite parties. It was more or less pleaded that every part of LPG Cylinder is C regulated by the Rules through various Notifications and that the price of steel constitutes 50% of the total manufacturing cost, so also the price of the paint, it being an essential raw material. All these factors, including the taxes which vary from State to State, determine the overall bidding pattern of the bidders. In para-5.2.3 of the common objection, it D was added that these 44 parties had nominated six agents for depositing their bids on their behalf and it was a common practice amongst the bidders to direct their agents to keep close watch on the rates offered by their competitors in respect of a particular State and this led to the possibility of copying and matching of the rates quoted in the price bids by many suppliers in a particular State, who may have appointed common E agents. Due to this reason, cutting and over-writing in the price bids for the tender in question was noticed by the Director General.

12. It was further pointed out that there were only 62 qualified tenderers in the whole country, out of whom 12 bidders were classified as new parties, meaning thereby that they had not supplied Cylinders in F last three years and were not required to bid in the tender. Out of the remaining 50 bidders, there were group companies controlled by single management. 13. The CCI in its detailed order began with considering the scope of constructed bid rigging agreement and cartel. In that the CCI also G considered the 18 famous observations by Lord Denning in case of RRTA vs. W. H. Smith & Sons Limited regarding the quiet and secret nature of the agreement between the parties. The CCI then went on to record its inference holding that there was element of agreement and considered the following factors in coming to the conclusion. They being:-H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 509 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] 1. Market conditions A 2. Small number of suppliers 3. Few new entrants 4. Active trade association 5. Repetitive bidding B 6. Identical products 7. Few or no substitutes 8. No significant technological changes C 9. Meeting of bidders in Mumbai and its agenda. 10. Appointing common agents 11. Identical bids despite varying cost.

14. After consideration of these factors, the CCI came to the conclusion that it did suggest collusive bidding. Thereafter, the CCI D analyzed these bids for each States and found that all 50 participating bidders had secured the order; that the orders were placed on the said 50 bidders who had quoted identical rates or near to identical rates in a particular pattern common to all the parties. CCI also highlighted the facts of absence of business justification. According to the CCI, the E material revealed that the supplies were effected at the higher cost. After discussing the concepts of standards of proof and appreciable adverse effect on competition, the CCI considered the various arguments and repelled those arguments. The CCI then went on to consider the case law, and in particular the judgment of this Court in Union of India vs. Hindustan Development Corporation1. It also took into consideration F the arguments raised by the individual parties and then came to record that cases of M/s. JBM Industries and Punjab Cylinders, however, were exceptional ones and they could be exonerated. After this the CCI went on to decide the penalty factor under Section 27 of the Act. 15. The COMPAT after discussing the findings of the CCI and G also taking note of the arguments of the appellants which were advanced before the CCI, proceeded with its own discussion. It started with the admitted facts of the case, and took note of the following such facts: 1 (1993) 3 SCC 499 H

510 SUPREME COURT REPORTS [2018] 12 S.C.R. A (A) The tender offers were to be made at Mumbai on 03.03.2010. Admittedly there were meetings in Hotel Sahara Star, Mumbai on 1st and 2nd March, 2010 which were attended by some of the appellants. The D.G. has held that 19 appellants were represented by various persons in that meeting. The fact of the meeting having been held was not disputed. B Though some of the appellants stated that they did not attend the meeting and those who attended the meeting maintained that nothing was disucssed about the tender, the same was not believed by the COMPAT and it held that these meetings did relate to the tender offers which were to be submitted on 03.03.2010. This C finding is premised on the basis that nobody came with the explanation as to what transpired in the meeting or gave any proof that prices were discussed. Minutes of the meeting were also not produced. (B) There is an association of the cylinder manufacturers. All the D parties, except few competing with each other, stated that they were not the members of that association. A feeble argument was also raised by some appellants that though they were the members but they were not the active members thereof. Some of the appellants also argued that they had abandoned the membership E by not contributing the subscription in the later years. However, the appellants could not deny the position that there was an association called Indian LPG Cylinder Manufacturers’ Association.

It was a registered association, its Memorandum of Association provided that one of the objectives was to prtoect F common interest and welfare of LPG cylinder manufacturers. According to COMPAT, there was a definite platform available for all cylinder manufacturers and practically all the appllants appear to be the members of that Association. (C) A common written reply was submitted by as many as 44 G parties. Further, the appellants had nominated six agents for depositing bids on their behalf. These common agents were instructed to keep a close watch on the price quoted by the competitors in a particular State. H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 511 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] Though some of the appellants had contended that they had A not appointed the common agents, the plea was not accepted by the COMPAT. The COMPAT, therefore, proceeded on the ‘admitted grounds’ that there was an association of cyliner manufactures; practically all the appellants were members of the said association; this association was an active association; it held B

meetings on the eve of entry tender obviously for discussing tenders, its conditions etc.; these meetings were attended by representatives of at least 19 appellants; and these appellants had six common agents at Mumbai who were instructed to watch the prices offered by the others. A dinner meeting as also a lunch were held and one Mr. Chandi Prasad Bhartia of M/s. Haldia C Precision Engineering Private Limited paid the bill for the same. Dinner and lunch held in Sahara hotel were attended by about 50 persons in all. From this the COMPAT inferred that there was no reason to disbelieve that the parties had an access to each other through their association which was an active association. The D existence of such an association under the aegis of which meetings took place just before the submission of tender has been noted as a very relevant factor by the COMPAT in affirming the findings of CCI on cartelisation and it summed up the position in the following manner:

E “26. What is important is not whether a particular appellant was a member of the association or not. The existence of an association is by itself sufficient, as it gives opportunity to the competitors to interact with each other and discuss the trade problems. There will be no necessity to prove that any party actually discussed the prices by actively taking part in the F meeting. If there is a direct evidence to that effect that is certainly a pointer towards the fact that such party had a tacit agreement with its competitors. However, the existence of an association and further holding of the meetings just one or two days prior to the last date of making offers and further G admission that the parties had appointed common agents with the instructions to keep watch on the prices quoted by the competitors would go a long way in providing plus factors in favour of the agreement between the parties. All these factors would form a back drop, in the light of which, the further H 512 SUPREME COURT REPORTS [2018] 12 S.C.R. A evidence about agreement would have to be appreciated. We have seen the comments of Director General as also the findings of the CCI. We are convinced that CCI has not committed any error in considering all these factors as plus factors to come to the conclusion that there was a concerted agreement between the parties on the basis of which the B

identical or near identical prices came to be quoted in tenders for the supply of cylinders to the 25 States. In view of this, we need not dilate on the individual claims by some of the appellants that they were not the members of the association or that they were only the dormant members or that they had abdicated C their membership. We also need not go on the claim that while the meeting was attended by the 19 parties as held by the D.G. and confirmed by the CCI, it was not attended by the rest of the appellants because that would be of no consequence. Once there was a meeting, there was every opportunity to discuss or to communicate to each other whatever transpired in the D meeting. 27. We have seen the order of the CCI and while commenting about the meeting, the CCI has painstakingly noted the details of that meeting. The CCI has referred to the evidence of Mr. Dinesh Goyal, who was an active member of the Indian LPG E Cylinder Manufacturers’ Association and noted that he had attended the meeting. He has also referred to the statement of Mr. Sandeep Bhartia of Carbac Group though initially he denied to have organized the conference, he later on had confirmed about such a conference having been held along with Mr.

F Sandeep Bhartia of Carbac Group. The CCI also noted that he admitted that in such meetings there were discussions on pre-bid issues. He also admitted that though there are about 50 competitors, in fact about 25 persons control the whole affairs. From this evidence, the CCI correctly deduced that pre-bid issues were discussed in that meeting. The CCI has G then referred to the evidence of Mr. Manvinder Singh of Bhiwadi Cylinders Limited, Mr. Chandi Prasad Bhartia of Haldia Precision Engineering P. Ltd., Mr. Vijay Kumar Agarwal of SM Sugar Pvt. Ltd., Mr. S. Kulandhaiswamy, MD of Lite Containers Pvt. Ltd. and Secretary of the Association, Mr. H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 513 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] Ramesh Kumar Batra, Director of Surya Shakti Vessels Pvt. A Ltd. and on that basis came to the correct conclusion that not only was the meetings held on 1st and 2nd March, but thorough discussions went on in those meeting on the pre-bid issue of the concerned tender. The CCI has also correctly noted about the agenda of the meeting and has also referred to an admission B made by one of the witnesses that the matching of the quotation was a matter of co-incidence and telephonic discussions do take place amongst the parties regarding the trends. We are thus thoroughly convinced about holding of the meeting, the discussion held therein and also the discussion regarding the pre-bid issue having been taken place in that meeting.” C

16. The COMPAT thereafter took up for discussion the argument of the appellants that the CCI should have enquired IOCL also. But rejected the same. Another significant argument which was canvassed before us also with great emphasis was that it was an oligopolistic market wherein there was a likelihood of each player being aware of actions of D the other and in such a situation price parallelism would be a common phenomena. Thus, merely because there was a price parallalism, it would not be construed as evidence of collusion. The COMPAT rejected this argument as well. In the process, it analysed the order of CCI, conclusion whereof was founded on the following factors: E (1) The prevailing market conditions were such that there was a constant demand for cylinders not only by IOCL but by other two oil manufacutring companies as well. Therefore, there was a constant need for the cylinders which facilitated factor for the collusion. F (2) There was small number of suppliers. Among the 50 participating companies, only 37 companies could be said to be independant bidding companies and there were seven groups consisting of 20 participating companies. This small number of suppliers should also be a facilitating factor. G (3) There were very few new entrants. (4) The existence of an active trade association in which all the bidders, except seven companies, were members would be another facilitating factor. H 514 SUPREME COURT REPORTS [2018] 12 S.C.R.

A (5) Few other factors like repetitive bidding, identical products, few or no substitutes and no significant technological changes were the additional factors which persuaded the CCI to arrive at such a conclusion. (6) These manufacturing companies had their factories at different B places in India, where the costs of the components would differ from State to State. Even the taxing structure, the labour conditions and other factors like cost of electricity etc. were bound to be different. Still the prices quoted were almost identical. (7) On the above considerations, the defence of the appellants C was rejected as unconvincing, thereby undergoing the factors considered by the CCI. 17. According to the COMPAT all these could not have been possible unless there were internal agreements between the appellants. The COMPAT has approved the finding of the CCI that owing to the collusion, the IOCL could not get lower or the competitive prices. The D rates quoted in 2010-2011 were higher as compared to the rate quoted in 2009-10. From the year 2006-07, the prices had collectively been raised on an average of 30% for making supplies in different states.

18. According to the COMPAT, the CCI was right in concluding that it had appreciable adverse effect on competition as the conduct of E the LPG cylinder manufacturers in coming together on a common platform and fixing the bid prices ensures that no new player could enter the relevant market and quote the prices independently. Thus, these manufacturers would make entry of a new player into the relevant market difficult, because such new player would necessarily have to first F negotiate with the existing players to get the business profitably. Other factors were driving existing competitiors out of the market and foreclosure of competition by hindering entry into the market. 19. It negated the argument of the appellants that when the IOCL was placing orders on the basis of negotiated rates there could be no G possibility of incentive to collude. According to it, even where the rates are fixed, the bid rigging can still take place to keep the big amounts to a pre-determined level. Such pre-determination can be by way of intentional manipulation by members of the bidding group and where the L-1 rates themselves get fixed like in the present case at higher level even if there are negotations the negotiators would have to take into consideration the H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 515 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.]

benchmark rates. There is aslo a possibility that such benchmark rates A could go higher in the subsequent tenders; known as rippel effect in long term. 20. The COMPAT also took note of the provisions of Section 3, as per which once the agreement is proved there is a presumption about the appreciable adverse effect on competition on the mere proof of the B agreement. Thus, onus shifts on the other side to prove otherwise which according to the COMPAT was not discharged by the appellants. The COMPAT thereafter took note of some arguments by certain counsel specific to their cases but did not find any substance in them. 21. Having examined the relevant provisions whereupon these C appeals centre around, we proceed to take note of the arguments that were advanced by various counsel appearing for the appellants and the manner in which respondents endeavoured to meet the same. 22. Ms. Madhavi Divan, learned counsel appearing in the appeal filed by Rajasthan Cylinders and Containers Ltd., attacked the very basis D and foundation on which CCI came to conclusion that there was an agreement or cartelisation by the appellants aimed at bid rigging. She premised her case on the following three propositions: (i) the inherent nature of the market of cylinder manufacturers itself precludes the possibility of competition; E (ii) alternatively, there is no collusive agreement or bid-rigging in the present case; and

(iii) further, in the alternative, even assuming that there is a collusive agreement or bid-rigging in the present case, there is no appreciable adverse effect on competition. F 23. On the first proposition, argument developed by Ms. Divan was that the Act prohibits anti-competitive practices, which would imply that there has to be a competition in the market, in the first place. As a corrolary, if there is no such competition, Section 3(1) of the Act does not get triggered. According to her, in the instant case, the fact would G show that there was a tight control and regulation by the IOCL and, thus, it did not lead any scope of competition at the very threshold. She stressed that the conditions of monopsony/oligopsony prevailed. For the existence of monopsony/oligopsony, she referred to the Glossary of Industrial Organization Economics and Competition Law published by H 516 SUPREME COURT REPORTS [2018] 12 S.C.R. A the Organisation for Economic Co-operation and Development (OECD), as per which a monopsony consists of a market with a single buyer. When there are only a few buyers, the market is described as an oligopsony. In general, when buyers have some influence over the price of their inputs they are said to have monopsony power. The ability of a firm to raise prices, even when it is a monopolist, can be reduced or B eliminated by monopsony or oligopsony buyers. To the extent that input prices can be controlled in this way, consumers may be better off.

24. According to her, these conditions were adequately present in the instant case. In her attempt to make this propositoin good, she highlighed the following features and conditions surrounding the contract: C (i) Extremely limited number of buyers and for this particular kind of market - a sole buyer, i.e., IOCL. IOCL controls 48% of the market share. There are no other purchasers of 14.2 Kg gas cylinders except for HPCL and BPCL, both of whom invite e-tenders, having a market share of 26% and 25% respectively. D (ii) The product is standardized and special to the extent that it is tightly controlled and regulated by the Government and also there are no other takers for it. (iii) There are entry barriers in the market. As per the Tender conditions, only those manufacturers having valid approval from E the Chief Controller of Explosives (CCOE) and Bureau of Indian Standards (BIS) license for manufacture of 14.2 kg LPG cylinders as per IS-3196 (Part 1) could submit bids for the tender. (iv) Even the machinery used to manufacture this product is special and will become obsolete and reduceable to scrap if IOCL and F the aforesaid two players were to discontinue contracts for supply of 14.2 kg cylinders. She pointed out that this was accepted in the Expert Report of Dr. Rughvir K.S. Khemani. (v) The tender conditions state that it can be rejected without furnishing reasons. Therefore, the lowest price is not sacrosanct G (clause 11 of the contract).

(vi) L2 and L3 have also been granted contracts irrespective of the price they have quoted. (vii) Effective price has no sanctity since not only L2 and L3 also H get contracts in addition to or in exclusion of L1 but further, the RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 517 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] final negotiated price is determined on the basis of privately A conducted negotiations with individual bidders for which the benchmark is not the price quoted by them but the internal estimates arrived on the basis of objective criteria.

(viii) In most States, the final negotiated price was concluded at a rate lower than the internal estimate. The internal estimate had B absolutely no correlation with the quoted rates by L1 or any other party. In this behalf, she pointed out that the IOCL had carried out the exercise of ascertaining the estimated cost of the cylinder through its experts. In the report given by the expert, the estimated cost per cylinder was arrived at Rs. 1106.61 paisa per cylinder. C As against this, the final negotiated price at which the appellants had supplied cylinders to the IOCL was much lesser. According to her, in the whole process the price determination was on the basis of internal estimates by IOCL which could not be influenced by the appellants at all. In fact, even after the tenderers submitted their bids, final price was the price negotiated by IOCL which D fact was accepted by Mr. Y. Ramana Rao of IOCL in his deposition recorded by the Director General of CCI. This, according to the learned counsel, clearly proved that there was no adverse effect on competition, in any case. (ix) The internal estimates were drawn up long after the price E bids were made, i.e., on 5th May, 2010. Price bids were opened on 23rd March, 2010 and negotiations were held only after the submission of Mott MacDonald Report on 05.05.2010.

(x) The pattern shows that since L1, L2 and even L3 were awarded the contract and not merely L1, quoting the lowest price did not F even determine the identity of the parties who were to get the contract, therefore, the manner in which the process was conducted or controlled by IOCL, completely leaves no scope for either determination of price or the identity of the parties who would get the contract. G 25. She submitted that in such market conditions where on account of the vertical agreement there is virtually no scope of competitive forces between horizontal players, the question of anti-competitive conduct by virture of horizontal agreements does not arise. There is no competition in the market even before a player enters the fray. Therefore, the first H 518 SUPREME COURT REPORTS [2018] 12 S.C.R. A premise for the applciation of Section 3, i.e., the presence of an otherwise competitive market is absent. The burden of proof is on the respondent— CCI to establish that there is competition in the market before it can justify invoking Section 3. There is no automatic presumption under Section 3 that there is competition in the market. B 26. From the aforesaid factors, Ms. Divan tried to deduce that price control was entirely in the hands of IOCL and in a situation like this, question of entering into any agreement with the motive of bid rigging or collusive bidding did not arise.

27. She also referred to LPG (Regulation of Supply and C Distribution) Order, 2000 published vide Notification dated 26th April, 2000 as per which only Government oil companies can supply LPG to domestic consumer of 14.2 kg LPG cylinders with dimensions as specified therein. Predicated thereupon, her submission was that the LPG supply in 14.2 kg gas cylinders is an essential commodity; the distribution of such cylinders takes place only through Government oil companies; the D price to the consumer is controlled by the Government; and parallel marketeers, supplier and distributor of LPG cylinders may do so only for cylinders and specifications other than 14.2 kg cylinders. This control of the Government, insofar as supply of 14.2 kg gas cylinders is concerned, would also show tight control over the pricing. In such a statutorily tight E control price fixing mechanism there could not be bid rigging, was the submission of Ms. Divan. She supported this submission by drawing the attention of the Court to the following observations in Ashoka Smokeless Coal India (P) Ltd. v. Union of India2 :

“109. It may be true that prices are required to be fixed having regard to the market forces. Demand and supply is a relevant F factor as regards fixation of the price. In a market governed by free economy where competition is the buzzword, producers may fix their own price. It is, however, difficult to give effect to the constitutional obligations of a State and the principles leading to a free economy at the same time. A level playing field is the key G factor for invoking the new economy. Such a level playing field can be achieved when there are a number of suppliers and when there are competitors in the market enabling the consumer to exercise choices for the purpose of procurement of goods. If the 2 (2007) 2 SCC 640 H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 519 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] policy of the open market is to be achieved the benefit of the A consumer must be kept uppermost in mind by the State. xxx xxx xxx

127. While fixing a fair and reasonable price in terms of the provisions of the Essential Commodities Act (although the price is B not dual), it is essential that price is actually fixed. Such price fixation is necessary in view of the fact that coal is an essential commodity. It is, therefore, vital that price is actually fixed and not kept variable. Fixation of price of coal is of utmost necessity as it is a mineral of grave national importance. Non-availability of coal and consequently, the other products may lead to hardship to C a section of citizens. It may entail closure of factories and other industries which in turn would lead to loss to the State exchequer; as they would be deprived of its taxes. It will lead to loss of employment of a large number of employees and would be detrimental to the avowed object of the Central Government to D encourage small-scale industries.” 28. She also referred to the following discussion in Excel Crop Care Limited v. Competition Commission of India & Anr.3: “52. We are here concerned with parallel behaviour. We are conscious of the argument put forth by Mr Venugopal that in an E oligopoly situation parallel behaviour may not, by itself, amount to a concerted practice. It would be apposite to take note of the following observations made by European Court of Justice in Dyestuffs [Imperial Chemical Industries Ltd. v. Commission of European Communities, 1972 ECR 619 (ECJ)] : F

“By its very nature, then, the concerted practice does not have all the elements of a contract but may inter alia arise out of coordination which becomes apparent from the behaviour of the participants. Although parallel behaviour may not itself be identified with a concerted practice, it may however amount to strong evidence of such a practice if it leads to G conditions of competition which do not respond to the normal conditions of the market, having regard to the nature of the products, the size and number of the undertakings, and the 3 (2017) 8 SCC 47 H 520 SUPREME COURT REPORTS [2018] 12 S.C.R. A volume of the said market. Such is the case especially where the parallel behaviour is such as to permit the parties to seek price equilibrium at a different level from that which would have resulted from competition, and to crystallise the status quo to the detriment of effective freedom of movement of the products in the [internal] market and free choice by consumers B of their suppliers.” (emphasis supplied) At the same time, the Court also added that the existence of a concerted practice could be appraised correctly by keeping in mind the following test: C “If the evidence upon which the contested decision is based is considered, not in isolation, but as a whole, account being taken of the specific features of the products in question.”

29. The learned counsel also referred to various judgments of other jurisdictions, primarily that of European Commission and the Court D of Justice of European Union, which we shall discuss at the appropriate stage. 30. Ms. Divan, also highlighted that in this entire scenario, it was necessary to have the views of IOCL. However, in a suo motu case, IOCL was not even served with any notice and therefore no evidence E was elicited from IOCL on the issue whether there was any autnomy left to the manufacturer in the matter of price determination. 31. She, thus, argued that merely because there was price parallelism, it could not have been the reason to arrive at a conclusion that there was a collusive agreement or bid rigging. She submitted that F in a monopsonistic market where there are few buyers, the price is set by the buyers, and the conditions are such that sellers can predict demand, there is a repetitive bidding process and the products are identical and specilized, the likelihood of price parallelism is natural. 32. Further, price parallelism is inevitable where the buyer has a G high degree of control and determines price, quantity, and even the identities of the awardees at its discretion. Referring to the following discussion in Union of India v. Hindustan Development Corporation4, she argued that mere identical pricing cannot lead to the conclusion of cartelisation: H 4 (1993) 1 SCC 467 RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 521

UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] “7. [….] (1) There is not enough of material to conclude that M/ A s H.D.C., Mukand and Bhartiya formed a cartel. Because of mere quoting identical tender offers by the said three manufacturers for which there is some basis, the conclusion that the said manufacturers had formed a cartel does not appear to be correct. However since the offers of the said three tenderers B were identical and the price was somewhat lower, the Tender Committee entertained a suspicion that a cartel had been formed and the same got further strengthened by the post-tender attitude of the said manufacturers which further resulted in entertaining the same suspicion by the other authorities in the hierarchy of decision making body including the Minister of Railways. [….] C 33. She pointed out that this principle has also been stated in paragraph 17 of the Union of India v. Hindustan Development Coproration5. Her submission was that identical pricing may be further explained by the fact that, given the high degree of predictability of prices, bidders may take a business decision to mirror prices of competitors in D certain States, by adjusting or averaging prices in others. 34. The learned counsel pointed out that the CCI arrived at an inference of a collusive agreement based, inter alia, on the presence of circumstances which have acted as ‘facilitating factors’ for collusion. These factors which describe the nature of the industry are: E

(i) Predictability of demand (ii) Small number of suppliers (iii) Few new entrants (iv) Active trade association F (v) Repetitive bidding (vi) Identical products (vii) Few or no substitutes (viii) No significant technological changes, i.e, a standardised G product in repsect of which there has been no change or alteration in design. 35. Her reply was that these are the characteristics which define the industry. Yet these very factors are relied upon to come to the 5 (1993) 3 SCC 499 H 522 SUPREME COURT REPORTS [2018] 12 S.C.R. A consclusion that there is ‘collusion’ and ‘bid rigging’. She submitted that if the very nature of the industry is such that there are very small numbers of suppliers, very few new entrants and a standard product being supplied to the same party year after year, such factors are beyond the control of the individual manufacturers and cannot be relied upon as factors to lead to a presumption that there is collusive conduct. In other words, the B very nature of the industry cannot be used as a factor to presume collusion because collusion itself requires a state of mind or intent whereas in this case, most of these factors are inherent in the nature of the industry as described by the CCI itself.

36. Adverting to her 2nd proposition, namely, there was no collusive C agreement or bid rigging in the present case, her submission was that CCI has relied on a dinner attended by some manufacturers on 1st March, 2010 and a lunch on 2nd March, 2010 as evidence of a price fixing agreement. Her response was that the factum of meetings of an association by itself in any case cannot lead to a conclusion of collusion. D Likewise, the COMPAT also upheld that inference based on the factum of the meetings of the Associaiton. The COMPAT went to the extent of holding that it is irrelevant whether a particular party was a member of the Association or not and the existence of Association is by itself sufficient. This approach was attacked as contrary to the fundamental right to form an association under Article 19(1)(c)(g) of the Constitution E of India. 37. So far as the meetings over dinner and lunch are concerned, both were hosted by individual members. In the case of the dinner meeting on 1st March, 2010, it was hosted by Mr. C.P. Bhartiya, MD of North India Wires. The lunch on 2nd March, 2010 was hosted by Mr. F Santosh Bhartiya of Haldia Precision. It is not as if that the Association paid or the expenses were shared by all members who attended.

38. She also submitted that insofar as appellant – Rajasthan Cylinders and Containers Limited is concerned, no representatives of appellant attended the said meeting. Further, many other members did G not attend the meeting. Even as per the findings of the Director General, only 12 persons representing 19 parties are said to have attended the meeting. Her submission was that as per the allegations, 45 persons had entered into an agreement of cartelisation which should not be established only with the said meeting which was not attended by all and in fact very few members. H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 523 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] 39. In any case, according to her, it was expressly stated by at A least two persons who attended the meeting that price was not discussed. These are Mr. Chandi Prasad Bhartia from Haldia Precision and Mr. Manvinder Singh of Bhiwadi Cylinders Ltd. 40. Her further submission on this aspect was that the inference drawn on the basis of six agents being nominated for depositing 44 bids B was also misconceived. The CCI holds that ‘this might have held to the possibility of copying and matching of the rates quoted in the price bids by many suppliers in a particular state.” 41. The test as laid down in the case of Ahlstrom Osakeyhtio v. Commission6 is: Is the concertation the only plausible explaination for C the conduct?

“126. Following that analysis, it must be stated that, in this case, concertation is not the only plausible explanation for the parallel conduct. To begin with, the system of price announcements may be regarded as constituting a rational response to the fact that the D pulp market constituted a long-term market and to the need felt by both buyers and sellers to limit commercial risks. Further, the similarity in the dates of price announcements may be regarded as a direct result of the high degree of market transparency, which does not have to be described as artificial. Finally, the parallelism E of prices and the price trends may be satisfactorily explained by the oligopolistic tendencies of the market and by the specific circumstances prevailing in certain period. Accordingly, the parallel conduct established by the Commission does not constitute evidence of concertation. This test is not met in the present case for reasons that are F enumerated.

42. Her third proposition was that in any case there was no appreciable adverse effect on competition. She tried to make this submission good by contending that when industry is an oligopoly, the price parallel or a finding of identical quoting of price does not by itself G lead to the conclusion of a conerted price. Moreoever, in the instant case, number of entrants had increased as 12 new entrants submitted their bid for the year 2010-11. Therefore, the finding of the CCI, upheld 6 31.3.1993, ECJ (paragraph 115, Internal P.1611) (“Woodpulp“) H 524 SUPREME COURT REPORTS [2018] 12 S.C.R. A by the COMPAT, that there has been a creation of barriers for new entrants is without any basis.

43. Other counsel who appeared on behalf of the appellants made their submission almost on the same lines, albeit, with further elaborations on certain aspects, some of which are taken note of hereafter. Mr. Jaiveer Shergil, who argued for the appellant—Om Containers (C.A. B No. 6369 of 2014) submitted that in order to attract the presumption contained in Section 3(3) about the appreciable adverse effect on competition, in the first instance, there has to be a finding that there has been an agreement of the kind set out in Section 3(3)(a) to (d). Since, the allegation against the appellants was that the agreement resulted in C bid rigging and case is covered under Section 3(3)(d) of the Act, it was necessary that there is a positive finding to the aforesaid effect, namely, that there was agreement which had resulted in bid rigging. Accordoing to him, since the definition of bid rigging in Explanation to Section 3(3) uses the words ‘means’, the definition is a hard and fast definition and no other meaning can be assigned to the expression than is put down in D the definition, as held in Punjab Land Developement & Reclamation Corporation Ltd. vs. Presiding Officer, Labour Court7in the following words:

“72. The definition has used the word ‘means’. When a statute says that a word or phrase shall “mean”— not merely that it shall E “include” — certain things or acts, “the definition is a hard-and-fast definition, and no other meaning can be assigned to the expression than is put down in definition” (per Esher, M.R., Gough v. Gough [(1891) 2 QB 665 : 65 LT 110] ). A definition is an explicit statement of the full connotation of a term.” F 44. Thus, according to him, for it to be a case of bid rigging, the agreement must be such which is defined in the Explanation to Section 3(3)(d) creating the effect of: a. Eliminating or reducing competition for bids or b. Adversely affecting the process for bidding or G c. Manipulating the process for bidding. 45. He referred to the judgment in S. Sundaram Pillai vs. V.R. Pattabiraman8, on the purpose of an ‘Explanation’, viz.: 7 (1990) 3 SCC 682 H 8 (1985) 1 SCC 591 RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 525 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] “46. We have now to consider as to what is the impact of the A Explanation on the proviso which deals with the question of willful default. Before, however, we embark on an inquiry into this difficult and delicate question, we must appreciate the intent, purpose and legal effect of an Explanation. It is now well settled that an Explanation added to a statutory provision is not a substantive B

provision in any sense of the term but as the plain meaning of the word itself shows it is merely meant to explain or clarify certain ambiguities which may have crept in the statutory provision. Sarathi in Interpretation of Statutes while dwelling on the various aspects of an Explanation observes as follows: C (a) The object of an Explanation is to understand the Act in the light of the explanation. (b) It does not ordinarily enlarge the scope of the original section which it explains, but only makes the meaning clear beyond dispute.” D 46. He submitted that there is no positive evidence of this nature at all and the CCI as well as COMPAT has proceeded on inferences as regards bid rigging and, therefore, such orders cannot be sustained. 47. In the absence thereof, submitted the learned counsel, doctrine of reverse burden which was put on the appellants would not apply. He E referred to the following judgments9 in support. 48. The counsel relied upon the following observations in CCI v. Artistes & Technicians of W.B. Film & Television:

“31. The Competition Act, 2002, as amended in 2007 and 2009, deals with anti-trust issues viz. regulation of anti-competitive F agreements, abuse of dominant position and a combination or acquisition falling within the provisions of the said Act. Since the majority view of CCI also accepted that the impugned activities of the Coordination Committee did not amount to abuse of dominant position, and it treated the same as anti-competitive having G appreciable adverse effect on competition, our discussion would be focused only on anti-competitive agreements. Section 3 of the 9 (i) Babu v. State of Kerala: [(2010) 9 SCC 189] (ii) Noor Aga v. State of Punjab; [(2008) 16 SCC 417] (iii) CCI v. Artistes & Technicians of W. B. Film & Television: [(2017) 5 SCC 17] H 526 SUPREME COURT REPORTS [2018] 12 S.C.R. A Act is the relevant section in this behalf. It is intended to curb or prohibit certain agreements. Therefore, in the first instance, it is to be found out that there existed an “agreement” which was entered into by enterprise or association of enterprises or person or association of persons. Thereafter, it needs to be determined as to whether such an agreement is anti-competitive agreement B within the meaning of the Act. Once it is found to be so, other provisions relating to the treatment that needs to be given thereto get attracted.”

49. Taking aid of the aforesaid legal principle, it was submitted that in the present case it will be seen that the CCI, rather arriving at a C finding with focus on the aforesaid factors, proceeded to analyse factors which attach to the general market conditions of the industry to ‘infer’ the ‘possibility’ of bid rigging and then concluded that the ‘facilitating factors’ which may be ‘considered conducive for cartelisation’ are present. The D.G. found that ‘in all the probability, prices were fixed D there at the meeting in Bombay in collusion with each other. Such an inference and assumption based on ‘higher chances’, ‘probability’, ‘tendencies’ or ‘likelihood’ by the CCI does not meet the requirement of the definition contained in Explanation to Section 3(3) and certainly does not constitute a finding of ‘bid rigging’ as defined therein. The Tribunal has also proceeded on the basis that it ‘is to be deduced....that these E meetings did relate to the tender offers’. There was, thus, not clearcut, precise and consistent evidence to support that the alleged bid rigging took place.

50. Next submission of Mr. Shergil was that apart from the complete absence of a finding of bid rigging, in the present IOCL tender, as a F matter of fact there canot be any bid rigging as defined in Section 3(3). To take the first ingredient, i.e., eliminating or reducing competition for bids, the report of D.G. itself finds that out of the 60 bidding parties 37 entities were not belonging to any single group and are independently controlled. Hence, straight away there is no case of ‘eliminating or G reducing competition for bids’ which is one of the possible ingredients of bid rigging as there were 37 entities who were free of mind to participate and bid of their own accord in the absence of any control by any cartel. 51. As regards the second and third requirement of bid rigging, i.e., adversely effecting or manipulating the bidding process, he argued H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 527 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] that the submission of bids by the appellant (even if identical) can have A no effect of ‘adversely effecting or manipulating the bidding process’ this being on account of the very nature of the present tender process. Although, bids are invited from bidders, IOCL has a fixed/base procurement price of Rs. 1106.61 per cylinder. IOCL then works out an estimated rate per State based on certain factors peculiar to that State B

such as octroi, freight etc. The bid offered by the L1 (lowest bidder) is then subject to further downward negotiations by IOCL as per the tender clause and a further finalised rate is arrived at. Such finalised rate is eventually even lower10 than the L1 bid amount. Thus, factually, logically and in reality any bid submitted by any party can never be one which is said to adversely affect or manipulate the bidding process. All of this C information is with IOCL as part of its bidding process preparations, estimates and financial workings and could easily have been taken into consideration. In support, Mr. Shergil also referred to the terms and conditions of the IOCL tender. 52. His further submission was that CCI, or for that matter D COMPAT, were wrong in getting influenced by the submissions of identical bids by the appellants as it could not be, ipso facto, inferance of bid rigging. Such identical prices could be for various reasons and he shared that the reasons given by Ms. Divan predicated her submissions on oligopsony/monopsony. In addition, he relied upon the judgment of Ethis Court in Union of India vs. Hindustan Development Corporation11 and, in particular, para 17 thereof, which is as under:

“17. Therefore, whether in a given case, there was formation of a cartel by some of the manufacturers which amounts to an unfair trade practice, depends upon the available evidence and the F surrounding circumstances. In the instant case, initially the Tender Committee formed the opinion that the three big manufacturers formed a cartel on the ground that the price initially quoted by them was identical and was only a cartel price. This, in our view, was only a suspicion which of course got strengthened by post-tender attitude of the said manufacturers who quoted a much G lesser price. As noticed above it cannot positively be concluded on the basis of these two circumstances alone. In the past these 10 With the sole exception of the Andaman & Nicobar Islands as it has unique implication of transport related costs. 11 (1993) 3 SCC 499 H 528 SUPREME COURT REPORTS [2018] 12 S.C.R. A three big manufacturers also offered their own quotations and they were allotted quantities on the basis of the existing practice. However, a mere quotation of identical price and an offer of further reduction by themselves would not entitle them automatically to corner the entire market by way of monopoly since the final allotment of quantities vested in the authorities who in their B

discretion can distribute the same to all the manufacturers including these three big manufacturers on certain basis. No doubt there was an apprehension that if such predatory price has to be accepted the smaller manufacturers will not be in a position to compete and may result in elimination of free competition. But C there again the authorities reserved a right to reject such lower price. Under these circumstances though the attitude of these three big manufacturers gave rise to a suspicion that they formed a cartel but there is not enough of material to conclude that in fact there was such formation of a cartel. However, such an opinion entertained by the concerned authorities including the Minister D was not malicious nor was actuated by any extraneous considerations. They entertained a reasonable suspicion based on the record and other surrounding circumstances and only acted in a bona fide manner in taking the stand that the three big manufacturers formed a cartel.” E 53. Some literature on the ‘theory of oligopolistic interdependence’ as well as judgments of the European Union and European Commission were also cited.

54. Mr. Pradeep Aggarwal, in addition, argued that though there was no positive finding of cartelisation and the conclusion was merely F presumptive, even if it is accepted that there was such an agreement of bid rigging or collusive bidding, there was no presumption of ‘appreciable adverse effect on competiiton’. In the alternative, he submitted that there was, in fact, no appreciable adverse effect on competition in the present case and the said presumption totally was rebutted by producing G sufficient evidence on record. 55. Various other counsel also argued on the same lines and in addition referred to facts or their specific cases and it is not necessary to state all those arguments to avoid repetition. 56. Per contra, Mr. Salman Khurshid, learned senior counsel H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 529 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] appearijng for CCI highlighted the purpose for which the Act is enacted A and, in particular, objective behind Section 3 of the Act, whch is taken note of by this Court in Excel Crop Care Limited as well as West Bengal Artists Association. Insofar as instant case is concerned, his submission was that it is a stark and clear-cut case of bid rigging as a result of anti-competitive agreement amongst LPG manufacturers in B

respect of a tender (Tender No. LPG-O/M/PT-03/09-10) floated by IOCL for procurement of approximately 1,05,00,000 (105 laks) LPG Cylinders. This is a matter of serious public concern because these cylinders were to be used to supply Liquefied Petroleum Gas (LPG) for domestic consumption across 25 States. A rise in price resulting from anti-competitive activities would affect the cost of living for the common C man, and has serious ramifications for the economy as a whole. 57. Mr. Khurshid referred to the findings of the CCI as approved by COMPAT and submitted that there was a strong economic evidence of collusion which is evident from the following aspects: D (a) Identical or near-identical bidding by all 50 empaneled LPG vendors resulting in bid rigging. (b) Results of the tender revealed that these bids were made in such a way that all the bidders were awarded some portion of the tender and no bidder was left empty handed, i.e., Market Sharing E Arrangement. (c) Geographical/Territorial allocation of market, i.e., the bids were placed in such a way that entities located in the northern parts of the country were awrded the tender in the northern States, entities located in the southern parts were awarded the tender in respect of southern States etc. F (d) No plausible economic rationale or explanation was forthcoming for the identical bids, despite obvious difference in cost of production, location, input cost etc.

(e) The overall effect of increase in price of procurement of LPG G Cylinders over previous years. 58. He also submitted that pattern of identical and near identical bids, which was all pervasive through out, could not be brushed aside lightly as that was the clear indicator of price bidding as a result of agreement between the parties. The analysis of the bids also shows H 530 SUPREME COURT REPORTS [2018] 12 S.C.R. A that it had already been decided amongst the LPG Cylinder manufacturers as to who the L1 and L2 bidders were going to be prior to submission of bids. For instance, in the State of Punjab, the L1 bidder (Shri Ram Cylinders) bid Rs. 1080 whereas the four L2 bidders placed identical bids at Rs, 1080.50, i.e., a difference of only 50 paise from the L-1 bid. Similarly, in Rajasthan, the L-1 bidder (M/s. Rajasthan Cylinders) B quoted Rs. 1130, whereas nine L2 bidders quoted identically by just 50 paise more, at Rs. 1130.50. This pattern is repeated across a number of States. 59. Not only this, in order to achieve the pre-decided outcome, some of the bidders hastily made corrections to their bid documents. C One such case is that of M/s. Jesmajo Industrial Fabrications (appellant in C.A. No. 4868 of 2014). In the bid documents, the bid of Rs. 1103 was cut-corrected to make it Rs. 1103.60 even though the calculation of VAT was done only on the figure of |Rs. 1103. The Director General has commented on this aspect as follows:

D “6.13…...That bids were submitted after mutual discussion is apparent from the tender documents of Jesmajo Industrial Fabricators P. Ltd. (Exhibit 4P). There are cuttings in the tender documents and financial bids of the company. Since, there were discussions among all oil companies, the company might have E decided to make alterations in the financial bids. However, even in the financial bids of the company, it is noted that despite alterations, errors have remained. It seems that the company had originally quoted Rs. 1103/- as the rate. Subsequently it changed the rates to Rs. 1103.60. However, VAT rate in the bid had been calculated on Rs. 1103/- only instead of Rs. 1103.60. Thus, while F other component of rates has been changed/altered, the calculation of VAT has been done on Rs. 1103 (originally quoted) instead of Rs. 1103.60 (altered quote). This appears to have been done to match rates with other bidders who have quoted the similar rates in the State of Karnataka and to let Sanghvi group be L-1 in that G State.” 60. Mr. Khurshid also refuted the submisison of the appellants that there was no competition and, therefore, Section 3 was not applicable. According to him, if the matters are examined on such basis most of the culprits will get away. The purpose of the Act was not only to eliminate cartelisation but also to promote competition. His submission was that H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 531 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.]

once the findings of the CCI and COMPAT are accepted that there was A an agreement, such an agreement was obviously for the purpose of curbing the competition. 61. Answering the argument of ‘price parallelism’ which according to the appellants resulted in identical and near identical bids, Mr. Khurshid argued that legal submission in this respect was settled by this Court in B Excel Crop Care case wherein such an argument was rejected in the following words: “48…It was argued that since dominant position is enjoyed by the buyer, it leads to parallel pricing and this conscious parallelism takes place leading to quoting the same price by the suppliers. C The explanation, thus, given for quoting identical price was the aforesaid economic forces and not because of any agreement or arrangement between the parties. It was submitted that merely because same price was quoted by the appellants in respect of the 2009 FCI tender, one could not jump to the conclusion that D there was some “agreement” as well between these parties, in the absence of any other evidence corroborating the said factum of quoting identical price. In respect of this submission, Mr Venugopal had also referred to a few judgments.

49. The aforesaid argument is highly misconceived. A neat and E pellucid reply of Mr. Kaul, which commands acceptance, is that argument of parallelism is not applicable in bid cases and it fits in the realm of market economy. It is for this reason that entire history of quoting identical price before coming into operation of Section 3 and which continued much after Section 3 of the Act was enforced, has been highlighted...” F 62. He also referred to the following findings of COMPAT with the submission that finding of facts need to be accepted:-“36. We are thoroughly convinced by this analysis that all this could not have been possible unless there were internal agreements between the concerns. What shocks us is that the quotations of G the price did match to the last decimal and the quotations in some cases were in odd figures like Rs.1127 in the State of Tamil Nadu. The record is replete with such odd figures. It was strange that in some of the oral statements of the representatives of these parties, who were examined by the DG, some of them could not even H 532 SUPREME COURT REPORTS [2018] 12 S.C.R.

A justify these identical prices and tried to say that it was a mere coincidence. We cannot accept the argument of coincidence as was rightly rejected by CCI. There can be no explanation for this kind of identical or near identical pricing. The CCI has rightly considered that the manufacturing cost of per cylinder varies in a wide spectrum ranging from Rs.870 to Rs.1095.89. If this was B the case, the prices had to be different, if they had been offered in a competitive spirit. Either before the CCI or before us no material was produced, which would be able to rebut the presumption arising from the identity of rates. The CCI, therefore, rightly concluded that this identity of prices was sinister and anti-competitive in C nature.” 63. Another feature which Mr. Salman Khurshid pointed out was that the analysis of bids revealed that there was a market sharing and territorial allocation of bids. This, according to him, could be discerned from the following evidence emerging from record:

D “Firstly, the economic evidence indicates that there was an understanding or arrangement or agreement by which each and every bidder would get some part of the allocation under the tender. This is clear from the fact that each and every one of the 50 bidders who submitted price bids received some part of the E allocation in one or more states, and no one went empty handed. In other words, the purpose of quoting identical bids in many instances was to achieve the objective of sharing the market, i.e., the IOCL requirement across 25 states was ‘shared’ by each of the 50 bidders, through concerted action and pre-decided understanding. F 64. Mr. Khurshid also highlighed that in spite of there being difference in location of appellant’s units and their input cost, the bids submitted by various tenderers were identical and there canot be any plausible economic rationale for such identical bidding. Therefore, the inference drawn by the CCI as well as COMPAT based on the aforesaid G features and factors was justified and valid in law. He also referred to certain judgements of this Court as well as other jurisdictions, such as, European Commission and the Court of Justice of European Union to which reference would be made at the appropriate stage. 65. Before we deal with the arguments advanced by various H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 533 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.]

counsel who appeared for the appellants and rebuttal thereto by Mr. A Salman Khurshid, learned senior counsel who appeared for CCI as also counsel for IOCL, we would like to reproduce the relevant provisions of the Act in the light of which these appeals are to be decided: 66. Section 2(c) defines “cartel” and reads as under: B “”Cartel” includes an association of producers, sellers, distributors, traders or service providers who, by agreement amongst themselves, limit, control or attempt to control the production, distribution, sale or price of, or, trade in goods or provision of services; C 67. Section 3 of the Act deals with and prohibits those agreements which cause and are likely to cause an appreciable adverse effect on competition within india. It reads as under: “Section 3 : Anti-competitive agreements:- (1) No enterprise or association of enterprises or person or D association of persons shall enter into any agreement in respect of production, supply, distribution, storage, acquisition or control of goods or provision of services, which causes or is likely to cause an appreciable adverse effect on competition within India. (2) Any agreement entered into in contravention of the provisions E contained in sub-section (1) shall be void.

(3) Any agreement entered into between enterprises or associations of enterprises or persons or associations of persons or between any person and enterprise or practice carried on, or decision taken by, any association of enterprises or association of persons, F including cartels, engaged in identical or similar trade of goods or provision of services, which— (a) directly or indirectly determines purchase or sale prices; (b) limits or controls production, supply, markets, technical development, investment or provision of services; G (c) shares the market or source of production or provision of services by way of allocation of geographical area of market, or type of goods or services, or number of customers in the market or any other similar way; H 534 SUPREME COURT REPORTS [2018] 12 S.C.R. A (d) directly or indirectly results in bid rigging or collusive bidding, shall be presumed to have an appreciable adverse effect on competition: Provided that nothing contained in this sub-section shall apply to any agreement entered into by way of joint ventures if such agreement increases efficiency in production, supply, distribution, storage, acquisition or control of goods or provision of B services.

Explanation.—For the purposes of this sub-section, “bid rigging” means any agreement, between enterprises or persons referred to in sub-section (3) engaged in identical or similar production or trading of goods or provision of services, which has the effect of C eliminating or reducing competition for bids or adversely affecting or manipulating the process for bidding.” 68. Certain factors are mentioned in Section 19 of the Act which have to be kept in mind while determining whether an agreement has an appreciable adverse effect on competition under Section 3. We reproduce D this Section hereinbelow: “Section 19(3) in the Competition Act, 2002: (3) The Commission shall, while determining whether an agreement has an appreciable adverse effect on competition under section 3, have due regard to all or any of the following factors, namely:— E (a) creation of barriers to new entrants in the market; (b) driving existing competitors out of the market; (c) foreclosure of competition by hindering entry into the market; F (d) accrual of benefits to consumers; (e) improvements in production or distribution of goods or provision of services; (f) promotion of technical, scientific and economic development by means of production or distribution of goods or provision of G services.” 69. In Excel Crop Care Limited, scope of Section 3 of the Act which prohibits three kinds of practices as anti-competitive, was taken note of as follows:

“20. Chapter II of the Act deals with three kinds of practices H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 535 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] which are treated as anti-competitive and prohibited. These are: A (a) where agreements are entered into by certain persons with a view to cause an appreciable adverse effect on competition; (b) where any enterprise or group of enterprises, which enjoys dominant position, abuses the said dominant position; and B (c) regulating the combination of enterprises by means of mergers or amalgamations to ensure that such mergers or amalgamations do not become anti-competitive or abuse the dominant position which they can attain.” 70. In that case also the Court was concerned with the 1st category, C namely, those cases where certain persons enter into agreements with a view to cause an appreciable adverse effect of competition. Purpose behind curbing such anti-competitive practices was mentioned in detail. It is not necessary to re-state the same in that expansive manner, however, we would still like to quote certain portions to capture the essence and Dpurpose of the Act:-

“21. In the instant case, we are concerned with the first type of practices, namely, anti-competitive agreements. The Act, which prohibits anti-competitive agreements, has a laudable purpose behind it. It is to ensure that there is a healthy competition in the market, as it brings about various benefits for the public at large E as well as economy of the nation. In fact, the ultimate goal of competition policy (or for that matter, even the consumer policies) is to enhance consumer well-being. These policies are directed at ensuring that markets function effectively. Competition policy towards the supply side of the market aims to ensure that F consumers have adequate and affordable choices. Another purpose in curbing anti-competitive agreements is to ensure “level playing field” for all market players that helps markets to be competitive. It sets “rules of the game” that protect the competition process itself, rather than competitors in the market. In this way, the pursuit of fair and effective competition can contribute to improvements G in economic efficiency, economic growth and development of consumer welfare……….. xxx xxx xxx 23. In fact, there is broad empirical evidence supporting the H 536 SUPREME COURT REPORTS [2018] 12 S.C.R. A proposition that competition is beneficial for the economy. Economists agree that it has an important role to play in improving productivity and, therefore, the growth prospects of an economy…..

24. Productivity is increased through competition by putting B pressure on firms to control costs as the producers strive to lower their production costs so that they can charge competitive prices. It also improves the quality of their goods and services so that they correspond to consumers’ demands. xxx xxx xxx C 26. When we recognise that competition has number of benefits, it clearly follows that cartels or anti-competitive agreements cause harm to consumers by fixing prices, limiting outputs or allocating markets. Effective enforcement against such practices has direct visible effects in terms of reduced prices in the market and this is also supported by various empirical studies. D xxx xxx xxx 27. Keeping in view the aforesaid objectives that need to be achieved, Indian Parliament enacted the Competition Act, 2002. Need to have such a law became all the more important in the E wake of liberalisation and privatisation as it was found that the law prevailing at that time, namely, Monopolies and Restrictive Trade Practices Act, 1969 was not equipped adequately enough to tackle the competition aspects of the Indian economy. The law enforcement agencies, which include CCI and COMPAT, have to ensure that these objectives are fulfilled by curbing anti-competitive F agreements.”

71. The Court also mentioned, in particular, that competition leads to economic efficiency, economic growth and development as well as consumers welfare. The Court also spelled out the manner in which competition contributed to increase economic growth and increased G productivity. 72. It follows from the above that whereas on the one hand the economic policy of the nation has ushered in the era of liberalisation and globalisation thereby giving freeplay to the private sector in the manner of conducting business, at the same time, in public interest and in the H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 537 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] interest of consumers, a regime of regulators has also been brought to A ensure certain checks and balances. Since competition among the enterprises or businessmen is treated as service for a public purpose and, therefore, there is a need to curb anti-competitive practices, the CCI is given the task (as a regulator) to ensure that no such anti-competitive practices are undertaken. In fact, Section 18 of the Act B

casts a specific and positive obligation on CCI to ‘eliminate’ anti-competitive practices and promote competition, interest of the consmuer and free trade. This objective was also emphasised by this Court in Competition Commission of India vs. Steel Authority of India Limited and Another12 which can be found in the following observations: C “6. As far as the objectives of competition laws are concerned, they vary from country to country and even within a country they seem to change and evolve over the time. However, it will be useful to refer to some of the common objectives of competition law. The main objective of competition law is to promote economic efficiency using competition as one of the means of assisting the D creation of market responsive to consumer preferences. The advantages of perfect competition are threefold: allocative efficiency, which ensures the effective allocation of resources, productive efficiency, which ensures that costs of production are kept at a minimum and dynamic efficiency, which promotes E innovative practices. These factors by and large have been accepted all over the world as the guiding principles for effective implementation of competition law.”

73. As mentioned above, one of the anti-competitive practices is cartelisation, the essential postulate whereof is agreement between Fenterprises or association of enterprises or persons or associations of persons in respect of production, supply, distribution, storage, acquisition or control of goods or provisions of service, which causes or is likely to cause an appreciable adverse effect on competition within India. Such an agreement is treated as void. The types of agreement which may fall foul of Section 3 are mentioned in sub-section (3) thereof. These include G sharing the market by way of allocation of geographical areas of market [clause (c)] and the agreements which result in bid-rigging or collusive bidding whether directly or indirectly [clause (d)]. There is a presumption that four types of agreements mentioned in sub-section (3) will have an appreciable adverse effect on competition. H 538 SUPREME COURT REPORTS [2018] 12 S.C.R. A 74. We may also state at this stage that Section 19(3) of the Act mentions the factors which are to be examined by the CCI while determining whether an agreement has an appreciable adverse effect on competition under Section 3. However, this inquiry would be needed in those cases which are not covered by clauses (a) to (d) of sub-section (3) of Section 3. Reason is simple. As already pointed out above, the B

agreements of nature mentioned in sub-section (3) are presumed to have an appreciable effect and, therefore, no further exercise is needed by the CCI once a finding is arrived at that a particular agreement fell in any of the aforesaid four categories. We may hasten to add, however, that agreements mentioned in Section 3(3) raise a presumption that such C agreements shall have an appreciable adverse effect on competition. It follows, as a fortiorari, that the presumption is rebuttable as these agreements are not treated as conclusive proof of the fact that it would result in appreciable adverse effect on competition. What follows is that once the CCI finds that case is covered by one or more of the clauses mentioned in sub-section (3) of Section 3, it need not undertake D

any further enquiry and burden would shift upon such enterprises or persons etc. to rebut the said presumption by leading adequate evidence. In case such an evidence is led, which dispels the presumption, then the CCI shall take into consideration the factors mentioned in Section 19 of the Act and to see as to whether all or any of these factors are established. E If the evidence collected by the CCI leads to one or more or all factors mentioned in Section 19(3), it would again be treated as an agreement which may cause or is likely to cause an appreciable adverse effect of competition, thereby compelling the CCI to take further remedial action in this behalf as provided under the Act. That, according to us, is the broad scheme when Sections 3 and 19 are to be read in conjuction. F 75. In these appeals, the Court is concerned with the alleged agreement entered into between the appellants falling in clause (d) of sub-section (3) of Section 3, which talks of bid rigging or collusive bidding. Therefore, it would be necessary to understand the meaning of the expression ‘bid rigging’ and ‘collusive bidding’. Explanation to Section G 3, which is reproduced, assigns meaning to ‘bid rigging’ and states : “S. 3: Explanation.—For the purposes of this sub-section, “bid rigging” means any agreement, between enterprises or persons referred to in sub-section (3) engaged in identical or similar production or H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 539

UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] trading of goods or provision of services, which has the effect of A eliminating or reducing competition for bids or adversely affecting or manipulating the process for bidding.” 76. The necessary ingredients of bid rigging, thus, are: (a) agreement between the parties; (b) these parties are engaged in identical or similar production or trading of goods or provisions of services; and B (c) the agreement has the effect of eliminating or reducing competition of bids or adversely affect or manipulating the process for bidding. 77. Though the expression ‘collusive bidding’ is not defined in the Act, it appears that both ‘bid rigging’ and ‘collusive bidding’ are overlapping concepts. This position stands accepted in Excel Crop C Care Limited case which should be found from the following discussion therefrom:

“38. Mr Neeraj Kishan Kaul, learned Additional Solicitor General, refuted the aforesaid submission with vehemence by urging that bid rigging and collusive bidding are not mutually exclusive and D these are overlapping concepts. Illustratively, he referred to the findings of CCI, as approved by COMPAT, in the instant case itself to the effect that the appellants herein had “manipulated the process of bidding” on the ground that bids were submitted on 8-5-2009 collusively, which was only the beginning of the anti-E competitive agreement between the parties and this continued through the opening of the price bids on 1-6-2009 and thereafter negotiations on 17-6-2009 when all the parties reduced their bids by same figure of Rs 2 to bring their bid down to Rs 386 per kg from Rs 388 per kg. From this example, he submitted that on 8-5-2009 there was a collusive bidding but with concerted negotiations F on 17-6-2009, in the continued process, it was rigging of the bid that was practiced by the appellants. We are inclined to agree with this pellucid submission of the learned Additional Solicitor General. 39. Richard Whish and David Bailey [Competition Law, 7th Edn., G p. 536.] , in their book, have given illustrations of various forms of collusive bidding/bid rigging, which include: (a) Level tendering/bidding (i.e. bidding at same price — as in the present case). H 540 SUPREME COURT REPORTS [2018] 12 S.C.R. A (b) Cover bidding/courtesy bidding. (c) Bid rotation.

(d) Bid allocation. 40. Even internationally, “collusive bidding” is not understood as B being different from “bid rigging”. These two expressions have been used interchangeably in the following international commentaries/glossaries and websites of competition authorities: (a) UNCTAD Competition Glossary dated 22-6-2016 “Bid rigging or collusive tendering is a manner in which conspiring C competitors may effectively raise prices where business contracts are awarded by means of soliciting competitive bids. Essentially, it relates to a situation where competitors agree in advance who will win the bid and at what price, undermining the very purpose of inviting tenders which is to procure goods or services on the most favourable prices and conditions.” D (b) OECD Glossary of Industrial Organisation Economics and Competition Law “Bid rigging is a particular form of collusive price-fixing behaviour by which firms coordinate their bids on procurement or project E contracts. There are two common forms of bid rigging. In the first, firms agree to submit common bids, thus eliminating price competition. In the second, firms agree on which firm will be the lowest bidder and rotate in such a way that each firm wins an agreed upon number or value of contracts.

F Since most (but not all) contracts open to bidding involve Governments, it is they who are most often the target of bid rigging. Bid rigging is one of the most widely prosecuted forms of collusion.” Collusive bidding (tendering) — See “bid rigging”. (This shows collusive bidding and bid rigging are treated as one G and the same.) (c) OECD Guidelines for fighting bid rigging “Bid rigging (or collusive tendering) occurs when businesses, that would otherwise be expected to compete, secretly conspire to H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 541 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] raise prices or lower the quality of goods or services for purchasers A who wish to acquire products or services through a bidding process.” (d) United States Office of the Inspector General, Investigations (Fraud Indicators Handbook) B “Collusive bidding, price fixing or bid rigging, are commonly used interchangeable terms which describe many forms of an illegal anti-competitive activity. The common thread throughout all these activities is that they involve any agreements or informal arrangements among independent competitors, which limit competition. Agreements among competitors which violate the C law include but are not limited to: (1) Agreements to adhere to published price lists. (2) Agreements to raise prices by a specified increment.(3) Agreements to establish, adhere to, or eliminate discounts. D (4) Agreements not to advertise prices.

(5) Agreements to maintain specified price differentials based on quantity, type or size of product.” (e) Australian Competition and Consumer Commission E “Bid rigging, also referred to as collusive tendering, occurs when two or more competitors agree they will not compete genuinely with each other for tenders, allowing one of the cartel members to ‘win’ the tender. Participants in a bid rigging cartel may take turns to be the ‘winner’ by agreeing about the way they submit tenders, including some competitors agreeing not to tender.” F 41. As the Liegeman of the law, it is our task, nay a duty, to give proper meaning and effect to the aforesaid “Explanation”. It can easily be discussed that the legislature had in mind that the two expressions are interchangeably used. It is also necessary to keep in mind the purport behind Section 3 and the objective it seeks to G achieve: 41.1. Sub-section (1) of Section 3 is couched in the negative terms which mandates that no enterprise or association of enterprises or person or association of persons shall enter into any agreement, H 542 SUPREME COURT REPORTS [2018] 12 S.C.R.

A when such agreement is in respect of production, supply, distribution, storage, acquisition or control of goods or provision of services and it causes or is likely to cause an appreciable adverse effect on competition within India. It can be discerned that first part relates to the parties which are prohibited from entering into such an agreement and embraces within it persons as well as B enterprises thereby signifying its very wide coverage. This becomes manifest from the reading of the definition of “enterprise” in Section 2(h) and that of “person” in Section 2(l) of the Act. The second part relates to the subject-matter of the agreement. Again it is very wide in its ambit and scope as it covers C production, supply, distribution, storage, acquisition or control of goods or provision of services. The third part pertains to the effect of such an agreement, namely, “appreciable adverse effect on competition”, and if this is the effect, purpose behind this provision is not to allow that. Obvious purpose is to thwart any such agreements which are anti-competitive in nature and this salubrious D provision aims at ensuring healthy competition. Sub-section (2) of Section 3 specifically makes such agreements as void.

41.2. Sub-section (3) mentions certain kinds of agreements which would be treated as ipso facto causing appreciable adverse effect on competition. It is in this backdrop and context that “Explanation” E beneath sub-section (3), which uses the expression “bid rigging”, has to be understood and given an appropriate meaning. It could never be the intention of the legislature to exclude “collusive bidding” by construing the expression “bid rigging” narrowly. No doubt, clause (d) of sub-section (3) of Section 3 uses both the F expressions “bid rigging” and “collusive bidding”, but the Explanation thereto refers to “bid rigging” only. However, it cannot be said that the intention was to exclude “collusive bidding”. Even if the Explanation does contain the expression “collusive bidding” specifically, while interpreting clause (d), it can be inferred that “collusive bidding” relates to the process of bidding as well. G Keeping in mind the principle of purposive interpretation, we are inclined to give this meaning to “collusive bidding”. It is more so when the expressions “bid rigging” and “collusive bidding” would be overlapping, under certain circumstances which was conceded by the learned counsel for the appellants as well. H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 543 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.]

42. We are, therefore, of the opinion that the two expressions are A to be interpreted using the principle of noscitur a sociis i.e. when two or more words which are susceptible to analogous meanings are coupled together, the words can take colour from each other. (See Leelabai Gajanan Pansare v. Oriental Insurance Co. Ltd.[Leelabai Gajanan Pansare v. Oriental Insurance Co. Ltd., B (2008) 9 SCC 720] , Thakorlal D. Vadgama v. State of Gujarat [Thakorlal D. Vadgama v. State of Gujarat, (1973) 2 SCC 413 : 1973 SCC (Cri) 835] and M.K. Ranganathan v. State of Madras[M.K. Ranganathan v. State of Madras, (1955) 2 SCR 374 : AIR 1955 SC 604] .)” C

78. The first proposition of Ms. Divan, viz. there is no competition, has two facets. First, the legal one which concerns the jurisdiction of the CCI to deal with such matters and the other is factual, which is to be examined on the basis of facts in these cases. Insofar as the first component is concerned, having regard to the aforesaid scheme of the Act, we are not convinced with the argument of Ms. Madhavi Divan D that there is no possibility of a competition in these cases and, therefore, CCI had no jurisdiction to carry out any such investigation. The scope and ambit of the provisions of Section 3 have been considered in detail in Excel Crop Care Limited case. This Section prohibits anti-competitive agreements and brings about the prime objective of the Competition E Act. These aspects were noted in Excel Crop Care Limited, relevant portions whereof are already extracted above. We may also quote the following portion from the judgment of this Court in Steel Authority of India Limited wherein objective behind the Act was highlighted in the following manner: F

“125. We have already noticed that the principal objects of the Act, in terms of its Preamble and the Statement of Objects and Reasons, are to eliminate practices having adverse effect on the competition, to promote and sustain competition in the market, to protect the interest of the consumers and ensure freedom of trade carried on by the participants in the market, in view of the economic G developments in the country. In other words, the Act requires not only protection of free trade but also protection of consumer interest. The delay in disposal of cases, as well as undue continuation of interim restraint orders, can adversely and H 544 SUPREME COURT REPORTS [2018] 12 S.C.R. A prejudicially affect the free economy of the country. Efforts to liberalise the Indian economy to bring it on a par with the best of the economies in this era of globalisation would be jeopardised if time-bound schedule and, in any case, expeditious disposal by the Commission is not adhered to. The scheme of various provisions of the Act which we have already referred to including Sections B 26, 29, 30, 31, 53-B(5) and 53-T and Regulations 12, 15, 16, 22, 32, 48 and 31 clearly show the legislative intent to ensure time-bound disposal of such matters.”

79. We would like to reemphasise that the purpose of the Act is not only to illuminate practices having adverse effect on the competition C but also to promote and sustain competition in the market. Enforcement provides remedies to avoid situation that will lead to decrease competition in the market. Therefore, effective enforcement is important not only to sanction anti-competitive conduct but also to deter future competitive practices. In the present case itself, there are sixty suppliers of the product D for which there are three buyers. After all, each supplier would like to be L-1 or L-2 so that it is able to get order for larger quantities than the other. In this sense, there would be a competition among them. Further, it would also be in the interest of the buyers like IOCL etc. that the elements of healthy competition persists in the market. In any case, it is the duty of the CCI to ensure that the conditions which have tendency to E kill the competition are to be curbed. It is also the function of the CCI to ensure that there is a competition so that benefits of such competition are reaped by the consumers. However, insofar as certain factual aspects highlighted by the appellants are concerned, they would be dealt with while examining the third proposition, as we deem it more appropriate to F discuss these two aspects together.

80. Second proposition of Ms. Divan was that there was no collusive bidding in the present case. The CCI and COMPAT have rejected this argument in view of the fact that there is an active trade association of the suppliers; a meeting took place couple of days before G the date of bidding; common changes were pointed out by these appellants who submitted bids on their behalf; and bids were of identical amounts despite varying cost, which were repetitive in nature. The respondents may be right in their submission that there may not be a direct evidence on the basis of which cartelisation or such agreement between the parties can be proved as these agreements are normally entered into in closed H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 545 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] doors. The standard of proof which is required is one of probability, A which is a principle accepted in Technip SA vs. SMS Holding (P) Ltd. & Ors.13 wherein the Court stated and discussed this aspect in the following manner: “54. The standard of proof required to establish such concert is one of probability and may be established B “if having regard to their relation etc., their conduct, and their common interest, that it may be inferred that they must be acting together: evidence of actual concerted acting is normally difficult to obtain, and is not insisted upon” [CIT v. East Coast Commercial Co. Ltd., (1967) 1 SCR 821 : AIR 1967 SC 768] C . (SCR p. 829 H)

55. While deciding whether a company was one in which the public were substantially interested within the meaning of Section 23-A of the Income Tax Act, 1922 this Court said: “The test is not whether they have actually acted in concert D but whether the circumstances are such that human experience tells us that it can safely be taken that they must be acting together. It is not necessary to state the kind of evidence that will prove such concerted actings. Each case must necessarily be decided on its own facts.” [CIT v. Jubilee Mills Ltd., (1963) E 48 ITR 9 (SC), p. 20] 56. In Guinness PLC and Distillers Co. PLC [Guinness PLC and Distillers Company PLC (Panel hearing on 25-8-1987 and 2-9-1987 at p. 10052 — Reasons for decisions of the Panel.)] the question before the Takeover Panel was whether Guinness had F acted in concert with Pipetec when Pipetec purchased shares in Distillers Company PLC. Various factors were taken into consideration to conclude that Guinness had acted in concert with Pipetec to get control over Distillers Company. The Panel said: “The nature of acting in concert requires that the definition be G drawn in deliberately wide terms. It covers an understanding as well as an agreement, and an informal as well as a formal arrangement, which leads to cooperation to purchase shares to acquire control of a company. This is necessary, as such 13 (2005) 5 SCC 465 H 546 SUPREME COURT REPORTS [2018] 12 S.C.R.

A arrangements are often informal, and the understanding may arise from a hint. The understanding may be tacit, and the definition covers situations where the parties act on the basis of a ‘nod or a wink’…. Unless persons declare this agreement or understanding, there is rarely direct evidence of action in concert, and the Panel must draw on its experience and B common sense to determine whether those involved in any dealings have some form of understanding and are acting in cooperation with each other.” [Guinness PLC and Distillers Company PLC (Panel hearing on 25-8-1987 and 2-9-1987 at p. 10052 — Reasons for decisions of the Panel.)] C 81. We would also like to reproduce the following discussion in Commissioner of Income Tax, Bombay City I, Bombay vs. Jubilee Mills Ltd., Bombay14:

“19. At the hearing a point was raised that it has to be proved as a fact that the persons constituting the group which owns shares D carrying more than seventy-five per cent of the voting power, were acting in unison. The test is not whether they have actually acted in concert but whether the circumstances are such that human experience tells us that it can safely be taken that they must be acting together. It is not necessary to state the kind of E evidence that will prove such concerted actings. Each case must necessarily be decided on its own facts. The exclusion of “public” in the manner indicated generally from more than 75% of the shares and the concentration of such a holding in a single person or a group acting in concert is what attracts Section 23(A).” 82. It is also significant to state that respondents had drawn F attention of this Court to OECD Policy Roundtables Prosecuting Cartels without Direct Evidence 2006 which discussed the nature of evidence that is required for proving cartel agreement, relevant portion thereof contained in para 2 of the said Policy is reproduced below: “Available evidence for proving cartel agreements G 2.1 Categories of evidence Evidence used to prove a cartel agreement can be classified into two types: direct and circumstantial. Circumstantial evidence, in 14 (1963) 48 ITR 9 H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 547 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.]

turn, consists of “communication” evidence and economic A evidence, which include firm conduct, market structure, and evidence of facilitating practices. Common types of direct evidence include: - A document or documents (including email messages) B essentially embodying the agreement, or parts of it, and identifying the parties to it. - Oral or written statements by co-operative cartel participants describing the operation of the cartel and their participation in it. There are different types of circumstantial evidence. One is C evidence that cartel operators met or otherwise communicated but does not describe the substance of their communications. It might be called communication evidence for purposes of this discussion. It includes: - Records of telephone conversations between competitors D (but not their substance), or of travel to a common destination or of participation in a meeting, for example during a trade conference. - other evidence that the parties communicated about the subject e.g., minutes or notes of a meeting showing that prices, E demand or capacity utilisation were discussed; internal documents evidencing knowledge or understanding of a competitors pricing strategy, such as an awareness of a future price increase by a rival. A broader category of circumstantial evidence is often called F

economic evidence. Economic evidence identifies primarily firm conduct that suggests that an agreement was reached, but also conduct of the industry as a whole, elements of market structure which suggest that secret price fixing was feasible, and certain practices that can be used to sustain a cartel agreement. G Conduct evidence is the single most important type of economic evidence. As noted earlier, observation of certain, suspicious conduct frequently triggers an investigation of a possible cartel. And as the section in this paper on economics highlights 11 careful analysis of the conduct of parties is important to identify behaviours H 548 SUPREME COURT REPORTS [2018] 12 S.C.R. A that can be characterised as contrary to the parties’ unilateral self-interest and which therefore supports the inference of an agreement. Conduct evidence includes, first and foremost: - Parallel pricing – changes in prices by rivals that are identical, or nearly so, and simultaneous, or nearly so. It includes other B forms of parallel conduct, such as capacity reductions, adoption of standardised terms of sale, and suspicious bidding patterns, e.g., a predictable rotation of winning bidders. Industry performance could also be described as conduct evidence. It includes: C - abnormally high profits; - stable market shares - A history of competition law violations.

Evidence related to market structure can be used primarily to D make the finding of a cartel agreement more plausible, even though market structure factors do not prove the existence of such an agreement. Relevant economic evidence relating to market structure includes: - high concentration; E - low concentration on the opposite side of the market; - high barriers to entry; - high degree of vertical integration; - Standardised or homogeneous product. F The evidentiary value of structural evidence can be limited, however. There can be highly concentrated industries selling homogeneous products in which all parties compete. Conversely, the absence of such evidence cannot be used to show that a cartel did not exist. Cartels are known to have existed in industries with G numerous competitors and differentiated products. A specific kind of economic conduct evidence is facilitating practices – practices that can make it easier for competitors to reach or sustain an agreement. It is important to note that conduct described as facilitating practices is not necessarily unlawful. But H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 549 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.]

where a competition authority has found other circumstantial A evidence pointing to the existence of a cartel agreement, the existence of facilitating practices can be an important complement. They can explain what kind of arrangements the parties set up to facilitate the formation of a cartel agreement, monitoring, detection of defection, and/or punishment, thus supporting the ‘collusion B story’ put together by the competition law enforcer. Facilitating practices include: - information exchanges; - price signalling; C - freight equalisation; - price protection and most favoured nation policies; - Unnecessarily restrictive product standards.” 83. Thus, even in the absence of proof of concluded formal agreement, when there are indicators that there was practical cooperation D between the parties which knowingly substitute the risk of competition, that would amount to anti-competitive practices. This has been discussed in Coordination Committee of Artistes and Technicians of West Bengal Film and Television & Ors. (see paras 44 and 45). Then, there are guidelines on the applicability of Article 101 of the Treaty on E the functioning of the E.U. to horizontal cooperation agreements which records as under: “60. Information exchange can only be addressed under Article 101 if it establishes or is part of an agreement, a concerted practice or a decision by an association of undertakings. The existence of F

an agreement, a concerted practice or decision by an association of undertakings does not prejudge whether the agreement, concerted practice or decision by an association of undertakings gives rise to a restriction of competition within the meaning of Article 101(1). In line with the case-law of the Court of Justice of the European Union, the concept of a concerted practice refers G to a form of coordination between undertakings by which, without it having reached the stage where an agreement properly so-called has been concluded, practical cooperation between them is knowingly substituted for the risks of competition. The criteria of coordination and cooperation necessary for determining the H 550 SUPREME COURT REPORTS [2018] 12 S.C.R. A existence of a concerted practice, far from requiring an actual plan to have been worked out, are to be understood in the light of the concept inherent in the provisions of the Treaty on competition, according to which each company must determine independently the policy which it intends to adopt on the internal market and the conditions which it intends to offer to its customers. B

61. This does not deprive companies of the right to adapt themselves intelligently to the existing or anticipated conduct of their competitors. It does, however, preclude any direct or indirect contact between competitors, the object or effect of which is to create conditions of competition which do not correspond to the C normal competitive conditions of the market in question, regard being had to the nature of the products or services offered, the size and number of the undertakings, and the volume of the said market. This precludes any direct or indirect contact between competitors, the object or effect of which is to influence conduct D on the market of an actual or potential competitor, or to disclose to such competitor the course of conduct which they themselves have decided to adopt or contemplate adopting on the market, thereby facilitating a collusive outcome on the market. Hence, information exchange can constitute a concerted practice if it reduces strategic uncertainty in the market thereby facilitating E collusion, that is to say, if the data exchanged is strategic. Consequently, sharing of strategic data between competitors amounts to concentration, because it reduces the independence of competitors’ conduct on the market and diminishes their incentives to compete.

F 62. A situation where only one undertaking discloses strategic information to its competitor(s) who accept(s) it can also constitute a concerted practice. Such disclosure could occur, for example, through contacts via mail, emails, phone calls, meetings etc. It is then irrelevant whether only one undertaking unilaterally informs G its competitors of its intended market behaviour, or whether all participating undertakings inform each other of the respective deliberations and intentions. When one undertaking alone reveals to its competitors strategic information concerning its future commercial policy, that reduces strategic uncertainty as to the H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 551 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] future operation of the market for all the competitors involved A and increases the risk of limiting competition and of collusive behaviour. For example, mere attendance at a meeting where a company disclose its pricing plans to its competitors is likely to be caught by Article 101, even in the absence of an explicit agreement to raise prices. When a company receives strategic data from a B competitor (be it in a meeting, by mail or electronically), it will be presumed to have accepted the information and adapted its market conduct accordingly unless it responds with a clear statement that it does not wish to receive such data.” (emphasis added) C

84. According to us, the real question in the present case is as to whether there was a possibility of such an agreement having regard to market conditions even when we proceed on the basis that meeting did take place. Possibility of such an agreement has been inferred by the CCI on the grounds that identical bidding takes place thereafter and Dvarious suppliers gave such a bid despite varying cost and also that they have appoined common changes etc. as pointed out above. 85. The first and foremost issue which needs to be considered is that whether there was a situation of monopsony or oligopsony.

86. From the aforesaid discussion, it is clear that as far as CCI is E concerned, it has come to the conclusion that there was a cartelisation among the appellants herein and a concerted decision was taken to rig the bids which were submitted persuant to the tenders issued by IOCL. On the other hand, the appellants argue that there was no such agreement and even if the bids of many bidders were identical in nature, the bids were driven by market conditions. Their plea is that there was a situation F of oligopsony and the modus which was adopted by IOCL in floating the tenders and awarding the contracts would show that the determination of price was entirely within the control of the IOCL. As per them, the way price was determined for supply of these cylinders, it had become an open secret known to everybody. Therefore, there was no question G of any competition and no possibility of adversely affecting that competition by entering into any contract. 87. The factors which have influenced the authorities below in coming to the conclusion that the appellants had colluded and formed a H 552 SUPREME COURT REPORTS [2018] 12 S.C.R. A cartel which led to bid rigging have already been noted above. To recaptulate, the authorities below have been influenced by the following factors: 1. Market conditions 2. Small number of suppliers B 3. Few new entrants 4. Active trade association 5. Repetitive bidding 6. Identical products C 7. Few or no substitutes

8. No significant technological changes 9. Meeting of bidders in Mumbai and its agenda. D 10. Appointing common agents 11. Identical bids despite varying cost. After deliberating on the aforesaid aspects, the CCI has concluded that there is an active trade association in which many of the appellants are members. That product in question, namely, gas cylinder is of a E particular specification which is needed by IOCL in large numbers every year and there are very few manufacturers and suppliers of this product to IOCL and two other buyers. For this identical product which is to be supplied by all the suppliers, there is no substitute and no significant technology change. Further, there is an active trade association in which most of the appellants are the members. Their interest is to ensure that F no new entrants are able to join. Further, the trade association also ensures that all the members are able to get some order. It is for this reason the bids submitted in various standards which are floated by IOCL at different places are almost identical despite varying cost. The authorities below attributed this identical bidding to the concerted action G of the appellants. This has been inferred from the fact that 2-3 days before the submission of bids, meeting of the association took place which most of the appellants attended. Not only this, common agents, six in number, were appointed who submitted the bids on behalf of these appellants. H

RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 553 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] 88. We may say at the outset that if these factors are taken into A consideration by themselves, they may lead to the inference that there was bid rigging. We may, particularly, emphasise the fact that there is an active trade association of the appellants and a meeting of the bidders was held in Mumbai just before the submission of the tenders. Another very important fact is that there were identical bids despite varying cost. B Further, products are identical and there are small number of suppliers with few new entrants. These have become the supporting factors which persuaded the CCI to come to the conclusion that these are suggestive of collusive bidding. 89. However, that is only one side of the coin. The aforesaid Cfactors are to be analysed keeping in mind the ground realities that were prevailing, which are pointed out by the appellants. These attendant circumstances are argued in detail by the counsel for the appellants which have already been taken note of. We may recapitulate the same in brief hereinbelow: D (i) In the present case there are only three buyers. Among them, IOCL is the biggest buyer with 48% market share. It is also a matter of record that all these appellants are manufacturers of

14.2 kg gas cylinders to the three buyers who are available in the market, namely, IOCL, HPCL and BPCL. If these three buyers do not purchase from any of the appellants, that particular appellant E would not be in a position to sell those cylinder to any other entity as there are no other buyers. (ii) There are only three buyers, it may not attract many to enter the field and manufacture these cylinders. It is because of limited number of buyers and for some reason if they do not purchase, F the manufacturer would be nowhere. That may deter the persons to enter the field. (iii) The manner in which the tenders are floated by IOCL and the rates at which these are awarded, are an indicator that it is the IOCL which calls the shots insofar as price control is concerned. G It has come in evidence that the IOCL undertakes the exercise of having its internal estimates about the cost of these cylinders. Their own expert arrived at a figue of Rs. 1106.61 paisa per cylinder. All the tenders which have been accepted are for a price lesser than the aforesaid estimate of IOCL itself. That apart, H 554 SUPREME COURT REPORTS [2018] 12 S.C.R.

A the modus adopted by the IOCL is that that final price is negotiated by it and the contract is not awarded at the rate quoted by bidder who turns out to be L-1. Negotiations are held with such a bidder who is L-1 which generaly leads to further reduction of price than the one quoted by L-1. Thereafter, the other bidders who may be L-2 or L-3 etc. are awarded the contract at the rate at which it is B awarded to L-1. Thus, ultimately, all the bidders supply the goods at the same rate which is fixed by the IOCL after negotiating with L-1 bidder. The only difference is that bidder who is L-1 would be able to receive the order for larger quantity than L-2 and L-2 may get an order of more quantity than L-3. C (iv) It has also come on record that there are very few suppliers. For the tender in question, there were 50 parties already in the fray and 12 new entrants were admitted. Number of 12, in such a scenario, cannot be treated as less. Therefore, the conclusion of CCI that the appellants ensured that there should not be entry D of new entrant may not be correct. (v) Since there are not many manufacturers and supplies are needed by the three buyers on regular basis, IOCL ensures that all those manufacturers whose bids are technically viable, are given some order for the supply of specific cylinder. For this E purpose, it has framed its broad policy as well. This also shows that control remains with IOCL.

Thus, the appellants appear to be correct when they say that all the participants in the bidding process were awarded contracts in some State or the other which was aimed at ensuring a bigger pool of manufacturers so that the supply of this essential product F is always maintained for the benefit of the general public. Had IOCL left some manufacturers empty handed, in all likelihood, they would have shut their shops. However, IOCL wanted all manufacturers to be in the fray in its own interest. Therefore, it was necessary to keep all parties afloat and this explains why all G 50 parties obtained order along with 12 new entrants. (vi) There is another very relevant factor pointed out by the appellants, viz., the governmental control which is regulated by law. As pointed out above, it is not only the three oil companies which can supply LPG to domestic consumers in 14.2 kg LPG H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 555 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] cylinders as mandated in the LPG (Regulation and Distribution) A Order, 2000 which is issued under the provisions of Essential Commodities Act, 1955, even the price at which the LPG cylinder is to be supplied to the consumer is controlled by the Government. Following features of the aforesaid LPG Order, 2000, are significant: B • The LPG supplied in 14.2 kg gas cylinders is an essential commodity.

• The distribution of LPG in 14.2 kgs cylinders takes place as part of a public distribution system defined under clause 2(1) of C the Order as “the system of distribution, marketing or selling of liquified petroleum gas by a Government Oil Company at the Government controlled or declared price through a distribution system approved by the Central or State Government”. • The price to the consumer is controlled by the Government. D • The supply of LPG to domestic consumers shall be made only in 14.2 kg gas cylinders. • According to clauses 4 and 5 read with Schedule III of the LPG Order, parallel marketeers who supply and distribute LPG E cylinders, may do so only for cylinders with size and specifications other than those specified in Schedule II. 90. The manner in which tendering process takes place would show that in such a competitive scenario, the bid which the different bidder would be submitting becomes obvious. It has come on record that Fjust a few days before the tender in question, another tender was floated by BPCL and on opening of the said tender the rates of L-1, L-2 etc. came to be known. In a scenario like this, that obviously becomes a guiding factor for the bidders to submit their bids.

91. When we keep in mind the aforesaid fact situation on the Gground, those very factors on the basis of which the CCI has come to the conclusion that there was cartelization, in fact, become valid explanations to the indicators pointed out by the CCI. We have already commented about the market conditions and small number of suppliers. We have also mentioned that 12 new entrants cannot be considered as H 556 SUPREME COURT REPORTS [2018] 12 S.C.R. A entry of very few new suppliers where the existing suppliers were only 50. Identical products along with market conditions for which there would be only three buyers, in fact, would go in favour of the appellants. The factor of repetitive bidding, though appears to be a factor against the appellants, was also possible in the aforesaid scneario. The prevailing conditions in fact rule out the possibility of much price variations and all B

the manufacturers are virtually forced to submit their bid with a price that is quite close to each other. Therefore, it became necessary to sustain themselves in the market. Hence, the factor that these suppliers are from different region having different cost of manufacture would lose its significance. It is a situation where prime condition is to quote C the price at which a particular manufacturer can bag an order even when its manufacturing cost is more than the manufacturing cost of others. The main purpose for such a manufacuring would be to remain in the fray and not to lose out. Therefore, it would be ready to accept lesser margin. This would answer why there were near identical bids despite varying cost. D

92. Insofar as meeting of bidders in Mumbai just before the date of submission of tender is concerned, some aspects pointed out by the appellants are not considered by the CCI or the COMPAT at all. No doubt, the meeting took place a couple of days before the date of tender. No doubt, the absence of agenda coming on record would not make E much difference. However, only 19 appellants had attended that meeting. Many others were not even members or did not attend the meeting. In spite thereof, even they quoted almost same rates as the one who attended the meeting. This would lead us to the inference that reason for quoting similar price was not the meeting but something else. The question is F what would be the other reason and whether the appellants have been able to satisfactorily explain that and rebut the presumption against them? 93. The explanation is market conditions leading to the situation of oligopsony that prevailed because of limited buyers and influence of buyers in the fixation of prices was all prevalent. This seems to be G convincing in the given set of facts. The situation of oligopsony can be both ways. There may be a situation where the sellers are few and they may control the market and by their concerted action indulge into cartelization. It may also be, as in the present case, a situation where buyers are few and that results in the situation of oligopsony with the control of buyers. H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 557

UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] 94. To recapitulate, the two prime factors against the appellants, A which are discussed by the CCI, are that there was a collusive tendering, which is inferred from the parallel behaviour of the appellants, namely, quoting almost the same rates in their bids. The parameters on the basis of which these aspects are to be judged are stated in Excel Crop Care Limited as follows: B “50. It needs to be emphasised that collusive tendering is a practice whereby firms agree amongst themselves to collaborate over their response to invitations to tender. Main purpose for such collusive tendering is the need to concert their bargaining power, though, such a collusive tendering has other benefits apart from the fact C that it can lead to higher prices. Motive may be that fewer contractors actually bother to price any particular deal so that overheads are kept lower. It may also be for the reason that a contractor can make a tender which it knows will not be accepted (because it has been agreed that another firm will tender at a lower price) and yet it indicates that the said contractor is still D interested in doing business, so that it will not be deleted from the tenderee’s list. It may also mean that a contractor can retain the business of its established, favoured customers without worrying that they will be poached by its competitors.

51. Collusive tendering takes many forms. Simplest form is to E agree to quote identical prices with the hope that all will receive their fair share of orders. That is what has happened in the present case. However, since such a conduct becomes suspicious and would easily attract the attention of the competition authorities, more subtle arrangements of different forms are also made F between colluding parties. One system which has been noticed by certain competition authorities in other countries is to notify intended quotes to each other, or more likely to a Central secretariat, which will then cost the order and eliminate those quotes that it considers would result in a loss to some or all members of the cartel. Another system, which has come to light, is to rotate orders. G In such a case, the firm whose turn is to receive an order will ensure that its quote is lower than the quotes of others. 52. We are here concerned with parallel behaviour. We are conscious of the argument put forth by Mr Venugopal that in an H 558 SUPREME COURT REPORTS [2018] 12 S.C.R. A oligopoly situation parallel behaviour may not, by itself, amount to a concerted practice. It would be apposite to take note of the following observations made by European Court of Justice in Dyestuffs:

“By its very nature, then, the concerted practice does not have B all the elements of a contract but may inter alia arise out of coordination which becomes apparent from the behaviour of the participants. Although parallel behaviour may not itself be identified with a concerted practice, it may however amount to strong evidence of such a practice if it leads to conditions of competition which do not respond to the normal C conditions of the market, having regard to the nature of the products, the size and number of the undertakings, and the volume of the said market. Such is the case especially where the parallel behaviour is such as to permit the parties to seek price equilibrium at a different level from that which would D have resulted from competition, and to crystallise the status quo to the detriment of effective freedom of movement of the products in the [internal] market and free choice by consumers of their suppliers.” (emphasis supplied) E At the same time, the Court also added that the existence of a concerted practice could be appraised correctly by keeping in mind the following test: “If the evidence upon which the contested decision is based is considered, not in isolation, but as a whole, account being taken F of the specific features of the products in question.” Having regard to the aforesaid principles in mind, we deal with the arugment on oligopsony raised by the appellants.

95. Monopsony consists of a market with a single buyer. When there are only few buyers the market is described as an oligopsony. G What is emphasised is that in such a situation a manufacturer with no buyers will have to exit from the trade. Therefore, first condition of oligopsony stands fulfilled. The other condition for the existence of oligopsony is whether the buyers have some influence over the price of their inputs. It is also to be seen as to whether the seller has any ability H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 559 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] to raise prices or it stood reduced/eliminated by the aforesaid buyers. A 96. On a hollistic view of the matter, we find that the appellants have been able to discharge the onus by referring to various indicators which go on to show that parallel behaviour was not the result of any concerted practice.

B 97. In Dyestuffs, the European Court held that parallel behaviour does not, by itself, amount to a concerted practice, though it may provide a strong evidence of such a practice. Nevertheless, it is a strong evidence of such a practice. However, before such an inference is drawn it has to be seen that this parallel behaviour has led to conditions of competition which do not correspond to the normal conditions of the market, having C regard to the nature of the products, size and volume of the undertaking of the said market. Thus, we examine the matter from the stand point of market economy where question of oligopsony assumes relevance. Whenever there is a situation of oligopsony, parallel pricing simplicitor would not lead to the conclusion that there was a concerted practice Dthere has to be other credible and corroborative evidence to show that in an oligopoly a reduction in price would swiftly attract the customers of the other two or three rivals, the effect upon whom would be so devastating that they would have to react by matching the cut. In Richard Whish & David Bailey in Oxford’s Competition Law15 discussed the “Theory of Oligopolistic Interdependence” as under: E “In an oligopoly a reduction in price would swiftly attract the customers of the other two or three rivals, the effect upon whom would be so devastating that they would have to react by matching the cut. Similarly an oligopolist could not increase its price unilaterally, because it would be deserted by its customers if it did F so. Thus the theory runs that in an oligopolistic market rivals are interdependent: they are acutely aware of each other’s presence and are bound to match one another’s marketing strategy. The result is that price competition between them will be minimal or non-existent; oligopoly G produces non-competitive stability…..

...Oligopolists recognize their interdependence as well as their own self-interest. By matching each other’s conduct they 15 Seventh Edition; Chapter 14-Horizontal agreements (w)-oligopoly, tacit collusion and collective dominance; Pg 561-562, 565-566 H 560 SUPREME COURT REPORTS [2018] 12 S.C.R. A will be able to achieve and charge a profit-maximising price which will be set at a supra-competitive level, without actually communicating with one another. There does not need to be any communication: the structure of the market is such that, through interdependence and mutual self-awareness, prices will rise towards the monopolistic level…. B …..The logical conclusion of the case against oligopoly is that, since it is the market structure itself which produces the problem, structural measures should be taken to remedy it by deconcentrating the market. Unless this is done, there will be an area of consciously parallel action in pricing strategies which is C beyond the reach of laws against cartels and yet which has serious implications for consumers welfare. xxx xxx xxx xxx xxx xxx D (iii) A regulatory approach

A different possibility would be to regulate the prices of undertakings that operate in an oligopolistic environment. This, however, would be a counsel of despair. As a matter of policy direct regulation should be a remedy of last resort. Competition E authorities should not be price regulators; they should be the guardians of the competitive process. Where markets are oligopolistic and entry is limited, competition authorities should be concerned with the question of whether there are barriers to entry and whether the state itself, for example through restrictive licensing rules, regulation or legislation, F is responsible for a lack of competition.” 98. In Theatre Enterprises v. Paramount Films16, the Supreme Court of United States held as under: “1-3 The crucial question is whether respondents’ conduct toward G petitioner stemmed from independent decision or from an agreement, tacit or express. To be sure, business behavior is admissible circumstantial evidence from which the fact finder may infer agreement. Interstate Circuit. 16 346 US 357 H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 561 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] But this Court has never held that proof of parallel business behavior A conclusively establishes agreement or, phrased differently, that such behavior itself constitutes a Sherman Act offence.

Circumstantial evidence of consciously parallel behavior may have made heavy inroads into the traditional judicial attitude toward conspiracy; but “conscious parallelism” has not yet read conspiracy B out of the Sherman Act entirely.” 99. In this regard, the test laid down by the Supreme Court of United States in Monsanto Co. v. Spray-Rite Service Corp.17is relevant and is reproduced hereunder: “The correct standard is that there must be evidence that tends C to exclude the possibility that the manufacturer and non-terminated distributors were acting independently. That is, there must be direct or circumstantial evidence that reasonably tends to prove that the manufacturer and others had a conscious commitment to a common scheme designed to achieve an D unlawful objective.” 100. This test was reiterated by the Supreme Court of United States in Matsushita v. Zenith Ratio Corp.18: “…..But antitrust law limits the range of permissible inferences from ambiguous evidence in a 1 case. Thus, in Monsanto Co. v. E Spray-Rite Service Corp., 465 U.S. 752 (1984), we held that conduct as consistent with permissible competition as with illegal conspiracy does not, standing alone, support an inference of antitrust conspiracy. Id., at 764. See also Cities Service, supra, at

280. To survive a motion for summary judgment or for a directed verdict, a plaintiff seeking damages for a violation of 1 must present F evidence “that tends to exclude the possibility” that the alleged conspirators acted independently. 465 U.S., at 764 ……. …...petitioners had no motive to enter into the alleged conspiracy. To the contrary, as presumably rational businesses, petitioners had every incentive not to engage in the conduct with which they are G charged, for its likely effect would be to generate losses for petitioners with no corresponding gains. Cf. Cities Service, 391 U.S., at 279 . The Court of Appeals did not take account of the 17 465 U.S. 752, 104 S.Ct. 1464, 79 L.Ed.2d 775 (1984) 18 475 U.S. 574 (1986) H 562 SUPREME COURT REPORTS [2018] 12 S.C.R. A absence of a plausible motive to enter into the alleged predatory pricing conspiracy. It focused instead on whether there was “direct evidence of concert of action.” 723 F.2d, at 304. The Court of Appeals erred in two respects: (i) the “direct evidence” on which the court relied had little, if any, relevance to the alleged predatory pricing conspiracy; and (ii) the court failed to consider the absence B of a plausible motive to engage in predatory pricing….. xxx xxx xxx

Lack of motive bears on the range of permissible conclusions that might be drawn from ambiguous evidence: if petitioners had no C rational economic motive to conspire, and if their conduct is consistent with other, equally plausible explanations, [475 U.S. 574, 597] the conduct does not give rise to an inference of conspiracy. See Cities Service, supra, at 278-280.” 101. Similarly, in Bell Atlantic Corp v. Twombly19, the U.S. Supreme Court held as under: D

“[1-3] Because §1 of the Sherman Act “does not prohibit [all] unreasonable restraints of trade … but only restraints effected by a contract, combination, or conspiracy,” Copperweld Corp. v. Independence Tube Corp.,467 U.S. 752, 775(1984), “[t]he crucial question” is whether the challenged anticompetitive E conduct “stem[s] from independent decision or from an agreement, tacit or express,” Theatre Enterprises, 346 U. S., at 540. While a showing of parallel “business behavior is admissible circumstantial evidence from which the fact finder may infer agreement,” it falls short of “conclusively establishing agreement or … itself F constituting a Sherman Act offense.” Id., at 540–541. Even “conscious parallelism,” a common reaction of “firms in a concentrated market [that] recogniz[e] their shared economic interests and their interdependence with respect to price and output decisions” is “not in itself unlawful.” Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U. S. 209, 227 (1993) ; see 6 G P. Areeda & H. Hovenkamp, Antitrust Law 1433a, p. 236 (2d ed. 2003) (hereinafter Areeda & Hovenkamp) (“The courts are nearly unanimous in saying that mere interdependent parallelism does not establish the contract, combination, or conspiracy required by 19 550 U.S. 544 H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 563 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.]

Sherman Act §1”); Turner, The Definition of Agreement Under A the Sherman Act: Conscious Parallelism and Refusals to Deal, 75Harv. L. Rev. 655, 672 (1962) (“[M]ere interdependence of basic price decisions is not conspiracy”). [4-6] The inadequacy of showing parallel conduct or interdependence, without more, mirrors the ambiguity of the B behavior: consistent with conspiracy, but just as much in line with a wide swath of rational and competitive business strategy unilaterally prompted by common perceptions of the market. See, e.g., AEI-Brookings Joint Center for Regulatory Studies, Epstein, Motions to Dismiss Antitrust Cases: Separating Fact from C Fantasy, Related Publication 06–08, pp. 3–4 (2006) (discussing problem of “false positives” in §1 suits). Accordingly, we have previously hedged against false inferences from identical behavior at a number of points in the trial sequence. An antitrust conspiracy plaintiff with evidence showing nothing beyond parallel conduct is not entitled to a directed verdict, see Theatre Enterprises, D supra; proof of a §1 conspiracy must include evidence tending to exclude the possibility of independent action, see Monsanto Co. v. Spray-Rite Service Corp., 465 U. S. 752 (1984); and at the summary judgment stage a §1 plaintiff’s offer of conspiracy evidence must tend to rule out the possibility that the defendants E were acting independently, see Matsushita Elec. Industrial Co. v. Zenith Radio Corp., 475 U.S. 574(1986).”

102. After taking note of the test that needs to be applied in such cases, which was laid down in Dyestuffs and accepted in Excel Crop Care Limited, we come to the conclusion that the inferences drawn by F the CCI on the basis of evidence collected by it are duly rebutted by the appellants and the appellants have been able to discharge the onus that shifted upon them on the basis of factors pointed out by the CCI. However, at that stage, the CCI failed to carry the matter further by having required and necessary inquiry that was needed in the instant case. G 103. We are emphasising here that in such a watertight tender policy of IOCL which gave IOCL full control over the tendering process, it was necessary to summon IOCL. This would have cleared many aspects which are shrouded in mystery and the dust has not been cleared. H 564 SUPREME COURT REPORTS [2018] 12 S.C.R. A 104. We, thus, arrive at a conclusion that there is no sufficient evidence to hold that there was any agreement between the appellants for bid rigging. Accordingly, we allow these appeals and set aside the order of the Authorities below. As a consequence, since no penalty is payable, appeals of the CCI are rendered infructuous and dismissed as such. All the pending applications stand disposed of. B No orders as to costs. Kalpana K. Tripathy Appeals disposed of. C D E F G H

[2018] 12 S.C.R. 495 495 RAJASTHAN CYLINDERS AND CONTAINERS LIMITED A v. UNION OF INDIA AND ANOTHER (Civil Appeal No. 3546 of 2014) OCTOBER 01, 2018 B [A. K. SIKRI AND ASHOK BHUSHAN, JJ.] Competition Act, 2002: ss. 3, 19 and 27 – Cartelisation, bid-rigging and collusive bidding – Suo-motu proceedings initiated by Competition Commission C of India – Against bidders/manufacturers of LPG Gas Cylinders – Investigating into complaint about unfair conditions in the tender floated by Indian Oil Corporation Ltd. (IOCL) for supply of LPG Gas Cylinders – Competition Commission held that there was collusive bidding – Competition Appellate Tribunal affirmed the findings of D Competition Commission – On appeal, held: It is duty of the Commission to ensure that the conditions which have tendency to kill the competition are to be curbed – There may not be direct evidence on the basis of which cartelisation or such agreement between the parties can be proved – The standard of proof for such agreement is one of probability – There is a presumption that four E types of agreements mentioned in s. 3(3) will have an appreciable effect on competition – However, the presumption is rebuttable as these agreements are not conclusive proof of the fact that it would result in appreciable adverse effects on competition – If evidence is led which rebuts the presumption, the Commission shall take into F

consideration the factors mentioned in s. 19 – If the evidence collected by the Commission leads to one or more or all the factors mentioned in s. 19(3), it would again be treated as an agreement which may cause or likely to cause an appreciable adverse effect on competition – In the present case, inferences drawn by the Commission, on the basis of the evidence collected by it, have been G duly rebutted by the appellants/manufacturers – They have been able to discharge the onus that shifted upon them – However, at that stage, the Commission failed to carry the matter further – Thus, there is no sufficient evidence to hold that there was any agreement between the appellants for bid rigging – Appeals by the manufactures H 495 496 SUPREME COURT REPORTS [2018] 12 S.C.R. A are allowed – Since no penalty is payable, appeals of the Commission are dismissed as infructuous – Evidence. Words and Phrases: “bid rigging” and “collusive rigging” – Meaning of, in the context of Competition Act, 2002. B Allowing the appeals filed by the manufactures and dismissing the appeals filed by the Competition Commission, the Court

HELD: 1. On the one hand the economic policy of the nation has ushered in the era of liberalisation and globalisation thereby C giving freeplay to the private sector in the manner of conducting business, at the same time, in public interest and in the interest of consumers, a regime of regulators has also been brought to ensure certain checks and balances. Since competition among the enterprises or businessmen is treated as service for a public D purpose and, therefore, there is a need to curb anti-competitive practices. The Competition Commission of India (CCI) is given the task (as a regulator) to ensure that no such anti-competitive practices are undertaken. In fact, Section 18 of the Act casts a specific and positive obligation on CCI to ‘eliminate’ anti-competitive practices and promote competition, interest of the E consmuer and free trade. [Para 72] [536-G-H; 537-A-B] Competition Commission of India vs. Steel Authority of India Limited and Another (2010) 10 SCC 744 : [2010] 11 SCR 112 – relied on. F 2. One of the anti-competitive practices is cartelisation, the essential postulate whereof is agreement between enterprises or association of enterprises or persons or associations of persons in respect of production, supply, distribution, storage, acquisition or control of goods or provisions of service, which causes or is likely to cause an appreciable adverse effect on competition within G

India. Such an agreement is treated as void. The types of agreement which may fall foul of Section 3 are mentioned in sub-section (3) thereof. These include sharing the market by way of allocation of geographical areas of market [clause (c)] and the agreements which result in bid-rigging or collusive bidding H whether directly or indirectly [clause (d)]. There is a presumption RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 497 UNION OF INDIA AND ANOTHER that four types of agreements mentioned in sub-section (3) will A have an appreciable adverse effect on competition. [Para 73] [537-F-H]

3. Section 19(3) of the Act mentions the factors which are to be examined by the CCI while determining whether an agreement has an appreciable adverse effect on competition under B Section 3. However, this inquiry would be needed in those cases which are not covered by clauses (a) to (d) of sub-section (3) of Section 3. Reason is simple. The agreeements of nature mentioned in sub-section (3) are presumed to have an appreciable effect and, therefore, no further exercise is needed by the CCI Conce a finding is arrived at, that a particular agreement fell in any of the aforesaid four categories. Agreements mentioned in Section 3(3) raise a presumption that such agreements shall have an appreciable adverse effect on competition. It follows, as a fortiorari, that the presumption is rebuttable as these agreements are not treated as conclusive proof of the fact that it would result D in appreciable adverse effect on competition. What follows is that once the CCI finds that case is covered by one or more of the clauses mentioned in sub-section (3) of Section 3, it need not undertake any further enquiry and burden would shift upon such enterprises or persons etc. to rebut the said presumption by Eleading adequate evidence. In case such an evidence is led, which dispels the presumption, then the CCI shall take into consideration the factors mentioned in Section 19 of the Act and to see as to whether all or any of these factors are established. If the evidence collected by the CCI leads to one or more or all factors mentioned in Section 19(3), it would again be treated as F an agreement which may cause or is likely to cause an appreciable adverse effect of competition, thereby compelling the CCI to take further remedial action in this behalf as provided under the Act. That, is the broad scheme when Sections 3 and 19 are to be read in conjuction. [Para 74] [538-A-F]

G 4. Explanation to Section 3, assigns meaning to ‘bid rigging’. The necessary ingredients of bid rigging, are: (a) agreement between the parties; (b) these parties are engaged in idential or similar production or trading of goods or provisions of services; and (c) the agreement has the effect of eliminating or reducing competition of bids or adversely affect or manipulating H 498 SUPREME COURT REPORTS [2018] 12 S.C.R. A the process for bidding. Though the expression ‘collusive bidding’ is not defined in the Act, it appears that both ‘bid rigging’ and ‘collusive bidding’ are overlapping concepts. [Paras 75, 77] [538-G; 539-B-C]

5. Having regard to the scheme of the Act, it cannot be said that there is no possibility of a competition in the present B cases and, therefore, CCI had no jurisdiction to carry out any such investigation. Section 3 prohibits anti-competitive agreements and brings about the prime objective of the Competition Act. The purpose of the Act is not only to illuminate practices having adverse effect on the competition but also to C promote and sustain competition in the market. Enforcement provides remedies to avoid situation that will lead to decrease competition in the market. Therefore, effective enforcement is important not only to sanction anti-competitive conduct but also to deter future competitive practices. In the present case itself, there are sixty suppliers of the product for which there are three D buyers. After all, each supplier would like to be L-1 or L-2 so that it is able to get order for larger quantities than the other. In this sense, there would be a competition among them. Further, it would also be in the interest of the buyers like IOCL etc. that the elements of healthy competition persists in the market. [Paras E 78 and 79] [543-D; 544-C-E] 6. It is the duty of the CCI to ensure that the conditions which have tendency to kill the competition are to be curbed. It is also the function of the CCI to ensure that there is a competition so that benefits of such competition are reaped by F the consumers. [Para 79] [544-E]

7. The CCI and Competition Appellate Tribunal (COMPAT) have held that there was collusive bidding in the present case in view of the fact that there is an active trade association of the suppliers; a meeting took place couple of days before the date of bidding; common changes were pointed out by these appellants G who submitted bids on their behalf; and bids were of identical amounts despite varying cost, which were repetitive in nature. The respondents may be right in their submission that there may not be a direct evidence on the basis of which cartelisation or such agreement between the parties can be proved as these H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 499 UNION OF INDIA AND ANOTHER agreements are normally entered into in closed doors. The A standard of proof which is required is one of probability. [Para 80] [544-G-H; 545-A] Technip SA v. SMS Holding (P) Ltd. & Ors. (2205) 5 SCC 465 : [2005] 1 Suppl. SCR 223; Commissioner of Income Tax, Bombay City I, Bombay v. Jubilee Mills B Ltd., Bombay (1963) 48 ITR 9 – relied on. 8. Even in the absence of proof of concluded formal agreement, when there are indicators that there was practical co-operation between the parties which knowingly substitute the risk of competition, that would amount to anti-competitive C practices.[Para 83] [549-D]

9. The manner in which tendering process takes place would show that in such a competitive scenario, the bid which the different bidder would be submitting becomes obvious. It has come on record that just a few days before the tender in D question, another tender was floated by BPCL and on opening of the said tender the rates of L-1, L-2 etc. came to be known. In a scenario like this, that obviously becomes a guiding factor for the bidders to submit their bids. Keeping in mind the aforesaid fact situation, the very factors on the basis of which the CCI has Ecome to the conclusion that there was cartelisation, in fact, become valid explanations to the indicators pointed out by the CCI. [Paras 90 and 91] [555-F-G]

10. Identical products along with market conditions for which there would be only three buyers, in fact, would go in favour of the appellants. The factor of repetitive bidding, though appears F to be a factor against the appellants, was also possible in the aforesaid scneario. The prevailing conditions in fact rule out the possibility of much price variations and all the manufacturers are virtually forced to submit their bid with a price that is quite close to each other. Therefore, it became necessary to sustain G themselves in the market. Hence, the factor that these suppliers are from different region having different cost of manufacture would lose its significance. It is a situation where prime condition is to quote the price at which a particular manufacturer can bag H 500 SUPREME COURT REPORTS [2018] 12 S.C.R. A an order even when its manufacturing cost is more than the manufacturing cost of others. The main purpose for such a manufacuring would be to remain in the fray and not to lose out. Therefore, it would be ready to accept lesser margin. This would answer why there were near identical bids despite varying cost. [Para 91] [556-A-D] B

11. Insofar as meeting of bidders in Mumbai just before the date of submission of tender is concerned, some aspects pointed out by the appellants are not considered by the CCI or the COMPAT at all. No doubt, the meeting took place a couple of days before the date of tender. No doubt, the absence of agenda C coming on record would not make much difference. However, only 19 appellants had attended that meeting. Many others were not even members or did not attend the meeting. In spite thereof, even they quoted almost same rates as the one who attended the meeting. This would lead the Court to the inference that reason D for quoting similar price was not the meeting but something else. [Para 92] [556-D-E] 12. Monopsony consists of a market with a single buyer. When there are only few buyers the market is described as an oligopsony. In such a situation, a manufacturer with no buyers E will have to exit from the trade. Therefore, first condition of oligopsony stands fulfilled. The other condition for the existence of oligopsony is whether the buyers have some influence over the price of their inputs. It is also to be seen as to whether the seller has any ability to raise prices or it stood reduced/eliminated by the aforesaid buyers. [Para 95] [558-G-H; 559-A] F

13. On a hollistic view of the matter, it is clear that the appellants have been able to discharge the onus by referring to various indicators which go on to show that parallel behaviour was not the result of any concerted practice. [Para 96] [559-A-B] 14. Parallel behaviour does not, by itself, amount to a G concerted practice, though it may provide a strong evidence of such a practice. Nevertheless, it is a strong evidence of such a practice. However, before such an inference is drawn it has to be seen that this parallel behaviour has led to conditions of competition which do not correspond to the normal conditions of H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 501 UNION OF INDIA AND ANOTHER the market, having regard to the nature of the products, size and A volume of the undertaking of the said market. [Para 97] [559-B-C] 15. Whenever there is a situation of oligopsony, parallel pricing simplicitor would not lead to the conclusion that there was a concerted practice there has to be other credible and corroborative evidence to show that in an oligopoly a reduction B in price would swiftly attract the customers of the other two or three rivals, the effect upon whom would be so devastating that they would have to react by matching the cut. [Para 97] [559-D-E]

16. The inferences drawn by the CCI on the basis of evidence collected by it are duly rebutted by the appellants and C the appellants have been able to discharge the onus that shifted upon them on the basis of factors pointed out by the CCI. However, at that stage, the CCI failed to carry the matter further by having required and necessary inquiry that was needed in the instant case. [Para 102] [563-F-G] D 17. In such a watertight tender policy of IOCL which gave IOCL full control over the tendering process, it was necessary to summon IOCL. This would have cleared many aspects which are shrouded in mystery and the dust has not been cleared. [Para 103] [563-G-H] E 18. Thus, there is no sufficient evidence to hold that there was any agreement between the appellants for bid rigging. As a consequence, since no penalty is payable, appeals of the CCI are rendered infructuous and dismissed as such. [Para 104] [564-A-B] Union of India vs. Hindustan Development Corporation F (1993) 3 SCC 499 : [1993] 3 SCR 128; Ashoka Smokeless Coal India (P) Ltd. v. Union of India (2007) 2 SCC 640 : [2006] 9 Suppl. SCR 954; Excel Crop Care Limited v. Competition Commission of India & Anr. (2017) 8 SCC 47 : [2017] 5 SCR 901; Union of India v. Hindustan Development Corporation (1993) 1 SCC G 467 : [1993] 3 SCR 108; Union of India v. Hindustan Development Coproration (1993) 3 SCC 499 : [1993] 3 SCR 128; Punjab Land Developement & Reclamation Corporation Ltd. v. Presiding Officer, Labour Court H

502 SUPREME COURT REPORTS [2018] 12 S.C.R. A (1990) 3 SCC 682 : [1990] 3 SCR 111; S. Sundaram Pillai v. V.R. Pattabiraman (1985) 1 SCC 591 : [1985] 2 SCR 643; CCI v. Artistes & Technicians of W.B. Film & Television (2017) 5 SCC 17 : [2017] 5 SCR 1; Union of India vs. Hindustan Development Corporation (1993) 3 SCC 499 : [1993] 3 SCR 128 – referred to. B Ahlstrom Osakeyhtio v. Commission 31.3.1993, ECJ (“Woodpulp”); Theatre Enterprises v. Paramount Films 346 US 357; Monsanto Co. v. Spray-Rite Service Corp. 346 US 357; Monsanto Co. v. Spray-Rite Service Corp. 465 U.S. 752, 104 S.Ct. 1464, 79 L.Ed. 2Nd 775 (1984); C Matsushita v. Zenith Ratio Corp. 475 U.S.574 (1986) Bell Atlantic Corp v. Twombly 550 U.S. 544 – referred to. Case Law Reference [1993] 3 SCR 128 referred to Para 14 D [2006] 9 Suppl. SCR 954 referred to Para 27 [2017] 5 SCR 901 referred to Para 28 [1993] 3 SCR 108 referred to Para 32 E [1993] 3 SCR 128 referred to Para 33 [1990] 3 SCR 111 referred to Para 43 [1985] 2 SCR 643 referred to Para 45 [2017] 5 SCR 1 referred to Para 48 F [1993] 3 SCR 128 referred to Para 52 [2010] 11 SCR 112 referred to Para 72 [2005] 1 Suppl. SCR 223 relied on Para 80 (1963) 48 ITR 9 relied on Para 81 G CIVIL APPELLATE JURISDICTION: Civil Appeal No. 3546 of 2014 From the Judgment and Order dated 20.12.2013 of the Competition Appellate Tribunal in Appeal No. 59 of 2012. H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 503 UNION OF INDIA AND ANOTHER WITH A

Civil Appeal Nos. 4280/2014, 4346/2014, 4649/2014, 4342/2014, 4879/2014, 4868/2014, 6033/2014, 5771/2014, 5772/2014, 5035/2014, 5773/2014, 5649/2014, 5650/2014, 5651/2014, 4972/2014, 6661/2014, 7214/2014, 7102/2014, 6365/2014, 6025/2014, 5993-5994/2014, 6868/2014, 5774/2014, 5775/2014, 5776/2014, 5832-5833/2014, 5777/2014, 5778/2014, B 6371/2014, 8953/2014, 6372/2014, 6373/2014, 6366/2014, 6367/2014, 6374/2014, 6368/2014, 6364/2014, 6369/2014, 6370/2014, 10579/2014, 1724/2015, 5277-5278/2016, 5281-5315/2016, 7359/2016.

Salman Khurshid, Prashanto Sen, Sr. Advs., Manan Verma, Ms. Nidhi Khanna, Mrs. Sarla Chandra, Ms. Diksha Rai, Arjun Krishnan, O. C P. Gaggar, Aditya Gaggar, Sudipto Sircar, Annam D. N. Rao, Annam Venkatesh, Rahul Mishra, Amol Sinha, Ms. Anshum Jain, Rahul Kochar, Naveen Kumar, Deepak Anand, N. S. Nandakumar, N. Krishn Kumar, K. Krishna Kumar, Sarad Kumar Singhania, Ms. Ruchi Kohli, Ms. Pallavi Langar, Jaiveer Shergill, Amit Goyal, Rajesh Sharma, Ms. Nidhi Singh DDubey, Shafiq Khan, Ms. Shalu Sharma, R. Satish Kumar, V. Susheetha, P. V. Yogeswaran, Rameshwar Prasad Goyal, S. Nanda Kumar, R. Sathis Kumar, M. S. Saron Kumar, Ms. N. Deepika Nanda Kumar, Naresh Kumar, Praveen Mahajan, Arjun Krishnan, Ankur Singh, Sumit Srivastava, Kaustav Som, Ms. Gitanjali Kapur, Antony Julian, Arpit Shukla, Udayan Verma, Sarvesh Mishra, Amit Sharma, Raj Bahadur, Ms. Anil Katiyar, E Kedar Nath Tripathy, Girija Ballav Das, B. B. Pradhan, Gaurav Agrawal, Manan Verma, Ms. Diksha Rai, Ms. Palak Mahajan, Kuljeet Rawal, Ashwani Kumar, Ms. Iti Sharma, Pradeep Aggarwal, Karan Khanna, Aniket Bhattacharya, Arjun Agggarwal, Advs. for the appearing parties. The Judgment of the Court was delivered by F A. K. SIKRI, J. 1. All these appeals are filed against the orders dated 20th December, 2013 passed by the Competition Appellate Tribunal (hereinafter referred to as ‘COMPAT’). The COMPAT by the said judgment has upheld the findings of the Competition Commission of India (for short, ‘CCI’) that the appellants/suppliers of Liquefied Petroleum G Gas (LPG) Cylinders to the Indian Oil Corporation Ltd. (for short, ‘IOCL’) had indulged in cartilisation, thereby influencing and rigging the prices, thus, violating the provisions of Section 3(3)(d) of the Competition Act, 2002 (for short, the ‘Act’). The CCI, as a result, imposed severe penalties in the form of fines under Section 27 of the Act. While maintaining the H

504 SUPREME COURT REPORTS [2018] 12 S.C.R. A order of the CCI insofar as it found the appellants guilty of contravention of Section 3(3)(d) and also under Section 3(3)(a) of the Act, the COMPAT has reduced the amount of penalty. These suppliers have filed the instant appeals on the ground that there was no cartilisation and they have not contravened the provisions of the Act. On the other hand, CCI has also come up in appeal challenging latter part of the order B whereby penalties inflicted on the suppliers stand reduced. For the sake of convenience these suppliers will be referred to as the appellants hereinafter.

2. We may point out at the outset that all these appellants are manufacturing gas cylinders of a particular specification having capacity C of 14.2 kg which are needed for use by the three oil companies in India, namely, IOCL, Bharat Petroleum Corporation Ltd. (BPCL) and Hindustan Petroleum Corporation Ltd. (HPCL) [all public sector companies]. It is also a matter of record that apart from the aforesaid three companies there are no other buyers for these cylinders D manufactured by the appellants. Insofar as IOCL is concerned, it is a leading market player in LPG as its market share is 48%. Thus, in case a particular manufacturer is not able to supply its cylinders to the aforesaid three companies, there is no other market for these cylinders and it may force that company to exit from its operations. We may also point out at this stage itself that inquiry was started against 47 companies. The CCI E exonerated two companies and found that 45 companies had entered into an arrangement/agreement insofar as statements of bids pursuant to tenders issued by IOCL are concerned. Out of these 45 companies one did not challenge the orders before the COMPAT and other 44 had filed appeals which have been decided by the COMPAT.

F 3. The manner in which the inquiry was undertaken by the CCI, culminating into the finding of guilt and imposition of penalty, is succintly and sequally recorded by the COMPAT in its impugned order. As there is no dispute about the said factual narration, it would be convenient to borrow the said discussion as recorded by the COMPAT. G 4. The suo-motu proceedings were started by the CCI on the basis of the information received by it in Case No. 10 of 2010 titled M/ s. Pankaj Gas Cylinders Ltd. Vs. Indian Oil Corporation Ltd. in that case a complaint was made by M/s. Pankaj Gas Cylinders before the CCI complaining about unfair conditions in the tender floated by IOCL H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 505 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] for the supply of 105 lakh 14.2 Kg. capacity LPG Cylinders with SC A valves in the year 2010-11, the tender No. being LPG-O/M/PT-03/09- 10. While considering the Director General’s investigation report in Case No. 10 of 2010, the CCI in pursuance of its duties under Section 18 felt that investigation was necessary in the case of all bidders who were the suppliers of 14.2 kg. LPG cylinders in that tender. In the investigation B

report in the said case, the Director General had noted that out of 63 bidders who participated in the tender, 50 bidders were qualified for opening of price bids, while 12 bidders were qualified as new vendors who were not required to submit price bids and one bidder was not qualified for the opening of the price bid. The technical bid of the subject tender was opened on 3.3.2010 and the price bids of 50 qualified bidders C were opened on 23.3.2010. According to the Director General, there was a similar pattern in the bids by all the 50 bidders who submitted price bids for various States. The bids of a large number of parties were exactly identical or near to identical for different States. The Director General had observed that there were strong indications of some sort of D agreement and understanding amongst the bidders to manipulate the process of bidding. It was on this basis the CCI directed further investigation in the matter. The Director General after careful consideration submitted a detailed investigation report to the CCI. After the CCI considered the Efreshly ordered investigation report, it directed that a copy of the report be sent to the parties seeking their objections. In all, 44 opposite parties submitted their objections. After giving them the opportunity to be heard, the CCI passed the order in question. As per the Director General’s report, the process of bidding Ffollowed by the IOCL in the tender was as under :-

i) The bidders would submit their quotations with the bid documents. ii) The existing bidders, who were existing suppliers, were required to submit the price bids and technical bids. G iii) The bidders were to quote for supplies in different States of India in keeping with their installed capacity. iv) After price bids were opened the bidders were arranged according to the rates in the categories of L-1, L-2 and L-3. H 506 SUPREME COURT REPORTS [2018] 12 S.C.R. A v) The rates for the supplies in different States were approved after negotiations with L-1 bidder. In case the L-1 bidder could not supply a required number of cylinders in a particular State, the orders of supplies went to L-2 and also L-3 bidder or likewise depending upon the requirement in that State as per fixed formula provided in the bid documents. B vi) Certain bidders were called new parties. They were required to submit only technical bids and to supply as per L-1 rates determined after the negotiations. vii) One bidder could quote for maximum eight States.

C 7. The Director General after analyzing the bids came to the conclusion that there was not only a similarity of pattern in the price bids submitted by the 50 bidders for making supply to the IOCL but the bids of large number of parties were exactly identical or near to identical in different States. It was also found that bidders, who belonged to same group, might have submitted identical rates. It was found that not only D there was identical pricing in case of group concerns but the rates of other entities not belonging to the group were also found to be identical. The D.G. painstakingly noted the names of group companies as well as non-group companies. He came to the conclusion that in all 37 entities could not be said to be belonging to any single group and were E independently controlled. The Director General found it unusual that unrelated firms had quoted identical rates in different States. The D.G. had analyzed the bidding pattern for the various parties for all the 25 States. He found that :- a. The orders were placed on all the 50 successful bidders. F b. The contracts were awarded to the sets of bidders who had quoted identical rates or near to identical rates in a particular pattern in almost all the States. c. There was a common pattern for quotation depending upon the State. In case of North East the rates were highest, quoted at Rs. G 1240 whereas in case of others rates were Rs.1100, Rs.1127 and Rs. 1151.

d. It was found that only for Andaman and Nicobar Islands there was a single party who had quoted the L-1 rate and got the formal H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 507 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] contract. In other States the contracts were bagged in a group on A the basis of identical or near to identical rates.

e. The similarity of the rates was found even in case of bidders whose factories and offices were not located at one and the same place in the States and where they were required to supply was far off from their factories located in different place. B 8.The D.G. had found further that though the factors like market conditions and small number of companies were different, there was a large scale collusion amongst the bidding parties. He also arrived at a finding to the effect that the LPG Cylinder Manufacturers had formed an Association in the name of Indian LPG Cylinders Manufacturers C Association and the members were interacting through this Association and were using the same as a platform. The date for submitting the bids in the case of the concerned tender was 3.3.2010 and just two days prior to it, two meetings were held on 1st and 2nd March, 2010 in Hotel Sahara Star in Mumbai. As many as 19 parties took part and discussed Dthe tender and, in all probability, prices were fixed there in collusion with each other. The D.G. reported that the bidders had agreed for allocation of territories, e.g., the bidders who quoted the bids for Western India had not generally quoted for Eastern India and that largely the bidders who quoted the lowest in the group in Northern India, had not quoted generally in Southern India. The D.G. also concluded that this behavior E created entry barrier and that there was no accrual of benefits of consumers nor were there any plus factors like improved production or distribution of the goods or the provision of services.

9. Ultimately, the D.G. came to the conclusion that there was a cartel like behavior on the part of the bidders and that the factors F necessary for the formation of cartel existed in the instant case. It was also found that there was certainly a ground to hold concerted action on the part of the bidders. The D.G. had also noted that the rates quoted for the year 2009-10 and in years previous to that were also identical in some cases. Thus, he came to the conclusion that the bids for the year G2010-11 had been manipulated by 50 participating bidders. It was thereafter that the CCI decided to supply the D.G.’s investigation report to the concerned parties and invite their objections. 10. A common reply came to be filed as also the individual replies. After considering the same, the CCI formulated the following issue for determination:- H 508 SUPREME COURT REPORTS [2018] 12 S.C.R. A “Whether there was any collusive agreement between the participating bidders which directly or indirectly resulted in bid rigging of the tender floated by IOCL in March 2010 for procurement of 14.2 kg. LPG cylinders in contravention of Section 3(3)(d) read with Section 3(1) of the Act?”

B 11. After considering the oral as well as written submissions, the CCI answered the issue against the Cylinders Manufacturers and inflicted the penalties against the present appellants. In its impugned order, while determining the issue, the CCI, in the first instance, considered the common replies to the DG’s report filed by as many as 44 opposite parties. It was more or less pleaded that every part of LPG Cylinder is C regulated by the Rules through various Notifications and that the price of steel constitutes 50% of the total manufacturing cost, so also the price of the paint, it being an essential raw material. All these factors, including the taxes which vary from State to State, determine the overall bidding pattern of the bidders. In para-5.2.3 of the common objection, it D was added that these 44 parties had nominated six agents for depositing their bids on their behalf and it was a common practice amongst the bidders to direct their agents to keep close watch on the rates offered by their competitors in respect of a particular State and this led to the possibility of copying and matching of the rates quoted in the price bids by many suppliers in a particular State, who may have appointed common E agents. Due to this reason, cutting and over-writing in the price bids for the tender in question was noticed by the Director General.

12. It was further pointed out that there were only 62 qualified tenderers in the whole country, out of whom 12 bidders were classified as new parties, meaning thereby that they had not supplied Cylinders in F last three years and were not required to bid in the tender. Out of the remaining 50 bidders, there were group companies controlled by single management. 13. The CCI in its detailed order began with considering the scope of constructed bid rigging agreement and cartel. In that the CCI also G considered the 18 famous observations by Lord Denning in case of RRTA vs. W. H. Smith & Sons Limited regarding the quiet and secret nature of the agreement between the parties. The CCI then went on to record its inference holding that there was element of agreement and considered the following factors in coming to the conclusion. They being:-H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 509 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] 1. Market conditions A 2. Small number of suppliers 3. Few new entrants 4. Active trade association 5. Repetitive bidding B 6. Identical products 7. Few or no substitutes 8. No significant technological changes C 9. Meeting of bidders in Mumbai and its agenda. 10. Appointing common agents 11. Identical bids despite varying cost.

14. After consideration of these factors, the CCI came to the conclusion that it did suggest collusive bidding. Thereafter, the CCI D analyzed these bids for each States and found that all 50 participating bidders had secured the order; that the orders were placed on the said 50 bidders who had quoted identical rates or near to identical rates in a particular pattern common to all the parties. CCI also highlighted the facts of absence of business justification. According to the CCI, the E material revealed that the supplies were effected at the higher cost. After discussing the concepts of standards of proof and appreciable adverse effect on competition, the CCI considered the various arguments and repelled those arguments. The CCI then went on to consider the case law, and in particular the judgment of this Court in Union of India vs. Hindustan Development Corporation1. It also took into consideration F the arguments raised by the individual parties and then came to record that cases of M/s. JBM Industries and Punjab Cylinders, however, were exceptional ones and they could be exonerated. After this the CCI went on to decide the penalty factor under Section 27 of the Act. 15. The COMPAT after discussing the findings of the CCI and G also taking note of the arguments of the appellants which were advanced before the CCI, proceeded with its own discussion. It started with the admitted facts of the case, and took note of the following such facts: 1 (1993) 3 SCC 499 H

510 SUPREME COURT REPORTS [2018] 12 S.C.R. A (A) The tender offers were to be made at Mumbai on 03.03.2010. Admittedly there were meetings in Hotel Sahara Star, Mumbai on 1st and 2nd March, 2010 which were attended by some of the appellants. The D.G. has held that 19 appellants were represented by various persons in that meeting. The fact of the meeting having been held was not disputed. B Though some of the appellants stated that they did not attend the meeting and those who attended the meeting maintained that nothing was disucssed about the tender, the same was not believed by the COMPAT and it held that these meetings did relate to the tender offers which were to be submitted on 03.03.2010. This C finding is premised on the basis that nobody came with the explanation as to what transpired in the meeting or gave any proof that prices were discussed. Minutes of the meeting were also not produced. (B) There is an association of the cylinder manufacturers. All the D parties, except few competing with each other, stated that they were not the members of that association. A feeble argument was also raised by some appellants that though they were the members but they were not the active members thereof. Some of the appellants also argued that they had abandoned the membership E by not contributing the subscription in the later years. However, the appellants could not deny the position that there was an association called Indian LPG Cylinder Manufacturers’ Association.

It was a registered association, its Memorandum of Association provided that one of the objectives was to prtoect F common interest and welfare of LPG cylinder manufacturers. According to COMPAT, there was a definite platform available for all cylinder manufacturers and practically all the appllants appear to be the members of that Association. (C) A common written reply was submitted by as many as 44 G parties. Further, the appellants had nominated six agents for depositing bids on their behalf. These common agents were instructed to keep a close watch on the price quoted by the competitors in a particular State. H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 511 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] Though some of the appellants had contended that they had A not appointed the common agents, the plea was not accepted by the COMPAT. The COMPAT, therefore, proceeded on the ‘admitted grounds’ that there was an association of cyliner manufactures; practically all the appellants were members of the said association; this association was an active association; it held B

meetings on the eve of entry tender obviously for discussing tenders, its conditions etc.; these meetings were attended by representatives of at least 19 appellants; and these appellants had six common agents at Mumbai who were instructed to watch the prices offered by the others. A dinner meeting as also a lunch were held and one Mr. Chandi Prasad Bhartia of M/s. Haldia C Precision Engineering Private Limited paid the bill for the same. Dinner and lunch held in Sahara hotel were attended by about 50 persons in all. From this the COMPAT inferred that there was no reason to disbelieve that the parties had an access to each other through their association which was an active association. The D existence of such an association under the aegis of which meetings took place just before the submission of tender has been noted as a very relevant factor by the COMPAT in affirming the findings of CCI on cartelisation and it summed up the position in the following manner:

E “26. What is important is not whether a particular appellant was a member of the association or not. The existence of an association is by itself sufficient, as it gives opportunity to the competitors to interact with each other and discuss the trade problems. There will be no necessity to prove that any party actually discussed the prices by actively taking part in the F meeting. If there is a direct evidence to that effect that is certainly a pointer towards the fact that such party had a tacit agreement with its competitors. However, the existence of an association and further holding of the meetings just one or two days prior to the last date of making offers and further G admission that the parties had appointed common agents with the instructions to keep watch on the prices quoted by the competitors would go a long way in providing plus factors in favour of the agreement between the parties. All these factors would form a back drop, in the light of which, the further H 512 SUPREME COURT REPORTS [2018] 12 S.C.R. A evidence about agreement would have to be appreciated. We have seen the comments of Director General as also the findings of the CCI. We are convinced that CCI has not committed any error in considering all these factors as plus factors to come to the conclusion that there was a concerted agreement between the parties on the basis of which the B

identical or near identical prices came to be quoted in tenders for the supply of cylinders to the 25 States. In view of this, we need not dilate on the individual claims by some of the appellants that they were not the members of the association or that they were only the dormant members or that they had abdicated C their membership. We also need not go on the claim that while the meeting was attended by the 19 parties as held by the D.G. and confirmed by the CCI, it was not attended by the rest of the appellants because that would be of no consequence. Once there was a meeting, there was every opportunity to discuss or to communicate to each other whatever transpired in the D meeting. 27. We have seen the order of the CCI and while commenting about the meeting, the CCI has painstakingly noted the details of that meeting. The CCI has referred to the evidence of Mr. Dinesh Goyal, who was an active member of the Indian LPG E Cylinder Manufacturers’ Association and noted that he had attended the meeting. He has also referred to the statement of Mr. Sandeep Bhartia of Carbac Group though initially he denied to have organized the conference, he later on had confirmed about such a conference having been held along with Mr.

F Sandeep Bhartia of Carbac Group. The CCI also noted that he admitted that in such meetings there were discussions on pre-bid issues. He also admitted that though there are about 50 competitors, in fact about 25 persons control the whole affairs. From this evidence, the CCI correctly deduced that pre-bid issues were discussed in that meeting. The CCI has G then referred to the evidence of Mr. Manvinder Singh of Bhiwadi Cylinders Limited, Mr. Chandi Prasad Bhartia of Haldia Precision Engineering P. Ltd., Mr. Vijay Kumar Agarwal of SM Sugar Pvt. Ltd., Mr. S. Kulandhaiswamy, MD of Lite Containers Pvt. Ltd. and Secretary of the Association, Mr. H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 513 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] Ramesh Kumar Batra, Director of Surya Shakti Vessels Pvt. A Ltd. and on that basis came to the correct conclusion that not only was the meetings held on 1st and 2nd March, but thorough discussions went on in those meeting on the pre-bid issue of the concerned tender. The CCI has also correctly noted about the agenda of the meeting and has also referred to an admission B made by one of the witnesses that the matching of the quotation was a matter of co-incidence and telephonic discussions do take place amongst the parties regarding the trends. We are thus thoroughly convinced about holding of the meeting, the discussion held therein and also the discussion regarding the pre-bid issue having been taken place in that meeting.” C

16. The COMPAT thereafter took up for discussion the argument of the appellants that the CCI should have enquired IOCL also. But rejected the same. Another significant argument which was canvassed before us also with great emphasis was that it was an oligopolistic market wherein there was a likelihood of each player being aware of actions of D the other and in such a situation price parallelism would be a common phenomena. Thus, merely because there was a price parallalism, it would not be construed as evidence of collusion. The COMPAT rejected this argument as well. In the process, it analysed the order of CCI, conclusion whereof was founded on the following factors: E (1) The prevailing market conditions were such that there was a constant demand for cylinders not only by IOCL but by other two oil manufacutring companies as well. Therefore, there was a constant need for the cylinders which facilitated factor for the collusion. F (2) There was small number of suppliers. Among the 50 participating companies, only 37 companies could be said to be independant bidding companies and there were seven groups consisting of 20 participating companies. This small number of suppliers should also be a facilitating factor. G (3) There were very few new entrants. (4) The existence of an active trade association in which all the bidders, except seven companies, were members would be another facilitating factor. H 514 SUPREME COURT REPORTS [2018] 12 S.C.R.

A (5) Few other factors like repetitive bidding, identical products, few or no substitutes and no significant technological changes were the additional factors which persuaded the CCI to arrive at such a conclusion. (6) These manufacturing companies had their factories at different B places in India, where the costs of the components would differ from State to State. Even the taxing structure, the labour conditions and other factors like cost of electricity etc. were bound to be different. Still the prices quoted were almost identical. (7) On the above considerations, the defence of the appellants C was rejected as unconvincing, thereby undergoing the factors considered by the CCI. 17. According to the COMPAT all these could not have been possible unless there were internal agreements between the appellants. The COMPAT has approved the finding of the CCI that owing to the collusion, the IOCL could not get lower or the competitive prices. The D rates quoted in 2010-2011 were higher as compared to the rate quoted in 2009-10. From the year 2006-07, the prices had collectively been raised on an average of 30% for making supplies in different states.

18. According to the COMPAT, the CCI was right in concluding that it had appreciable adverse effect on competition as the conduct of E the LPG cylinder manufacturers in coming together on a common platform and fixing the bid prices ensures that no new player could enter the relevant market and quote the prices independently. Thus, these manufacturers would make entry of a new player into the relevant market difficult, because such new player would necessarily have to first F negotiate with the existing players to get the business profitably. Other factors were driving existing competitiors out of the market and foreclosure of competition by hindering entry into the market. 19. It negated the argument of the appellants that when the IOCL was placing orders on the basis of negotiated rates there could be no G possibility of incentive to collude. According to it, even where the rates are fixed, the bid rigging can still take place to keep the big amounts to a pre-determined level. Such pre-determination can be by way of intentional manipulation by members of the bidding group and where the L-1 rates themselves get fixed like in the present case at higher level even if there are negotations the negotiators would have to take into consideration the H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 515 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.]

benchmark rates. There is aslo a possibility that such benchmark rates A could go higher in the subsequent tenders; known as rippel effect in long term. 20. The COMPAT also took note of the provisions of Section 3, as per which once the agreement is proved there is a presumption about the appreciable adverse effect on competition on the mere proof of the B agreement. Thus, onus shifts on the other side to prove otherwise which according to the COMPAT was not discharged by the appellants. The COMPAT thereafter took note of some arguments by certain counsel specific to their cases but did not find any substance in them. 21. Having examined the relevant provisions whereupon these C appeals centre around, we proceed to take note of the arguments that were advanced by various counsel appearing for the appellants and the manner in which respondents endeavoured to meet the same. 22. Ms. Madhavi Divan, learned counsel appearing in the appeal filed by Rajasthan Cylinders and Containers Ltd., attacked the very basis D and foundation on which CCI came to conclusion that there was an agreement or cartelisation by the appellants aimed at bid rigging. She premised her case on the following three propositions: (i) the inherent nature of the market of cylinder manufacturers itself precludes the possibility of competition; E (ii) alternatively, there is no collusive agreement or bid-rigging in the present case; and

(iii) further, in the alternative, even assuming that there is a collusive agreement or bid-rigging in the present case, there is no appreciable adverse effect on competition. F 23. On the first proposition, argument developed by Ms. Divan was that the Act prohibits anti-competitive practices, which would imply that there has to be a competition in the market, in the first place. As a corrolary, if there is no such competition, Section 3(1) of the Act does not get triggered. According to her, in the instant case, the fact would G show that there was a tight control and regulation by the IOCL and, thus, it did not lead any scope of competition at the very threshold. She stressed that the conditions of monopsony/oligopsony prevailed. For the existence of monopsony/oligopsony, she referred to the Glossary of Industrial Organization Economics and Competition Law published by H 516 SUPREME COURT REPORTS [2018] 12 S.C.R. A the Organisation for Economic Co-operation and Development (OECD), as per which a monopsony consists of a market with a single buyer. When there are only a few buyers, the market is described as an oligopsony. In general, when buyers have some influence over the price of their inputs they are said to have monopsony power. The ability of a firm to raise prices, even when it is a monopolist, can be reduced or B eliminated by monopsony or oligopsony buyers. To the extent that input prices can be controlled in this way, consumers may be better off.

24. According to her, these conditions were adequately present in the instant case. In her attempt to make this propositoin good, she highlighed the following features and conditions surrounding the contract: C (i) Extremely limited number of buyers and for this particular kind of market - a sole buyer, i.e., IOCL. IOCL controls 48% of the market share. There are no other purchasers of 14.2 Kg gas cylinders except for HPCL and BPCL, both of whom invite e-tenders, having a market share of 26% and 25% respectively. D (ii) The product is standardized and special to the extent that it is tightly controlled and regulated by the Government and also there are no other takers for it. (iii) There are entry barriers in the market. As per the Tender conditions, only those manufacturers having valid approval from E the Chief Controller of Explosives (CCOE) and Bureau of Indian Standards (BIS) license for manufacture of 14.2 kg LPG cylinders as per IS-3196 (Part 1) could submit bids for the tender. (iv) Even the machinery used to manufacture this product is special and will become obsolete and reduceable to scrap if IOCL and F the aforesaid two players were to discontinue contracts for supply of 14.2 kg cylinders. She pointed out that this was accepted in the Expert Report of Dr. Rughvir K.S. Khemani. (v) The tender conditions state that it can be rejected without furnishing reasons. Therefore, the lowest price is not sacrosanct G (clause 11 of the contract).

(vi) L2 and L3 have also been granted contracts irrespective of the price they have quoted. (vii) Effective price has no sanctity since not only L2 and L3 also H get contracts in addition to or in exclusion of L1 but further, the RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 517 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] final negotiated price is determined on the basis of privately A conducted negotiations with individual bidders for which the benchmark is not the price quoted by them but the internal estimates arrived on the basis of objective criteria.

(viii) In most States, the final negotiated price was concluded at a rate lower than the internal estimate. The internal estimate had B absolutely no correlation with the quoted rates by L1 or any other party. In this behalf, she pointed out that the IOCL had carried out the exercise of ascertaining the estimated cost of the cylinder through its experts. In the report given by the expert, the estimated cost per cylinder was arrived at Rs. 1106.61 paisa per cylinder. C As against this, the final negotiated price at which the appellants had supplied cylinders to the IOCL was much lesser. According to her, in the whole process the price determination was on the basis of internal estimates by IOCL which could not be influenced by the appellants at all. In fact, even after the tenderers submitted their bids, final price was the price negotiated by IOCL which D fact was accepted by Mr. Y. Ramana Rao of IOCL in his deposition recorded by the Director General of CCI. This, according to the learned counsel, clearly proved that there was no adverse effect on competition, in any case. (ix) The internal estimates were drawn up long after the price E bids were made, i.e., on 5th May, 2010. Price bids were opened on 23rd March, 2010 and negotiations were held only after the submission of Mott MacDonald Report on 05.05.2010.

(x) The pattern shows that since L1, L2 and even L3 were awarded the contract and not merely L1, quoting the lowest price did not F even determine the identity of the parties who were to get the contract, therefore, the manner in which the process was conducted or controlled by IOCL, completely leaves no scope for either determination of price or the identity of the parties who would get the contract. G 25. She submitted that in such market conditions where on account of the vertical agreement there is virtually no scope of competitive forces between horizontal players, the question of anti-competitive conduct by virture of horizontal agreements does not arise. There is no competition in the market even before a player enters the fray. Therefore, the first H 518 SUPREME COURT REPORTS [2018] 12 S.C.R. A premise for the applciation of Section 3, i.e., the presence of an otherwise competitive market is absent. The burden of proof is on the respondent— CCI to establish that there is competition in the market before it can justify invoking Section 3. There is no automatic presumption under Section 3 that there is competition in the market. B 26. From the aforesaid factors, Ms. Divan tried to deduce that price control was entirely in the hands of IOCL and in a situation like this, question of entering into any agreement with the motive of bid rigging or collusive bidding did not arise.

27. She also referred to LPG (Regulation of Supply and C Distribution) Order, 2000 published vide Notification dated 26th April, 2000 as per which only Government oil companies can supply LPG to domestic consumer of 14.2 kg LPG cylinders with dimensions as specified therein. Predicated thereupon, her submission was that the LPG supply in 14.2 kg gas cylinders is an essential commodity; the distribution of such cylinders takes place only through Government oil companies; the D price to the consumer is controlled by the Government; and parallel marketeers, supplier and distributor of LPG cylinders may do so only for cylinders and specifications other than 14.2 kg cylinders. This control of the Government, insofar as supply of 14.2 kg gas cylinders is concerned, would also show tight control over the pricing. In such a statutorily tight E control price fixing mechanism there could not be bid rigging, was the submission of Ms. Divan. She supported this submission by drawing the attention of the Court to the following observations in Ashoka Smokeless Coal India (P) Ltd. v. Union of India2 :

“109. It may be true that prices are required to be fixed having regard to the market forces. Demand and supply is a relevant F factor as regards fixation of the price. In a market governed by free economy where competition is the buzzword, producers may fix their own price. It is, however, difficult to give effect to the constitutional obligations of a State and the principles leading to a free economy at the same time. A level playing field is the key G factor for invoking the new economy. Such a level playing field can be achieved when there are a number of suppliers and when there are competitors in the market enabling the consumer to exercise choices for the purpose of procurement of goods. If the 2 (2007) 2 SCC 640 H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 519 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] policy of the open market is to be achieved the benefit of the A consumer must be kept uppermost in mind by the State. xxx xxx xxx

127. While fixing a fair and reasonable price in terms of the provisions of the Essential Commodities Act (although the price is B not dual), it is essential that price is actually fixed. Such price fixation is necessary in view of the fact that coal is an essential commodity. It is, therefore, vital that price is actually fixed and not kept variable. Fixation of price of coal is of utmost necessity as it is a mineral of grave national importance. Non-availability of coal and consequently, the other products may lead to hardship to C a section of citizens. It may entail closure of factories and other industries which in turn would lead to loss to the State exchequer; as they would be deprived of its taxes. It will lead to loss of employment of a large number of employees and would be detrimental to the avowed object of the Central Government to D encourage small-scale industries.” 28. She also referred to the following discussion in Excel Crop Care Limited v. Competition Commission of India & Anr.3: “52. We are here concerned with parallel behaviour. We are conscious of the argument put forth by Mr Venugopal that in an E oligopoly situation parallel behaviour may not, by itself, amount to a concerted practice. It would be apposite to take note of the following observations made by European Court of Justice in Dyestuffs [Imperial Chemical Industries Ltd. v. Commission of European Communities, 1972 ECR 619 (ECJ)] : F

“By its very nature, then, the concerted practice does not have all the elements of a contract but may inter alia arise out of coordination which becomes apparent from the behaviour of the participants. Although parallel behaviour may not itself be identified with a concerted practice, it may however amount to strong evidence of such a practice if it leads to G conditions of competition which do not respond to the normal conditions of the market, having regard to the nature of the products, the size and number of the undertakings, and the 3 (2017) 8 SCC 47 H 520 SUPREME COURT REPORTS [2018] 12 S.C.R. A volume of the said market. Such is the case especially where the parallel behaviour is such as to permit the parties to seek price equilibrium at a different level from that which would have resulted from competition, and to crystallise the status quo to the detriment of effective freedom of movement of the products in the [internal] market and free choice by consumers B of their suppliers.” (emphasis supplied) At the same time, the Court also added that the existence of a concerted practice could be appraised correctly by keeping in mind the following test: C “If the evidence upon which the contested decision is based is considered, not in isolation, but as a whole, account being taken of the specific features of the products in question.”

29. The learned counsel also referred to various judgments of other jurisdictions, primarily that of European Commission and the Court D of Justice of European Union, which we shall discuss at the appropriate stage. 30. Ms. Divan, also highlighted that in this entire scenario, it was necessary to have the views of IOCL. However, in a suo motu case, IOCL was not even served with any notice and therefore no evidence E was elicited from IOCL on the issue whether there was any autnomy left to the manufacturer in the matter of price determination. 31. She, thus, argued that merely because there was price parallelism, it could not have been the reason to arrive at a conclusion that there was a collusive agreement or bid rigging. She submitted that F in a monopsonistic market where there are few buyers, the price is set by the buyers, and the conditions are such that sellers can predict demand, there is a repetitive bidding process and the products are identical and specilized, the likelihood of price parallelism is natural. 32. Further, price parallelism is inevitable where the buyer has a G high degree of control and determines price, quantity, and even the identities of the awardees at its discretion. Referring to the following discussion in Union of India v. Hindustan Development Corporation4, she argued that mere identical pricing cannot lead to the conclusion of cartelisation: H 4 (1993) 1 SCC 467 RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 521

UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] “7. [….] (1) There is not enough of material to conclude that M/ A s H.D.C., Mukand and Bhartiya formed a cartel. Because of mere quoting identical tender offers by the said three manufacturers for which there is some basis, the conclusion that the said manufacturers had formed a cartel does not appear to be correct. However since the offers of the said three tenderers B were identical and the price was somewhat lower, the Tender Committee entertained a suspicion that a cartel had been formed and the same got further strengthened by the post-tender attitude of the said manufacturers which further resulted in entertaining the same suspicion by the other authorities in the hierarchy of decision making body including the Minister of Railways. [….] C 33. She pointed out that this principle has also been stated in paragraph 17 of the Union of India v. Hindustan Development Coproration5. Her submission was that identical pricing may be further explained by the fact that, given the high degree of predictability of prices, bidders may take a business decision to mirror prices of competitors in D certain States, by adjusting or averaging prices in others. 34. The learned counsel pointed out that the CCI arrived at an inference of a collusive agreement based, inter alia, on the presence of circumstances which have acted as ‘facilitating factors’ for collusion. These factors which describe the nature of the industry are: E

(i) Predictability of demand (ii) Small number of suppliers (iii) Few new entrants (iv) Active trade association F (v) Repetitive bidding (vi) Identical products (vii) Few or no substitutes (viii) No significant technological changes, i.e, a standardised G product in repsect of which there has been no change or alteration in design. 35. Her reply was that these are the characteristics which define the industry. Yet these very factors are relied upon to come to the 5 (1993) 3 SCC 499 H 522 SUPREME COURT REPORTS [2018] 12 S.C.R. A consclusion that there is ‘collusion’ and ‘bid rigging’. She submitted that if the very nature of the industry is such that there are very small numbers of suppliers, very few new entrants and a standard product being supplied to the same party year after year, such factors are beyond the control of the individual manufacturers and cannot be relied upon as factors to lead to a presumption that there is collusive conduct. In other words, the B very nature of the industry cannot be used as a factor to presume collusion because collusion itself requires a state of mind or intent whereas in this case, most of these factors are inherent in the nature of the industry as described by the CCI itself.

36. Adverting to her 2nd proposition, namely, there was no collusive C agreement or bid rigging in the present case, her submission was that CCI has relied on a dinner attended by some manufacturers on 1st March, 2010 and a lunch on 2nd March, 2010 as evidence of a price fixing agreement. Her response was that the factum of meetings of an association by itself in any case cannot lead to a conclusion of collusion. D Likewise, the COMPAT also upheld that inference based on the factum of the meetings of the Associaiton. The COMPAT went to the extent of holding that it is irrelevant whether a particular party was a member of the Association or not and the existence of Association is by itself sufficient. This approach was attacked as contrary to the fundamental right to form an association under Article 19(1)(c)(g) of the Constitution E of India. 37. So far as the meetings over dinner and lunch are concerned, both were hosted by individual members. In the case of the dinner meeting on 1st March, 2010, it was hosted by Mr. C.P. Bhartiya, MD of North India Wires. The lunch on 2nd March, 2010 was hosted by Mr. F Santosh Bhartiya of Haldia Precision. It is not as if that the Association paid or the expenses were shared by all members who attended.

38. She also submitted that insofar as appellant – Rajasthan Cylinders and Containers Limited is concerned, no representatives of appellant attended the said meeting. Further, many other members did G not attend the meeting. Even as per the findings of the Director General, only 12 persons representing 19 parties are said to have attended the meeting. Her submission was that as per the allegations, 45 persons had entered into an agreement of cartelisation which should not be established only with the said meeting which was not attended by all and in fact very few members. H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 523 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] 39. In any case, according to her, it was expressly stated by at A least two persons who attended the meeting that price was not discussed. These are Mr. Chandi Prasad Bhartia from Haldia Precision and Mr. Manvinder Singh of Bhiwadi Cylinders Ltd. 40. Her further submission on this aspect was that the inference drawn on the basis of six agents being nominated for depositing 44 bids B was also misconceived. The CCI holds that ‘this might have held to the possibility of copying and matching of the rates quoted in the price bids by many suppliers in a particular state.” 41. The test as laid down in the case of Ahlstrom Osakeyhtio v. Commission6 is: Is the concertation the only plausible explaination for C the conduct?

“126. Following that analysis, it must be stated that, in this case, concertation is not the only plausible explanation for the parallel conduct. To begin with, the system of price announcements may be regarded as constituting a rational response to the fact that the D pulp market constituted a long-term market and to the need felt by both buyers and sellers to limit commercial risks. Further, the similarity in the dates of price announcements may be regarded as a direct result of the high degree of market transparency, which does not have to be described as artificial. Finally, the parallelism E of prices and the price trends may be satisfactorily explained by the oligopolistic tendencies of the market and by the specific circumstances prevailing in certain period. Accordingly, the parallel conduct established by the Commission does not constitute evidence of concertation. This test is not met in the present case for reasons that are F enumerated.

42. Her third proposition was that in any case there was no appreciable adverse effect on competition. She tried to make this submission good by contending that when industry is an oligopoly, the price parallel or a finding of identical quoting of price does not by itself G lead to the conclusion of a conerted price. Moreoever, in the instant case, number of entrants had increased as 12 new entrants submitted their bid for the year 2010-11. Therefore, the finding of the CCI, upheld 6 31.3.1993, ECJ (paragraph 115, Internal P.1611) (“Woodpulp“) H 524 SUPREME COURT REPORTS [2018] 12 S.C.R. A by the COMPAT, that there has been a creation of barriers for new entrants is without any basis.

43. Other counsel who appeared on behalf of the appellants made their submission almost on the same lines, albeit, with further elaborations on certain aspects, some of which are taken note of hereafter. Mr. Jaiveer Shergil, who argued for the appellant—Om Containers (C.A. B No. 6369 of 2014) submitted that in order to attract the presumption contained in Section 3(3) about the appreciable adverse effect on competition, in the first instance, there has to be a finding that there has been an agreement of the kind set out in Section 3(3)(a) to (d). Since, the allegation against the appellants was that the agreement resulted in C bid rigging and case is covered under Section 3(3)(d) of the Act, it was necessary that there is a positive finding to the aforesaid effect, namely, that there was agreement which had resulted in bid rigging. Accordoing to him, since the definition of bid rigging in Explanation to Section 3(3) uses the words ‘means’, the definition is a hard and fast definition and no other meaning can be assigned to the expression than is put down in D the definition, as held in Punjab Land Developement & Reclamation Corporation Ltd. vs. Presiding Officer, Labour Court7in the following words:

“72. The definition has used the word ‘means’. When a statute says that a word or phrase shall “mean”— not merely that it shall E “include” — certain things or acts, “the definition is a hard-and-fast definition, and no other meaning can be assigned to the expression than is put down in definition” (per Esher, M.R., Gough v. Gough [(1891) 2 QB 665 : 65 LT 110] ). A definition is an explicit statement of the full connotation of a term.” F 44. Thus, according to him, for it to be a case of bid rigging, the agreement must be such which is defined in the Explanation to Section 3(3)(d) creating the effect of: a. Eliminating or reducing competition for bids or b. Adversely affecting the process for bidding or G c. Manipulating the process for bidding. 45. He referred to the judgment in S. Sundaram Pillai vs. V.R. Pattabiraman8, on the purpose of an ‘Explanation’, viz.: 7 (1990) 3 SCC 682 H 8 (1985) 1 SCC 591 RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 525 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] “46. We have now to consider as to what is the impact of the A Explanation on the proviso which deals with the question of willful default. Before, however, we embark on an inquiry into this difficult and delicate question, we must appreciate the intent, purpose and legal effect of an Explanation. It is now well settled that an Explanation added to a statutory provision is not a substantive B

provision in any sense of the term but as the plain meaning of the word itself shows it is merely meant to explain or clarify certain ambiguities which may have crept in the statutory provision. Sarathi in Interpretation of Statutes while dwelling on the various aspects of an Explanation observes as follows: C (a) The object of an Explanation is to understand the Act in the light of the explanation. (b) It does not ordinarily enlarge the scope of the original section which it explains, but only makes the meaning clear beyond dispute.” D 46. He submitted that there is no positive evidence of this nature at all and the CCI as well as COMPAT has proceeded on inferences as regards bid rigging and, therefore, such orders cannot be sustained. 47. In the absence thereof, submitted the learned counsel, doctrine of reverse burden which was put on the appellants would not apply. He E referred to the following judgments9 in support. 48. The counsel relied upon the following observations in CCI v. Artistes & Technicians of W.B. Film & Television:

“31. The Competition Act, 2002, as amended in 2007 and 2009, deals with anti-trust issues viz. regulation of anti-competitive F agreements, abuse of dominant position and a combination or acquisition falling within the provisions of the said Act. Since the majority view of CCI also accepted that the impugned activities of the Coordination Committee did not amount to abuse of dominant position, and it treated the same as anti-competitive having G appreciable adverse effect on competition, our discussion would be focused only on anti-competitive agreements. Section 3 of the 9 (i) Babu v. State of Kerala: [(2010) 9 SCC 189] (ii) Noor Aga v. State of Punjab; [(2008) 16 SCC 417] (iii) CCI v. Artistes & Technicians of W. B. Film & Television: [(2017) 5 SCC 17] H 526 SUPREME COURT REPORTS [2018] 12 S.C.R. A Act is the relevant section in this behalf. It is intended to curb or prohibit certain agreements. Therefore, in the first instance, it is to be found out that there existed an “agreement” which was entered into by enterprise or association of enterprises or person or association of persons. Thereafter, it needs to be determined as to whether such an agreement is anti-competitive agreement B within the meaning of the Act. Once it is found to be so, other provisions relating to the treatment that needs to be given thereto get attracted.”

49. Taking aid of the aforesaid legal principle, it was submitted that in the present case it will be seen that the CCI, rather arriving at a C finding with focus on the aforesaid factors, proceeded to analyse factors which attach to the general market conditions of the industry to ‘infer’ the ‘possibility’ of bid rigging and then concluded that the ‘facilitating factors’ which may be ‘considered conducive for cartelisation’ are present. The D.G. found that ‘in all the probability, prices were fixed D there at the meeting in Bombay in collusion with each other. Such an inference and assumption based on ‘higher chances’, ‘probability’, ‘tendencies’ or ‘likelihood’ by the CCI does not meet the requirement of the definition contained in Explanation to Section 3(3) and certainly does not constitute a finding of ‘bid rigging’ as defined therein. The Tribunal has also proceeded on the basis that it ‘is to be deduced....that these E meetings did relate to the tender offers’. There was, thus, not clearcut, precise and consistent evidence to support that the alleged bid rigging took place.

50. Next submission of Mr. Shergil was that apart from the complete absence of a finding of bid rigging, in the present IOCL tender, as a F matter of fact there canot be any bid rigging as defined in Section 3(3). To take the first ingredient, i.e., eliminating or reducing competition for bids, the report of D.G. itself finds that out of the 60 bidding parties 37 entities were not belonging to any single group and are independently controlled. Hence, straight away there is no case of ‘eliminating or G reducing competition for bids’ which is one of the possible ingredients of bid rigging as there were 37 entities who were free of mind to participate and bid of their own accord in the absence of any control by any cartel. 51. As regards the second and third requirement of bid rigging, i.e., adversely effecting or manipulating the bidding process, he argued H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 527 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] that the submission of bids by the appellant (even if identical) can have A no effect of ‘adversely effecting or manipulating the bidding process’ this being on account of the very nature of the present tender process. Although, bids are invited from bidders, IOCL has a fixed/base procurement price of Rs. 1106.61 per cylinder. IOCL then works out an estimated rate per State based on certain factors peculiar to that State B

such as octroi, freight etc. The bid offered by the L1 (lowest bidder) is then subject to further downward negotiations by IOCL as per the tender clause and a further finalised rate is arrived at. Such finalised rate is eventually even lower10 than the L1 bid amount. Thus, factually, logically and in reality any bid submitted by any party can never be one which is said to adversely affect or manipulate the bidding process. All of this C information is with IOCL as part of its bidding process preparations, estimates and financial workings and could easily have been taken into consideration. In support, Mr. Shergil also referred to the terms and conditions of the IOCL tender. 52. His further submission was that CCI, or for that matter D COMPAT, were wrong in getting influenced by the submissions of identical bids by the appellants as it could not be, ipso facto, inferance of bid rigging. Such identical prices could be for various reasons and he shared that the reasons given by Ms. Divan predicated her submissions on oligopsony/monopsony. In addition, he relied upon the judgment of Ethis Court in Union of India vs. Hindustan Development Corporation11 and, in particular, para 17 thereof, which is as under:

“17. Therefore, whether in a given case, there was formation of a cartel by some of the manufacturers which amounts to an unfair trade practice, depends upon the available evidence and the F surrounding circumstances. In the instant case, initially the Tender Committee formed the opinion that the three big manufacturers formed a cartel on the ground that the price initially quoted by them was identical and was only a cartel price. This, in our view, was only a suspicion which of course got strengthened by post-tender attitude of the said manufacturers who quoted a much G lesser price. As noticed above it cannot positively be concluded on the basis of these two circumstances alone. In the past these 10 With the sole exception of the Andaman & Nicobar Islands as it has unique implication of transport related costs. 11 (1993) 3 SCC 499 H 528 SUPREME COURT REPORTS [2018] 12 S.C.R. A three big manufacturers also offered their own quotations and they were allotted quantities on the basis of the existing practice. However, a mere quotation of identical price and an offer of further reduction by themselves would not entitle them automatically to corner the entire market by way of monopoly since the final allotment of quantities vested in the authorities who in their B

discretion can distribute the same to all the manufacturers including these three big manufacturers on certain basis. No doubt there was an apprehension that if such predatory price has to be accepted the smaller manufacturers will not be in a position to compete and may result in elimination of free competition. But C there again the authorities reserved a right to reject such lower price. Under these circumstances though the attitude of these three big manufacturers gave rise to a suspicion that they formed a cartel but there is not enough of material to conclude that in fact there was such formation of a cartel. However, such an opinion entertained by the concerned authorities including the Minister D was not malicious nor was actuated by any extraneous considerations. They entertained a reasonable suspicion based on the record and other surrounding circumstances and only acted in a bona fide manner in taking the stand that the three big manufacturers formed a cartel.” E 53. Some literature on the ‘theory of oligopolistic interdependence’ as well as judgments of the European Union and European Commission were also cited.

54. Mr. Pradeep Aggarwal, in addition, argued that though there was no positive finding of cartelisation and the conclusion was merely F presumptive, even if it is accepted that there was such an agreement of bid rigging or collusive bidding, there was no presumption of ‘appreciable adverse effect on competiiton’. In the alternative, he submitted that there was, in fact, no appreciable adverse effect on competition in the present case and the said presumption totally was rebutted by producing G sufficient evidence on record. 55. Various other counsel also argued on the same lines and in addition referred to facts or their specific cases and it is not necessary to state all those arguments to avoid repetition. 56. Per contra, Mr. Salman Khurshid, learned senior counsel H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 529 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] appearijng for CCI highlighted the purpose for which the Act is enacted A and, in particular, objective behind Section 3 of the Act, whch is taken note of by this Court in Excel Crop Care Limited as well as West Bengal Artists Association. Insofar as instant case is concerned, his submission was that it is a stark and clear-cut case of bid rigging as a result of anti-competitive agreement amongst LPG manufacturers in B

respect of a tender (Tender No. LPG-O/M/PT-03/09-10) floated by IOCL for procurement of approximately 1,05,00,000 (105 laks) LPG Cylinders. This is a matter of serious public concern because these cylinders were to be used to supply Liquefied Petroleum Gas (LPG) for domestic consumption across 25 States. A rise in price resulting from anti-competitive activities would affect the cost of living for the common C man, and has serious ramifications for the economy as a whole. 57. Mr. Khurshid referred to the findings of the CCI as approved by COMPAT and submitted that there was a strong economic evidence of collusion which is evident from the following aspects: D (a) Identical or near-identical bidding by all 50 empaneled LPG vendors resulting in bid rigging. (b) Results of the tender revealed that these bids were made in such a way that all the bidders were awarded some portion of the tender and no bidder was left empty handed, i.e., Market Sharing E Arrangement. (c) Geographical/Territorial allocation of market, i.e., the bids were placed in such a way that entities located in the northern parts of the country were awrded the tender in the northern States, entities located in the southern parts were awarded the tender in respect of southern States etc. F (d) No plausible economic rationale or explanation was forthcoming for the identical bids, despite obvious difference in cost of production, location, input cost etc.

(e) The overall effect of increase in price of procurement of LPG G Cylinders over previous years. 58. He also submitted that pattern of identical and near identical bids, which was all pervasive through out, could not be brushed aside lightly as that was the clear indicator of price bidding as a result of agreement between the parties. The analysis of the bids also shows H 530 SUPREME COURT REPORTS [2018] 12 S.C.R. A that it had already been decided amongst the LPG Cylinder manufacturers as to who the L1 and L2 bidders were going to be prior to submission of bids. For instance, in the State of Punjab, the L1 bidder (Shri Ram Cylinders) bid Rs. 1080 whereas the four L2 bidders placed identical bids at Rs, 1080.50, i.e., a difference of only 50 paise from the L-1 bid. Similarly, in Rajasthan, the L-1 bidder (M/s. Rajasthan Cylinders) B quoted Rs. 1130, whereas nine L2 bidders quoted identically by just 50 paise more, at Rs. 1130.50. This pattern is repeated across a number of States. 59. Not only this, in order to achieve the pre-decided outcome, some of the bidders hastily made corrections to their bid documents. C One such case is that of M/s. Jesmajo Industrial Fabrications (appellant in C.A. No. 4868 of 2014). In the bid documents, the bid of Rs. 1103 was cut-corrected to make it Rs. 1103.60 even though the calculation of VAT was done only on the figure of |Rs. 1103. The Director General has commented on this aspect as follows:

D “6.13…...That bids were submitted after mutual discussion is apparent from the tender documents of Jesmajo Industrial Fabricators P. Ltd. (Exhibit 4P). There are cuttings in the tender documents and financial bids of the company. Since, there were discussions among all oil companies, the company might have E decided to make alterations in the financial bids. However, even in the financial bids of the company, it is noted that despite alterations, errors have remained. It seems that the company had originally quoted Rs. 1103/- as the rate. Subsequently it changed the rates to Rs. 1103.60. However, VAT rate in the bid had been calculated on Rs. 1103/- only instead of Rs. 1103.60. Thus, while F other component of rates has been changed/altered, the calculation of VAT has been done on Rs. 1103 (originally quoted) instead of Rs. 1103.60 (altered quote). This appears to have been done to match rates with other bidders who have quoted the similar rates in the State of Karnataka and to let Sanghvi group be L-1 in that G State.” 60. Mr. Khurshid also refuted the submisison of the appellants that there was no competition and, therefore, Section 3 was not applicable. According to him, if the matters are examined on such basis most of the culprits will get away. The purpose of the Act was not only to eliminate cartelisation but also to promote competition. His submission was that H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 531 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.]

once the findings of the CCI and COMPAT are accepted that there was A an agreement, such an agreement was obviously for the purpose of curbing the competition. 61. Answering the argument of ‘price parallelism’ which according to the appellants resulted in identical and near identical bids, Mr. Khurshid argued that legal submission in this respect was settled by this Court in B Excel Crop Care case wherein such an argument was rejected in the following words: “48…It was argued that since dominant position is enjoyed by the buyer, it leads to parallel pricing and this conscious parallelism takes place leading to quoting the same price by the suppliers. C The explanation, thus, given for quoting identical price was the aforesaid economic forces and not because of any agreement or arrangement between the parties. It was submitted that merely because same price was quoted by the appellants in respect of the 2009 FCI tender, one could not jump to the conclusion that D there was some “agreement” as well between these parties, in the absence of any other evidence corroborating the said factum of quoting identical price. In respect of this submission, Mr Venugopal had also referred to a few judgments.

49. The aforesaid argument is highly misconceived. A neat and E pellucid reply of Mr. Kaul, which commands acceptance, is that argument of parallelism is not applicable in bid cases and it fits in the realm of market economy. It is for this reason that entire history of quoting identical price before coming into operation of Section 3 and which continued much after Section 3 of the Act was enforced, has been highlighted...” F 62. He also referred to the following findings of COMPAT with the submission that finding of facts need to be accepted:-“36. We are thoroughly convinced by this analysis that all this could not have been possible unless there were internal agreements between the concerns. What shocks us is that the quotations of G the price did match to the last decimal and the quotations in some cases were in odd figures like Rs.1127 in the State of Tamil Nadu. The record is replete with such odd figures. It was strange that in some of the oral statements of the representatives of these parties, who were examined by the DG, some of them could not even H 532 SUPREME COURT REPORTS [2018] 12 S.C.R.

A justify these identical prices and tried to say that it was a mere coincidence. We cannot accept the argument of coincidence as was rightly rejected by CCI. There can be no explanation for this kind of identical or near identical pricing. The CCI has rightly considered that the manufacturing cost of per cylinder varies in a wide spectrum ranging from Rs.870 to Rs.1095.89. If this was B the case, the prices had to be different, if they had been offered in a competitive spirit. Either before the CCI or before us no material was produced, which would be able to rebut the presumption arising from the identity of rates. The CCI, therefore, rightly concluded that this identity of prices was sinister and anti-competitive in C nature.” 63. Another feature which Mr. Salman Khurshid pointed out was that the analysis of bids revealed that there was a market sharing and territorial allocation of bids. This, according to him, could be discerned from the following evidence emerging from record:

D “Firstly, the economic evidence indicates that there was an understanding or arrangement or agreement by which each and every bidder would get some part of the allocation under the tender. This is clear from the fact that each and every one of the 50 bidders who submitted price bids received some part of the E allocation in one or more states, and no one went empty handed. In other words, the purpose of quoting identical bids in many instances was to achieve the objective of sharing the market, i.e., the IOCL requirement across 25 states was ‘shared’ by each of the 50 bidders, through concerted action and pre-decided understanding. F 64. Mr. Khurshid also highlighed that in spite of there being difference in location of appellant’s units and their input cost, the bids submitted by various tenderers were identical and there canot be any plausible economic rationale for such identical bidding. Therefore, the inference drawn by the CCI as well as COMPAT based on the aforesaid G features and factors was justified and valid in law. He also referred to certain judgements of this Court as well as other jurisdictions, such as, European Commission and the Court of Justice of European Union to which reference would be made at the appropriate stage. 65. Before we deal with the arguments advanced by various H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 533 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.]

counsel who appeared for the appellants and rebuttal thereto by Mr. A Salman Khurshid, learned senior counsel who appeared for CCI as also counsel for IOCL, we would like to reproduce the relevant provisions of the Act in the light of which these appeals are to be decided: 66. Section 2(c) defines “cartel” and reads as under: B “”Cartel” includes an association of producers, sellers, distributors, traders or service providers who, by agreement amongst themselves, limit, control or attempt to control the production, distribution, sale or price of, or, trade in goods or provision of services; C 67. Section 3 of the Act deals with and prohibits those agreements which cause and are likely to cause an appreciable adverse effect on competition within india. It reads as under: “Section 3 : Anti-competitive agreements:- (1) No enterprise or association of enterprises or person or D association of persons shall enter into any agreement in respect of production, supply, distribution, storage, acquisition or control of goods or provision of services, which causes or is likely to cause an appreciable adverse effect on competition within India. (2) Any agreement entered into in contravention of the provisions E contained in sub-section (1) shall be void.

(3) Any agreement entered into between enterprises or associations of enterprises or persons or associations of persons or between any person and enterprise or practice carried on, or decision taken by, any association of enterprises or association of persons, F including cartels, engaged in identical or similar trade of goods or provision of services, which— (a) directly or indirectly determines purchase or sale prices; (b) limits or controls production, supply, markets, technical development, investment or provision of services; G (c) shares the market or source of production or provision of services by way of allocation of geographical area of market, or type of goods or services, or number of customers in the market or any other similar way; H 534 SUPREME COURT REPORTS [2018] 12 S.C.R. A (d) directly or indirectly results in bid rigging or collusive bidding, shall be presumed to have an appreciable adverse effect on competition: Provided that nothing contained in this sub-section shall apply to any agreement entered into by way of joint ventures if such agreement increases efficiency in production, supply, distribution, storage, acquisition or control of goods or provision of B services.

Explanation.—For the purposes of this sub-section, “bid rigging” means any agreement, between enterprises or persons referred to in sub-section (3) engaged in identical or similar production or trading of goods or provision of services, which has the effect of C eliminating or reducing competition for bids or adversely affecting or manipulating the process for bidding.” 68. Certain factors are mentioned in Section 19 of the Act which have to be kept in mind while determining whether an agreement has an appreciable adverse effect on competition under Section 3. We reproduce D this Section hereinbelow: “Section 19(3) in the Competition Act, 2002: (3) The Commission shall, while determining whether an agreement has an appreciable adverse effect on competition under section 3, have due regard to all or any of the following factors, namely:— E (a) creation of barriers to new entrants in the market; (b) driving existing competitors out of the market; (c) foreclosure of competition by hindering entry into the market; F (d) accrual of benefits to consumers; (e) improvements in production or distribution of goods or provision of services; (f) promotion of technical, scientific and economic development by means of production or distribution of goods or provision of G services.” 69. In Excel Crop Care Limited, scope of Section 3 of the Act which prohibits three kinds of practices as anti-competitive, was taken note of as follows:

“20. Chapter II of the Act deals with three kinds of practices H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 535 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] which are treated as anti-competitive and prohibited. These are: A (a) where agreements are entered into by certain persons with a view to cause an appreciable adverse effect on competition; (b) where any enterprise or group of enterprises, which enjoys dominant position, abuses the said dominant position; and B (c) regulating the combination of enterprises by means of mergers or amalgamations to ensure that such mergers or amalgamations do not become anti-competitive or abuse the dominant position which they can attain.” 70. In that case also the Court was concerned with the 1st category, C namely, those cases where certain persons enter into agreements with a view to cause an appreciable adverse effect of competition. Purpose behind curbing such anti-competitive practices was mentioned in detail. It is not necessary to re-state the same in that expansive manner, however, we would still like to quote certain portions to capture the essence and Dpurpose of the Act:-

“21. In the instant case, we are concerned with the first type of practices, namely, anti-competitive agreements. The Act, which prohibits anti-competitive agreements, has a laudable purpose behind it. It is to ensure that there is a healthy competition in the market, as it brings about various benefits for the public at large E as well as economy of the nation. In fact, the ultimate goal of competition policy (or for that matter, even the consumer policies) is to enhance consumer well-being. These policies are directed at ensuring that markets function effectively. Competition policy towards the supply side of the market aims to ensure that F consumers have adequate and affordable choices. Another purpose in curbing anti-competitive agreements is to ensure “level playing field” for all market players that helps markets to be competitive. It sets “rules of the game” that protect the competition process itself, rather than competitors in the market. In this way, the pursuit of fair and effective competition can contribute to improvements G in economic efficiency, economic growth and development of consumer welfare……….. xxx xxx xxx 23. In fact, there is broad empirical evidence supporting the H 536 SUPREME COURT REPORTS [2018] 12 S.C.R. A proposition that competition is beneficial for the economy. Economists agree that it has an important role to play in improving productivity and, therefore, the growth prospects of an economy…..

24. Productivity is increased through competition by putting B pressure on firms to control costs as the producers strive to lower their production costs so that they can charge competitive prices. It also improves the quality of their goods and services so that they correspond to consumers’ demands. xxx xxx xxx C 26. When we recognise that competition has number of benefits, it clearly follows that cartels or anti-competitive agreements cause harm to consumers by fixing prices, limiting outputs or allocating markets. Effective enforcement against such practices has direct visible effects in terms of reduced prices in the market and this is also supported by various empirical studies. D xxx xxx xxx 27. Keeping in view the aforesaid objectives that need to be achieved, Indian Parliament enacted the Competition Act, 2002. Need to have such a law became all the more important in the E wake of liberalisation and privatisation as it was found that the law prevailing at that time, namely, Monopolies and Restrictive Trade Practices Act, 1969 was not equipped adequately enough to tackle the competition aspects of the Indian economy. The law enforcement agencies, which include CCI and COMPAT, have to ensure that these objectives are fulfilled by curbing anti-competitive F agreements.”

71. The Court also mentioned, in particular, that competition leads to economic efficiency, economic growth and development as well as consumers welfare. The Court also spelled out the manner in which competition contributed to increase economic growth and increased G productivity. 72. It follows from the above that whereas on the one hand the economic policy of the nation has ushered in the era of liberalisation and globalisation thereby giving freeplay to the private sector in the manner of conducting business, at the same time, in public interest and in the H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 537 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] interest of consumers, a regime of regulators has also been brought to A ensure certain checks and balances. Since competition among the enterprises or businessmen is treated as service for a public purpose and, therefore, there is a need to curb anti-competitive practices, the CCI is given the task (as a regulator) to ensure that no such anti-competitive practices are undertaken. In fact, Section 18 of the Act B

casts a specific and positive obligation on CCI to ‘eliminate’ anti-competitive practices and promote competition, interest of the consmuer and free trade. This objective was also emphasised by this Court in Competition Commission of India vs. Steel Authority of India Limited and Another12 which can be found in the following observations: C “6. As far as the objectives of competition laws are concerned, they vary from country to country and even within a country they seem to change and evolve over the time. However, it will be useful to refer to some of the common objectives of competition law. The main objective of competition law is to promote economic efficiency using competition as one of the means of assisting the D creation of market responsive to consumer preferences. The advantages of perfect competition are threefold: allocative efficiency, which ensures the effective allocation of resources, productive efficiency, which ensures that costs of production are kept at a minimum and dynamic efficiency, which promotes E innovative practices. These factors by and large have been accepted all over the world as the guiding principles for effective implementation of competition law.”

73. As mentioned above, one of the anti-competitive practices is cartelisation, the essential postulate whereof is agreement between Fenterprises or association of enterprises or persons or associations of persons in respect of production, supply, distribution, storage, acquisition or control of goods or provisions of service, which causes or is likely to cause an appreciable adverse effect on competition within India. Such an agreement is treated as void. The types of agreement which may fall foul of Section 3 are mentioned in sub-section (3) thereof. These include G sharing the market by way of allocation of geographical areas of market [clause (c)] and the agreements which result in bid-rigging or collusive bidding whether directly or indirectly [clause (d)]. There is a presumption that four types of agreements mentioned in sub-section (3) will have an appreciable adverse effect on competition. H 538 SUPREME COURT REPORTS [2018] 12 S.C.R. A 74. We may also state at this stage that Section 19(3) of the Act mentions the factors which are to be examined by the CCI while determining whether an agreement has an appreciable adverse effect on competition under Section 3. However, this inquiry would be needed in those cases which are not covered by clauses (a) to (d) of sub-section (3) of Section 3. Reason is simple. As already pointed out above, the B

agreements of nature mentioned in sub-section (3) are presumed to have an appreciable effect and, therefore, no further exercise is needed by the CCI once a finding is arrived at that a particular agreement fell in any of the aforesaid four categories. We may hasten to add, however, that agreements mentioned in Section 3(3) raise a presumption that such C agreements shall have an appreciable adverse effect on competition. It follows, as a fortiorari, that the presumption is rebuttable as these agreements are not treated as conclusive proof of the fact that it would result in appreciable adverse effect on competition. What follows is that once the CCI finds that case is covered by one or more of the clauses mentioned in sub-section (3) of Section 3, it need not undertake D

any further enquiry and burden would shift upon such enterprises or persons etc. to rebut the said presumption by leading adequate evidence. In case such an evidence is led, which dispels the presumption, then the CCI shall take into consideration the factors mentioned in Section 19 of the Act and to see as to whether all or any of these factors are established. E If the evidence collected by the CCI leads to one or more or all factors mentioned in Section 19(3), it would again be treated as an agreement which may cause or is likely to cause an appreciable adverse effect of competition, thereby compelling the CCI to take further remedial action in this behalf as provided under the Act. That, according to us, is the broad scheme when Sections 3 and 19 are to be read in conjuction. F 75. In these appeals, the Court is concerned with the alleged agreement entered into between the appellants falling in clause (d) of sub-section (3) of Section 3, which talks of bid rigging or collusive bidding. Therefore, it would be necessary to understand the meaning of the expression ‘bid rigging’ and ‘collusive bidding’. Explanation to Section G 3, which is reproduced, assigns meaning to ‘bid rigging’ and states : “S. 3: Explanation.—For the purposes of this sub-section, “bid rigging” means any agreement, between enterprises or persons referred to in sub-section (3) engaged in identical or similar production or H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 539

UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] trading of goods or provision of services, which has the effect of A eliminating or reducing competition for bids or adversely affecting or manipulating the process for bidding.” 76. The necessary ingredients of bid rigging, thus, are: (a) agreement between the parties; (b) these parties are engaged in identical or similar production or trading of goods or provisions of services; and B (c) the agreement has the effect of eliminating or reducing competition of bids or adversely affect or manipulating the process for bidding. 77. Though the expression ‘collusive bidding’ is not defined in the Act, it appears that both ‘bid rigging’ and ‘collusive bidding’ are overlapping concepts. This position stands accepted in Excel Crop C Care Limited case which should be found from the following discussion therefrom:

“38. Mr Neeraj Kishan Kaul, learned Additional Solicitor General, refuted the aforesaid submission with vehemence by urging that bid rigging and collusive bidding are not mutually exclusive and D these are overlapping concepts. Illustratively, he referred to the findings of CCI, as approved by COMPAT, in the instant case itself to the effect that the appellants herein had “manipulated the process of bidding” on the ground that bids were submitted on 8-5-2009 collusively, which was only the beginning of the anti-E competitive agreement between the parties and this continued through the opening of the price bids on 1-6-2009 and thereafter negotiations on 17-6-2009 when all the parties reduced their bids by same figure of Rs 2 to bring their bid down to Rs 386 per kg from Rs 388 per kg. From this example, he submitted that on 8-5-2009 there was a collusive bidding but with concerted negotiations F on 17-6-2009, in the continued process, it was rigging of the bid that was practiced by the appellants. We are inclined to agree with this pellucid submission of the learned Additional Solicitor General. 39. Richard Whish and David Bailey [Competition Law, 7th Edn., G p. 536.] , in their book, have given illustrations of various forms of collusive bidding/bid rigging, which include: (a) Level tendering/bidding (i.e. bidding at same price — as in the present case). H 540 SUPREME COURT REPORTS [2018] 12 S.C.R. A (b) Cover bidding/courtesy bidding. (c) Bid rotation.

(d) Bid allocation. 40. Even internationally, “collusive bidding” is not understood as B being different from “bid rigging”. These two expressions have been used interchangeably in the following international commentaries/glossaries and websites of competition authorities: (a) UNCTAD Competition Glossary dated 22-6-2016 “Bid rigging or collusive tendering is a manner in which conspiring C competitors may effectively raise prices where business contracts are awarded by means of soliciting competitive bids. Essentially, it relates to a situation where competitors agree in advance who will win the bid and at what price, undermining the very purpose of inviting tenders which is to procure goods or services on the most favourable prices and conditions.” D (b) OECD Glossary of Industrial Organisation Economics and Competition Law “Bid rigging is a particular form of collusive price-fixing behaviour by which firms coordinate their bids on procurement or project E contracts. There are two common forms of bid rigging. In the first, firms agree to submit common bids, thus eliminating price competition. In the second, firms agree on which firm will be the lowest bidder and rotate in such a way that each firm wins an agreed upon number or value of contracts.

F Since most (but not all) contracts open to bidding involve Governments, it is they who are most often the target of bid rigging. Bid rigging is one of the most widely prosecuted forms of collusion.” Collusive bidding (tendering) — See “bid rigging”. (This shows collusive bidding and bid rigging are treated as one G and the same.) (c) OECD Guidelines for fighting bid rigging “Bid rigging (or collusive tendering) occurs when businesses, that would otherwise be expected to compete, secretly conspire to H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 541 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] raise prices or lower the quality of goods or services for purchasers A who wish to acquire products or services through a bidding process.” (d) United States Office of the Inspector General, Investigations (Fraud Indicators Handbook) B “Collusive bidding, price fixing or bid rigging, are commonly used interchangeable terms which describe many forms of an illegal anti-competitive activity. The common thread throughout all these activities is that they involve any agreements or informal arrangements among independent competitors, which limit competition. Agreements among competitors which violate the C law include but are not limited to: (1) Agreements to adhere to published price lists. (2) Agreements to raise prices by a specified increment.(3) Agreements to establish, adhere to, or eliminate discounts. D (4) Agreements not to advertise prices.

(5) Agreements to maintain specified price differentials based on quantity, type or size of product.” (e) Australian Competition and Consumer Commission E “Bid rigging, also referred to as collusive tendering, occurs when two or more competitors agree they will not compete genuinely with each other for tenders, allowing one of the cartel members to ‘win’ the tender. Participants in a bid rigging cartel may take turns to be the ‘winner’ by agreeing about the way they submit tenders, including some competitors agreeing not to tender.” F 41. As the Liegeman of the law, it is our task, nay a duty, to give proper meaning and effect to the aforesaid “Explanation”. It can easily be discussed that the legislature had in mind that the two expressions are interchangeably used. It is also necessary to keep in mind the purport behind Section 3 and the objective it seeks to G achieve: 41.1. Sub-section (1) of Section 3 is couched in the negative terms which mandates that no enterprise or association of enterprises or person or association of persons shall enter into any agreement, H 542 SUPREME COURT REPORTS [2018] 12 S.C.R.

A when such agreement is in respect of production, supply, distribution, storage, acquisition or control of goods or provision of services and it causes or is likely to cause an appreciable adverse effect on competition within India. It can be discerned that first part relates to the parties which are prohibited from entering into such an agreement and embraces within it persons as well as B enterprises thereby signifying its very wide coverage. This becomes manifest from the reading of the definition of “enterprise” in Section 2(h) and that of “person” in Section 2(l) of the Act. The second part relates to the subject-matter of the agreement. Again it is very wide in its ambit and scope as it covers C production, supply, distribution, storage, acquisition or control of goods or provision of services. The third part pertains to the effect of such an agreement, namely, “appreciable adverse effect on competition”, and if this is the effect, purpose behind this provision is not to allow that. Obvious purpose is to thwart any such agreements which are anti-competitive in nature and this salubrious D provision aims at ensuring healthy competition. Sub-section (2) of Section 3 specifically makes such agreements as void.

41.2. Sub-section (3) mentions certain kinds of agreements which would be treated as ipso facto causing appreciable adverse effect on competition. It is in this backdrop and context that “Explanation” E beneath sub-section (3), which uses the expression “bid rigging”, has to be understood and given an appropriate meaning. It could never be the intention of the legislature to exclude “collusive bidding” by construing the expression “bid rigging” narrowly. No doubt, clause (d) of sub-section (3) of Section 3 uses both the F expressions “bid rigging” and “collusive bidding”, but the Explanation thereto refers to “bid rigging” only. However, it cannot be said that the intention was to exclude “collusive bidding”. Even if the Explanation does contain the expression “collusive bidding” specifically, while interpreting clause (d), it can be inferred that “collusive bidding” relates to the process of bidding as well. G Keeping in mind the principle of purposive interpretation, we are inclined to give this meaning to “collusive bidding”. It is more so when the expressions “bid rigging” and “collusive bidding” would be overlapping, under certain circumstances which was conceded by the learned counsel for the appellants as well. H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 543 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.]

42. We are, therefore, of the opinion that the two expressions are A to be interpreted using the principle of noscitur a sociis i.e. when two or more words which are susceptible to analogous meanings are coupled together, the words can take colour from each other. (See Leelabai Gajanan Pansare v. Oriental Insurance Co. Ltd.[Leelabai Gajanan Pansare v. Oriental Insurance Co. Ltd., B (2008) 9 SCC 720] , Thakorlal D. Vadgama v. State of Gujarat [Thakorlal D. Vadgama v. State of Gujarat, (1973) 2 SCC 413 : 1973 SCC (Cri) 835] and M.K. Ranganathan v. State of Madras[M.K. Ranganathan v. State of Madras, (1955) 2 SCR 374 : AIR 1955 SC 604] .)” C

78. The first proposition of Ms. Divan, viz. there is no competition, has two facets. First, the legal one which concerns the jurisdiction of the CCI to deal with such matters and the other is factual, which is to be examined on the basis of facts in these cases. Insofar as the first component is concerned, having regard to the aforesaid scheme of the Act, we are not convinced with the argument of Ms. Madhavi Divan D that there is no possibility of a competition in these cases and, therefore, CCI had no jurisdiction to carry out any such investigation. The scope and ambit of the provisions of Section 3 have been considered in detail in Excel Crop Care Limited case. This Section prohibits anti-competitive agreements and brings about the prime objective of the Competition E Act. These aspects were noted in Excel Crop Care Limited, relevant portions whereof are already extracted above. We may also quote the following portion from the judgment of this Court in Steel Authority of India Limited wherein objective behind the Act was highlighted in the following manner: F

“125. We have already noticed that the principal objects of the Act, in terms of its Preamble and the Statement of Objects and Reasons, are to eliminate practices having adverse effect on the competition, to promote and sustain competition in the market, to protect the interest of the consumers and ensure freedom of trade carried on by the participants in the market, in view of the economic G developments in the country. In other words, the Act requires not only protection of free trade but also protection of consumer interest. The delay in disposal of cases, as well as undue continuation of interim restraint orders, can adversely and H 544 SUPREME COURT REPORTS [2018] 12 S.C.R. A prejudicially affect the free economy of the country. Efforts to liberalise the Indian economy to bring it on a par with the best of the economies in this era of globalisation would be jeopardised if time-bound schedule and, in any case, expeditious disposal by the Commission is not adhered to. The scheme of various provisions of the Act which we have already referred to including Sections B 26, 29, 30, 31, 53-B(5) and 53-T and Regulations 12, 15, 16, 22, 32, 48 and 31 clearly show the legislative intent to ensure time-bound disposal of such matters.”

79. We would like to reemphasise that the purpose of the Act is not only to illuminate practices having adverse effect on the competition C but also to promote and sustain competition in the market. Enforcement provides remedies to avoid situation that will lead to decrease competition in the market. Therefore, effective enforcement is important not only to sanction anti-competitive conduct but also to deter future competitive practices. In the present case itself, there are sixty suppliers of the product D for which there are three buyers. After all, each supplier would like to be L-1 or L-2 so that it is able to get order for larger quantities than the other. In this sense, there would be a competition among them. Further, it would also be in the interest of the buyers like IOCL etc. that the elements of healthy competition persists in the market. In any case, it is the duty of the CCI to ensure that the conditions which have tendency to E kill the competition are to be curbed. It is also the function of the CCI to ensure that there is a competition so that benefits of such competition are reaped by the consumers. However, insofar as certain factual aspects highlighted by the appellants are concerned, they would be dealt with while examining the third proposition, as we deem it more appropriate to F discuss these two aspects together.

80. Second proposition of Ms. Divan was that there was no collusive bidding in the present case. The CCI and COMPAT have rejected this argument in view of the fact that there is an active trade association of the suppliers; a meeting took place couple of days before G the date of bidding; common changes were pointed out by these appellants who submitted bids on their behalf; and bids were of identical amounts despite varying cost, which were repetitive in nature. The respondents may be right in their submission that there may not be a direct evidence on the basis of which cartelisation or such agreement between the parties can be proved as these agreements are normally entered into in closed H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 545 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] doors. The standard of proof which is required is one of probability, A which is a principle accepted in Technip SA vs. SMS Holding (P) Ltd. & Ors.13 wherein the Court stated and discussed this aspect in the following manner: “54. The standard of proof required to establish such concert is one of probability and may be established B “if having regard to their relation etc., their conduct, and their common interest, that it may be inferred that they must be acting together: evidence of actual concerted acting is normally difficult to obtain, and is not insisted upon” [CIT v. East Coast Commercial Co. Ltd., (1967) 1 SCR 821 : AIR 1967 SC 768] C . (SCR p. 829 H)

55. While deciding whether a company was one in which the public were substantially interested within the meaning of Section 23-A of the Income Tax Act, 1922 this Court said: “The test is not whether they have actually acted in concert D but whether the circumstances are such that human experience tells us that it can safely be taken that they must be acting together. It is not necessary to state the kind of evidence that will prove such concerted actings. Each case must necessarily be decided on its own facts.” [CIT v. Jubilee Mills Ltd., (1963) E 48 ITR 9 (SC), p. 20] 56. In Guinness PLC and Distillers Co. PLC [Guinness PLC and Distillers Company PLC (Panel hearing on 25-8-1987 and 2-9-1987 at p. 10052 — Reasons for decisions of the Panel.)] the question before the Takeover Panel was whether Guinness had F acted in concert with Pipetec when Pipetec purchased shares in Distillers Company PLC. Various factors were taken into consideration to conclude that Guinness had acted in concert with Pipetec to get control over Distillers Company. The Panel said: “The nature of acting in concert requires that the definition be G drawn in deliberately wide terms. It covers an understanding as well as an agreement, and an informal as well as a formal arrangement, which leads to cooperation to purchase shares to acquire control of a company. This is necessary, as such 13 (2005) 5 SCC 465 H 546 SUPREME COURT REPORTS [2018] 12 S.C.R.

A arrangements are often informal, and the understanding may arise from a hint. The understanding may be tacit, and the definition covers situations where the parties act on the basis of a ‘nod or a wink’…. Unless persons declare this agreement or understanding, there is rarely direct evidence of action in concert, and the Panel must draw on its experience and B common sense to determine whether those involved in any dealings have some form of understanding and are acting in cooperation with each other.” [Guinness PLC and Distillers Company PLC (Panel hearing on 25-8-1987 and 2-9-1987 at p. 10052 — Reasons for decisions of the Panel.)] C 81. We would also like to reproduce the following discussion in Commissioner of Income Tax, Bombay City I, Bombay vs. Jubilee Mills Ltd., Bombay14:

“19. At the hearing a point was raised that it has to be proved as a fact that the persons constituting the group which owns shares D carrying more than seventy-five per cent of the voting power, were acting in unison. The test is not whether they have actually acted in concert but whether the circumstances are such that human experience tells us that it can safely be taken that they must be acting together. It is not necessary to state the kind of E evidence that will prove such concerted actings. Each case must necessarily be decided on its own facts. The exclusion of “public” in the manner indicated generally from more than 75% of the shares and the concentration of such a holding in a single person or a group acting in concert is what attracts Section 23(A).” 82. It is also significant to state that respondents had drawn F attention of this Court to OECD Policy Roundtables Prosecuting Cartels without Direct Evidence 2006 which discussed the nature of evidence that is required for proving cartel agreement, relevant portion thereof contained in para 2 of the said Policy is reproduced below: “Available evidence for proving cartel agreements G 2.1 Categories of evidence Evidence used to prove a cartel agreement can be classified into two types: direct and circumstantial. Circumstantial evidence, in 14 (1963) 48 ITR 9 H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 547 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.]

turn, consists of “communication” evidence and economic A evidence, which include firm conduct, market structure, and evidence of facilitating practices. Common types of direct evidence include: - A document or documents (including email messages) B essentially embodying the agreement, or parts of it, and identifying the parties to it. - Oral or written statements by co-operative cartel participants describing the operation of the cartel and their participation in it. There are different types of circumstantial evidence. One is C evidence that cartel operators met or otherwise communicated but does not describe the substance of their communications. It might be called communication evidence for purposes of this discussion. It includes: - Records of telephone conversations between competitors D (but not their substance), or of travel to a common destination or of participation in a meeting, for example during a trade conference. - other evidence that the parties communicated about the subject e.g., minutes or notes of a meeting showing that prices, E demand or capacity utilisation were discussed; internal documents evidencing knowledge or understanding of a competitors pricing strategy, such as an awareness of a future price increase by a rival. A broader category of circumstantial evidence is often called F

economic evidence. Economic evidence identifies primarily firm conduct that suggests that an agreement was reached, but also conduct of the industry as a whole, elements of market structure which suggest that secret price fixing was feasible, and certain practices that can be used to sustain a cartel agreement. G Conduct evidence is the single most important type of economic evidence. As noted earlier, observation of certain, suspicious conduct frequently triggers an investigation of a possible cartel. And as the section in this paper on economics highlights 11 careful analysis of the conduct of parties is important to identify behaviours H 548 SUPREME COURT REPORTS [2018] 12 S.C.R. A that can be characterised as contrary to the parties’ unilateral self-interest and which therefore supports the inference of an agreement. Conduct evidence includes, first and foremost: - Parallel pricing – changes in prices by rivals that are identical, or nearly so, and simultaneous, or nearly so. It includes other B forms of parallel conduct, such as capacity reductions, adoption of standardised terms of sale, and suspicious bidding patterns, e.g., a predictable rotation of winning bidders. Industry performance could also be described as conduct evidence. It includes: C - abnormally high profits; - stable market shares - A history of competition law violations.

Evidence related to market structure can be used primarily to D make the finding of a cartel agreement more plausible, even though market structure factors do not prove the existence of such an agreement. Relevant economic evidence relating to market structure includes: - high concentration; E - low concentration on the opposite side of the market; - high barriers to entry; - high degree of vertical integration; - Standardised or homogeneous product. F The evidentiary value of structural evidence can be limited, however. There can be highly concentrated industries selling homogeneous products in which all parties compete. Conversely, the absence of such evidence cannot be used to show that a cartel did not exist. Cartels are known to have existed in industries with G numerous competitors and differentiated products. A specific kind of economic conduct evidence is facilitating practices – practices that can make it easier for competitors to reach or sustain an agreement. It is important to note that conduct described as facilitating practices is not necessarily unlawful. But H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 549 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.]

where a competition authority has found other circumstantial A evidence pointing to the existence of a cartel agreement, the existence of facilitating practices can be an important complement. They can explain what kind of arrangements the parties set up to facilitate the formation of a cartel agreement, monitoring, detection of defection, and/or punishment, thus supporting the ‘collusion B story’ put together by the competition law enforcer. Facilitating practices include: - information exchanges; - price signalling; C - freight equalisation; - price protection and most favoured nation policies; - Unnecessarily restrictive product standards.” 83. Thus, even in the absence of proof of concluded formal agreement, when there are indicators that there was practical cooperation D between the parties which knowingly substitute the risk of competition, that would amount to anti-competitive practices. This has been discussed in Coordination Committee of Artistes and Technicians of West Bengal Film and Television & Ors. (see paras 44 and 45). Then, there are guidelines on the applicability of Article 101 of the Treaty on E the functioning of the E.U. to horizontal cooperation agreements which records as under: “60. Information exchange can only be addressed under Article 101 if it establishes or is part of an agreement, a concerted practice or a decision by an association of undertakings. The existence of F

an agreement, a concerted practice or decision by an association of undertakings does not prejudge whether the agreement, concerted practice or decision by an association of undertakings gives rise to a restriction of competition within the meaning of Article 101(1). In line with the case-law of the Court of Justice of the European Union, the concept of a concerted practice refers G to a form of coordination between undertakings by which, without it having reached the stage where an agreement properly so-called has been concluded, practical cooperation between them is knowingly substituted for the risks of competition. The criteria of coordination and cooperation necessary for determining the H 550 SUPREME COURT REPORTS [2018] 12 S.C.R. A existence of a concerted practice, far from requiring an actual plan to have been worked out, are to be understood in the light of the concept inherent in the provisions of the Treaty on competition, according to which each company must determine independently the policy which it intends to adopt on the internal market and the conditions which it intends to offer to its customers. B

61. This does not deprive companies of the right to adapt themselves intelligently to the existing or anticipated conduct of their competitors. It does, however, preclude any direct or indirect contact between competitors, the object or effect of which is to create conditions of competition which do not correspond to the C normal competitive conditions of the market in question, regard being had to the nature of the products or services offered, the size and number of the undertakings, and the volume of the said market. This precludes any direct or indirect contact between competitors, the object or effect of which is to influence conduct D on the market of an actual or potential competitor, or to disclose to such competitor the course of conduct which they themselves have decided to adopt or contemplate adopting on the market, thereby facilitating a collusive outcome on the market. Hence, information exchange can constitute a concerted practice if it reduces strategic uncertainty in the market thereby facilitating E collusion, that is to say, if the data exchanged is strategic. Consequently, sharing of strategic data between competitors amounts to concentration, because it reduces the independence of competitors’ conduct on the market and diminishes their incentives to compete.

F 62. A situation where only one undertaking discloses strategic information to its competitor(s) who accept(s) it can also constitute a concerted practice. Such disclosure could occur, for example, through contacts via mail, emails, phone calls, meetings etc. It is then irrelevant whether only one undertaking unilaterally informs G its competitors of its intended market behaviour, or whether all participating undertakings inform each other of the respective deliberations and intentions. When one undertaking alone reveals to its competitors strategic information concerning its future commercial policy, that reduces strategic uncertainty as to the H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 551 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] future operation of the market for all the competitors involved A and increases the risk of limiting competition and of collusive behaviour. For example, mere attendance at a meeting where a company disclose its pricing plans to its competitors is likely to be caught by Article 101, even in the absence of an explicit agreement to raise prices. When a company receives strategic data from a B competitor (be it in a meeting, by mail or electronically), it will be presumed to have accepted the information and adapted its market conduct accordingly unless it responds with a clear statement that it does not wish to receive such data.” (emphasis added) C

84. According to us, the real question in the present case is as to whether there was a possibility of such an agreement having regard to market conditions even when we proceed on the basis that meeting did take place. Possibility of such an agreement has been inferred by the CCI on the grounds that identical bidding takes place thereafter and Dvarious suppliers gave such a bid despite varying cost and also that they have appoined common changes etc. as pointed out above. 85. The first and foremost issue which needs to be considered is that whether there was a situation of monopsony or oligopsony.

86. From the aforesaid discussion, it is clear that as far as CCI is E concerned, it has come to the conclusion that there was a cartelisation among the appellants herein and a concerted decision was taken to rig the bids which were submitted persuant to the tenders issued by IOCL. On the other hand, the appellants argue that there was no such agreement and even if the bids of many bidders were identical in nature, the bids were driven by market conditions. Their plea is that there was a situation F of oligopsony and the modus which was adopted by IOCL in floating the tenders and awarding the contracts would show that the determination of price was entirely within the control of the IOCL. As per them, the way price was determined for supply of these cylinders, it had become an open secret known to everybody. Therefore, there was no question G of any competition and no possibility of adversely affecting that competition by entering into any contract. 87. The factors which have influenced the authorities below in coming to the conclusion that the appellants had colluded and formed a H 552 SUPREME COURT REPORTS [2018] 12 S.C.R. A cartel which led to bid rigging have already been noted above. To recaptulate, the authorities below have been influenced by the following factors: 1. Market conditions 2. Small number of suppliers B 3. Few new entrants 4. Active trade association 5. Repetitive bidding 6. Identical products C 7. Few or no substitutes

8. No significant technological changes 9. Meeting of bidders in Mumbai and its agenda. D 10. Appointing common agents 11. Identical bids despite varying cost. After deliberating on the aforesaid aspects, the CCI has concluded that there is an active trade association in which many of the appellants are members. That product in question, namely, gas cylinder is of a E particular specification which is needed by IOCL in large numbers every year and there are very few manufacturers and suppliers of this product to IOCL and two other buyers. For this identical product which is to be supplied by all the suppliers, there is no substitute and no significant technology change. Further, there is an active trade association in which most of the appellants are the members. Their interest is to ensure that F no new entrants are able to join. Further, the trade association also ensures that all the members are able to get some order. It is for this reason the bids submitted in various standards which are floated by IOCL at different places are almost identical despite varying cost. The authorities below attributed this identical bidding to the concerted action G of the appellants. This has been inferred from the fact that 2-3 days before the submission of bids, meeting of the association took place which most of the appellants attended. Not only this, common agents, six in number, were appointed who submitted the bids on behalf of these appellants. H

RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 553 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] 88. We may say at the outset that if these factors are taken into A consideration by themselves, they may lead to the inference that there was bid rigging. We may, particularly, emphasise the fact that there is an active trade association of the appellants and a meeting of the bidders was held in Mumbai just before the submission of the tenders. Another very important fact is that there were identical bids despite varying cost. B Further, products are identical and there are small number of suppliers with few new entrants. These have become the supporting factors which persuaded the CCI to come to the conclusion that these are suggestive of collusive bidding. 89. However, that is only one side of the coin. The aforesaid Cfactors are to be analysed keeping in mind the ground realities that were prevailing, which are pointed out by the appellants. These attendant circumstances are argued in detail by the counsel for the appellants which have already been taken note of. We may recapitulate the same in brief hereinbelow: D (i) In the present case there are only three buyers. Among them, IOCL is the biggest buyer with 48% market share. It is also a matter of record that all these appellants are manufacturers of

14.2 kg gas cylinders to the three buyers who are available in the market, namely, IOCL, HPCL and BPCL. If these three buyers do not purchase from any of the appellants, that particular appellant E would not be in a position to sell those cylinder to any other entity as there are no other buyers. (ii) There are only three buyers, it may not attract many to enter the field and manufacture these cylinders. It is because of limited number of buyers and for some reason if they do not purchase, F the manufacturer would be nowhere. That may deter the persons to enter the field. (iii) The manner in which the tenders are floated by IOCL and the rates at which these are awarded, are an indicator that it is the IOCL which calls the shots insofar as price control is concerned. G It has come in evidence that the IOCL undertakes the exercise of having its internal estimates about the cost of these cylinders. Their own expert arrived at a figue of Rs. 1106.61 paisa per cylinder. All the tenders which have been accepted are for a price lesser than the aforesaid estimate of IOCL itself. That apart, H 554 SUPREME COURT REPORTS [2018] 12 S.C.R.

A the modus adopted by the IOCL is that that final price is negotiated by it and the contract is not awarded at the rate quoted by bidder who turns out to be L-1. Negotiations are held with such a bidder who is L-1 which generaly leads to further reduction of price than the one quoted by L-1. Thereafter, the other bidders who may be L-2 or L-3 etc. are awarded the contract at the rate at which it is B awarded to L-1. Thus, ultimately, all the bidders supply the goods at the same rate which is fixed by the IOCL after negotiating with L-1 bidder. The only difference is that bidder who is L-1 would be able to receive the order for larger quantity than L-2 and L-2 may get an order of more quantity than L-3. C (iv) It has also come on record that there are very few suppliers. For the tender in question, there were 50 parties already in the fray and 12 new entrants were admitted. Number of 12, in such a scenario, cannot be treated as less. Therefore, the conclusion of CCI that the appellants ensured that there should not be entry D of new entrant may not be correct. (v) Since there are not many manufacturers and supplies are needed by the three buyers on regular basis, IOCL ensures that all those manufacturers whose bids are technically viable, are given some order for the supply of specific cylinder. For this E purpose, it has framed its broad policy as well. This also shows that control remains with IOCL.

Thus, the appellants appear to be correct when they say that all the participants in the bidding process were awarded contracts in some State or the other which was aimed at ensuring a bigger pool of manufacturers so that the supply of this essential product F is always maintained for the benefit of the general public. Had IOCL left some manufacturers empty handed, in all likelihood, they would have shut their shops. However, IOCL wanted all manufacturers to be in the fray in its own interest. Therefore, it was necessary to keep all parties afloat and this explains why all G 50 parties obtained order along with 12 new entrants. (vi) There is another very relevant factor pointed out by the appellants, viz., the governmental control which is regulated by law. As pointed out above, it is not only the three oil companies which can supply LPG to domestic consumers in 14.2 kg LPG H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 555 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] cylinders as mandated in the LPG (Regulation and Distribution) A Order, 2000 which is issued under the provisions of Essential Commodities Act, 1955, even the price at which the LPG cylinder is to be supplied to the consumer is controlled by the Government. Following features of the aforesaid LPG Order, 2000, are significant: B • The LPG supplied in 14.2 kg gas cylinders is an essential commodity.

• The distribution of LPG in 14.2 kgs cylinders takes place as part of a public distribution system defined under clause 2(1) of C the Order as “the system of distribution, marketing or selling of liquified petroleum gas by a Government Oil Company at the Government controlled or declared price through a distribution system approved by the Central or State Government”. • The price to the consumer is controlled by the Government. D • The supply of LPG to domestic consumers shall be made only in 14.2 kg gas cylinders. • According to clauses 4 and 5 read with Schedule III of the LPG Order, parallel marketeers who supply and distribute LPG E cylinders, may do so only for cylinders with size and specifications other than those specified in Schedule II. 90. The manner in which tendering process takes place would show that in such a competitive scenario, the bid which the different bidder would be submitting becomes obvious. It has come on record that Fjust a few days before the tender in question, another tender was floated by BPCL and on opening of the said tender the rates of L-1, L-2 etc. came to be known. In a scenario like this, that obviously becomes a guiding factor for the bidders to submit their bids.

91. When we keep in mind the aforesaid fact situation on the Gground, those very factors on the basis of which the CCI has come to the conclusion that there was cartelization, in fact, become valid explanations to the indicators pointed out by the CCI. We have already commented about the market conditions and small number of suppliers. We have also mentioned that 12 new entrants cannot be considered as H 556 SUPREME COURT REPORTS [2018] 12 S.C.R. A entry of very few new suppliers where the existing suppliers were only 50. Identical products along with market conditions for which there would be only three buyers, in fact, would go in favour of the appellants. The factor of repetitive bidding, though appears to be a factor against the appellants, was also possible in the aforesaid scneario. The prevailing conditions in fact rule out the possibility of much price variations and all B

the manufacturers are virtually forced to submit their bid with a price that is quite close to each other. Therefore, it became necessary to sustain themselves in the market. Hence, the factor that these suppliers are from different region having different cost of manufacture would lose its significance. It is a situation where prime condition is to quote C the price at which a particular manufacturer can bag an order even when its manufacturing cost is more than the manufacturing cost of others. The main purpose for such a manufacuring would be to remain in the fray and not to lose out. Therefore, it would be ready to accept lesser margin. This would answer why there were near identical bids despite varying cost. D

92. Insofar as meeting of bidders in Mumbai just before the date of submission of tender is concerned, some aspects pointed out by the appellants are not considered by the CCI or the COMPAT at all. No doubt, the meeting took place a couple of days before the date of tender. No doubt, the absence of agenda coming on record would not make E much difference. However, only 19 appellants had attended that meeting. Many others were not even members or did not attend the meeting. In spite thereof, even they quoted almost same rates as the one who attended the meeting. This would lead us to the inference that reason for quoting similar price was not the meeting but something else. The question is F what would be the other reason and whether the appellants have been able to satisfactorily explain that and rebut the presumption against them? 93. The explanation is market conditions leading to the situation of oligopsony that prevailed because of limited buyers and influence of buyers in the fixation of prices was all prevalent. This seems to be G convincing in the given set of facts. The situation of oligopsony can be both ways. There may be a situation where the sellers are few and they may control the market and by their concerted action indulge into cartelization. It may also be, as in the present case, a situation where buyers are few and that results in the situation of oligopsony with the control of buyers. H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 557

UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] 94. To recapitulate, the two prime factors against the appellants, A which are discussed by the CCI, are that there was a collusive tendering, which is inferred from the parallel behaviour of the appellants, namely, quoting almost the same rates in their bids. The parameters on the basis of which these aspects are to be judged are stated in Excel Crop Care Limited as follows: B “50. It needs to be emphasised that collusive tendering is a practice whereby firms agree amongst themselves to collaborate over their response to invitations to tender. Main purpose for such collusive tendering is the need to concert their bargaining power, though, such a collusive tendering has other benefits apart from the fact C that it can lead to higher prices. Motive may be that fewer contractors actually bother to price any particular deal so that overheads are kept lower. It may also be for the reason that a contractor can make a tender which it knows will not be accepted (because it has been agreed that another firm will tender at a lower price) and yet it indicates that the said contractor is still D interested in doing business, so that it will not be deleted from the tenderee’s list. It may also mean that a contractor can retain the business of its established, favoured customers without worrying that they will be poached by its competitors.

51. Collusive tendering takes many forms. Simplest form is to E agree to quote identical prices with the hope that all will receive their fair share of orders. That is what has happened in the present case. However, since such a conduct becomes suspicious and would easily attract the attention of the competition authorities, more subtle arrangements of different forms are also made F between colluding parties. One system which has been noticed by certain competition authorities in other countries is to notify intended quotes to each other, or more likely to a Central secretariat, which will then cost the order and eliminate those quotes that it considers would result in a loss to some or all members of the cartel. Another system, which has come to light, is to rotate orders. G In such a case, the firm whose turn is to receive an order will ensure that its quote is lower than the quotes of others. 52. We are here concerned with parallel behaviour. We are conscious of the argument put forth by Mr Venugopal that in an H 558 SUPREME COURT REPORTS [2018] 12 S.C.R. A oligopoly situation parallel behaviour may not, by itself, amount to a concerted practice. It would be apposite to take note of the following observations made by European Court of Justice in Dyestuffs:

“By its very nature, then, the concerted practice does not have B all the elements of a contract but may inter alia arise out of coordination which becomes apparent from the behaviour of the participants. Although parallel behaviour may not itself be identified with a concerted practice, it may however amount to strong evidence of such a practice if it leads to conditions of competition which do not respond to the normal C conditions of the market, having regard to the nature of the products, the size and number of the undertakings, and the volume of the said market. Such is the case especially where the parallel behaviour is such as to permit the parties to seek price equilibrium at a different level from that which would D have resulted from competition, and to crystallise the status quo to the detriment of effective freedom of movement of the products in the [internal] market and free choice by consumers of their suppliers.” (emphasis supplied) E At the same time, the Court also added that the existence of a concerted practice could be appraised correctly by keeping in mind the following test: “If the evidence upon which the contested decision is based is considered, not in isolation, but as a whole, account being taken F of the specific features of the products in question.” Having regard to the aforesaid principles in mind, we deal with the arugment on oligopsony raised by the appellants.

95. Monopsony consists of a market with a single buyer. When there are only few buyers the market is described as an oligopsony. G What is emphasised is that in such a situation a manufacturer with no buyers will have to exit from the trade. Therefore, first condition of oligopsony stands fulfilled. The other condition for the existence of oligopsony is whether the buyers have some influence over the price of their inputs. It is also to be seen as to whether the seller has any ability H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 559 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] to raise prices or it stood reduced/eliminated by the aforesaid buyers. A 96. On a hollistic view of the matter, we find that the appellants have been able to discharge the onus by referring to various indicators which go on to show that parallel behaviour was not the result of any concerted practice.

B 97. In Dyestuffs, the European Court held that parallel behaviour does not, by itself, amount to a concerted practice, though it may provide a strong evidence of such a practice. Nevertheless, it is a strong evidence of such a practice. However, before such an inference is drawn it has to be seen that this parallel behaviour has led to conditions of competition which do not correspond to the normal conditions of the market, having C regard to the nature of the products, size and volume of the undertaking of the said market. Thus, we examine the matter from the stand point of market economy where question of oligopsony assumes relevance. Whenever there is a situation of oligopsony, parallel pricing simplicitor would not lead to the conclusion that there was a concerted practice Dthere has to be other credible and corroborative evidence to show that in an oligopoly a reduction in price would swiftly attract the customers of the other two or three rivals, the effect upon whom would be so devastating that they would have to react by matching the cut. In Richard Whish & David Bailey in Oxford’s Competition Law15 discussed the “Theory of Oligopolistic Interdependence” as under: E “In an oligopoly a reduction in price would swiftly attract the customers of the other two or three rivals, the effect upon whom would be so devastating that they would have to react by matching the cut. Similarly an oligopolist could not increase its price unilaterally, because it would be deserted by its customers if it did F so. Thus the theory runs that in an oligopolistic market rivals are interdependent: they are acutely aware of each other’s presence and are bound to match one another’s marketing strategy. The result is that price competition between them will be minimal or non-existent; oligopoly G produces non-competitive stability…..

...Oligopolists recognize their interdependence as well as their own self-interest. By matching each other’s conduct they 15 Seventh Edition; Chapter 14-Horizontal agreements (w)-oligopoly, tacit collusion and collective dominance; Pg 561-562, 565-566 H 560 SUPREME COURT REPORTS [2018] 12 S.C.R. A will be able to achieve and charge a profit-maximising price which will be set at a supra-competitive level, without actually communicating with one another. There does not need to be any communication: the structure of the market is such that, through interdependence and mutual self-awareness, prices will rise towards the monopolistic level…. B …..The logical conclusion of the case against oligopoly is that, since it is the market structure itself which produces the problem, structural measures should be taken to remedy it by deconcentrating the market. Unless this is done, there will be an area of consciously parallel action in pricing strategies which is C beyond the reach of laws against cartels and yet which has serious implications for consumers welfare. xxx xxx xxx xxx xxx xxx D (iii) A regulatory approach

A different possibility would be to regulate the prices of undertakings that operate in an oligopolistic environment. This, however, would be a counsel of despair. As a matter of policy direct regulation should be a remedy of last resort. Competition E authorities should not be price regulators; they should be the guardians of the competitive process. Where markets are oligopolistic and entry is limited, competition authorities should be concerned with the question of whether there are barriers to entry and whether the state itself, for example through restrictive licensing rules, regulation or legislation, F is responsible for a lack of competition.” 98. In Theatre Enterprises v. Paramount Films16, the Supreme Court of United States held as under: “1-3 The crucial question is whether respondents’ conduct toward G petitioner stemmed from independent decision or from an agreement, tacit or express. To be sure, business behavior is admissible circumstantial evidence from which the fact finder may infer agreement. Interstate Circuit. 16 346 US 357 H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 561 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.] But this Court has never held that proof of parallel business behavior A conclusively establishes agreement or, phrased differently, that such behavior itself constitutes a Sherman Act offence.

Circumstantial evidence of consciously parallel behavior may have made heavy inroads into the traditional judicial attitude toward conspiracy; but “conscious parallelism” has not yet read conspiracy B out of the Sherman Act entirely.” 99. In this regard, the test laid down by the Supreme Court of United States in Monsanto Co. v. Spray-Rite Service Corp.17is relevant and is reproduced hereunder: “The correct standard is that there must be evidence that tends C to exclude the possibility that the manufacturer and non-terminated distributors were acting independently. That is, there must be direct or circumstantial evidence that reasonably tends to prove that the manufacturer and others had a conscious commitment to a common scheme designed to achieve an D unlawful objective.” 100. This test was reiterated by the Supreme Court of United States in Matsushita v. Zenith Ratio Corp.18: “…..But antitrust law limits the range of permissible inferences from ambiguous evidence in a 1 case. Thus, in Monsanto Co. v. E Spray-Rite Service Corp., 465 U.S. 752 (1984), we held that conduct as consistent with permissible competition as with illegal conspiracy does not, standing alone, support an inference of antitrust conspiracy. Id., at 764. See also Cities Service, supra, at

280. To survive a motion for summary judgment or for a directed verdict, a plaintiff seeking damages for a violation of 1 must present F evidence “that tends to exclude the possibility” that the alleged conspirators acted independently. 465 U.S., at 764 ……. …...petitioners had no motive to enter into the alleged conspiracy. To the contrary, as presumably rational businesses, petitioners had every incentive not to engage in the conduct with which they are G charged, for its likely effect would be to generate losses for petitioners with no corresponding gains. Cf. Cities Service, 391 U.S., at 279 . The Court of Appeals did not take account of the 17 465 U.S. 752, 104 S.Ct. 1464, 79 L.Ed.2d 775 (1984) 18 475 U.S. 574 (1986) H 562 SUPREME COURT REPORTS [2018] 12 S.C.R. A absence of a plausible motive to enter into the alleged predatory pricing conspiracy. It focused instead on whether there was “direct evidence of concert of action.” 723 F.2d, at 304. The Court of Appeals erred in two respects: (i) the “direct evidence” on which the court relied had little, if any, relevance to the alleged predatory pricing conspiracy; and (ii) the court failed to consider the absence B of a plausible motive to engage in predatory pricing….. xxx xxx xxx

Lack of motive bears on the range of permissible conclusions that might be drawn from ambiguous evidence: if petitioners had no C rational economic motive to conspire, and if their conduct is consistent with other, equally plausible explanations, [475 U.S. 574, 597] the conduct does not give rise to an inference of conspiracy. See Cities Service, supra, at 278-280.” 101. Similarly, in Bell Atlantic Corp v. Twombly19, the U.S. Supreme Court held as under: D

“[1-3] Because §1 of the Sherman Act “does not prohibit [all] unreasonable restraints of trade … but only restraints effected by a contract, combination, or conspiracy,” Copperweld Corp. v. Independence Tube Corp.,467 U.S. 752, 775(1984), “[t]he crucial question” is whether the challenged anticompetitive E conduct “stem[s] from independent decision or from an agreement, tacit or express,” Theatre Enterprises, 346 U. S., at 540. While a showing of parallel “business behavior is admissible circumstantial evidence from which the fact finder may infer agreement,” it falls short of “conclusively establishing agreement or … itself F constituting a Sherman Act offense.” Id., at 540–541. Even “conscious parallelism,” a common reaction of “firms in a concentrated market [that] recogniz[e] their shared economic interests and their interdependence with respect to price and output decisions” is “not in itself unlawful.” Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U. S. 209, 227 (1993) ; see 6 G P. Areeda & H. Hovenkamp, Antitrust Law 1433a, p. 236 (2d ed. 2003) (hereinafter Areeda & Hovenkamp) (“The courts are nearly unanimous in saying that mere interdependent parallelism does not establish the contract, combination, or conspiracy required by 19 550 U.S. 544 H RAJASTHAN CYLINDERS AND CONTAINERS LIMITED v. 563 UNION OF INDIA AND ANOTHER [A. K. SIKRI, J.]

Sherman Act §1”); Turner, The Definition of Agreement Under A the Sherman Act: Conscious Parallelism and Refusals to Deal, 75Harv. L. Rev. 655, 672 (1962) (“[M]ere interdependence of basic price decisions is not conspiracy”). [4-6] The inadequacy of showing parallel conduct or interdependence, without more, mirrors the ambiguity of the B behavior: consistent with conspiracy, but just as much in line with a wide swath of rational and competitive business strategy unilaterally prompted by common perceptions of the market. See, e.g., AEI-Brookings Joint Center for Regulatory Studies, Epstein, Motions to Dismiss Antitrust Cases: Separating Fact from C Fantasy, Related Publication 06–08, pp. 3–4 (2006) (discussing problem of “false positives” in §1 suits). Accordingly, we have previously hedged against false inferences from identical behavior at a number of points in the trial sequence. An antitrust conspiracy plaintiff with evidence showing nothing beyond parallel conduct is not entitled to a directed verdict, see Theatre Enterprises, D supra; proof of a §1 conspiracy must include evidence tending to exclude the possibility of independent action, see Monsanto Co. v. Spray-Rite Service Corp., 465 U. S. 752 (1984); and at the summary judgment stage a §1 plaintiff’s offer of conspiracy evidence must tend to rule out the possibility that the defendants E were acting independently, see Matsushita Elec. Industrial Co. v. Zenith Radio Corp., 475 U.S. 574(1986).”

102. After taking note of the test that needs to be applied in such cases, which was laid down in Dyestuffs and accepted in Excel Crop Care Limited, we come to the conclusion that the inferences drawn by F the CCI on the basis of evidence collected by it are duly rebutted by the appellants and the appellants have been able to discharge the onus that shifted upon them on the basis of factors pointed out by the CCI. However, at that stage, the CCI failed to carry the matter further by having required and necessary inquiry that was needed in the instant case. G 103. We are emphasising here that in such a watertight tender policy of IOCL which gave IOCL full control over the tendering process, it was necessary to summon IOCL. This would have cleared many aspects which are shrouded in mystery and the dust has not been cleared. H 564 SUPREME COURT REPORTS [2018] 12 S.C.R. A 104. We, thus, arrive at a conclusion that there is no sufficient evidence to hold that there was any agreement between the appellants for bid rigging. Accordingly, we allow these appeals and set aside the order of the Authorities below. As a consequence, since no penalty is payable, appeals of the CCI are rendered infructuous and dismissed as such. All the pending applications stand disposed of. B No orders as to costs. Kalpana K. Tripathy Appeals disposed of. C D E F G H