Treasury Committee
Oral evidence: Manipulation of Benchmarks, HC 491
Wednesday 2 July 2014
Ordered by the House of Commons to be published on 2 July 2014
Members present: Mr Andrew Tyrie (Chair); Steve Baker, Mark Garnier, Mr Andrew Love, John Mann, Mr Pat McFadden, Teresa Pearce, Mr David Ruffley, John Thurso
Questions 1-139.
Witnesses: David Bailey, Head of Markets Infrastructure and Policy, Financial Conduct Authority, and Finbarr Hutcheson, President, ICE Benchmark Administration gave evidence.
Q1 Chair: Thank you very much for giving evidence to us this afternoon. You will be in a much better position to restrain yourselves if you think there is anything that you are saying that could prejudice an inquiry that you yourselves have under way or have been briefed on. We hope that is not going to cramp your style in any way, but if you need to signal that then I would expect you to do so.
Can I begin by asking you both whether you think that other benchmarks are vulnerable to this clear conflict of interest that has been exposed between the submitters of the data and those who hold contracts that are linked to the benchmark? Mr Bailey.
David Bailey: Thank you, Chairman. Perhaps I will start off by answering that question. As you know, there are a variety of benchmarks that cover a whole range of products, and all benchmarks are susceptible to people attempting to manipulate them. They are susceptible to a variety of conflicts of interest and what is very important is that the process by which they are fixed is robust and has the appropriate governance and oversight. That is very much from an FCA perspective what we have sought to put in place via the “Wheatley Review” and via the work we have done to implement the findings of the “Wheatley Review” and also why we have led international work in IOSCO and in the Financial Stability Board in order to set the right standards for benchmarks. When benchmarks meet those standards and principles, the chances of any conflicts of interest that an individual benchmark might face being abused are minimised.
Q2 Chair: What is the answer to the question: which ones are vulnerable to these conflicts?
David Bailey: There are a variety of different benchmarks and they follow a variety of different ways of setting the benchmarks. There might be a transaction—
Q3 Chair: Just take us through the lists that you think are vulnerable.
David Bailey: As I mentioned previously, I think there is a very wide variety of benchmarks, all of which may be susceptible to someone trying to manipulate them. The important thing is having the right governance and the right controls to identify any attempts to abuse a conflict of interest or manipulate a benchmark and take appropriate action.
Q4 Chair: Your business plan said that you would look at how banks control the conflicts of interest between their obligations to clients and trading positions. When are we going to see that?
David Bailey: There are a couple of pieces of work that I would highlight, Mr Chairman. The first one is we said after we had implemented the “Wheatley Review”, we would conduct a review of the systems and controls of submitting banks. That work is under way. We also said in this year’s business plan that we would conduct further thematic work on controls around trader behaviour. We are scoping out that work and we expect that work to be conducted in the course of the next year.
Q5 Chair: When are we going to see it?
David Bailey: The work is just being scoped out, so I do not have an exact timetable. We are expecting it to take place over the course of the next year. [Interruption.]
Q6 Chair: Is there someone with an electronic device that is beeping, and if so, would you please close it down? It is the policeman at the back.
The LIBOR scam also brought to light evidence that there could be collusion between submitting banks. Are there other benchmarks vulnerable to that?
David Bailey: As I mentioned earlier, Mr Chairman, there are a variety of benchmarks set by a variety of methods, whether that is transaction-based, auction-based or submission-based. Many of those benchmarks will have a number of participants and therefore those participants may attempt to collude to affect the outcome of that benchmark. The important thing is having the right systems and controls and governance, the right standards, which is what we are promoting via our work in the IOSCO principles and the work in the FSB to promote that internationally.
Q7 Chair: Is it possible that there could have been collusion between members of the gold-fixing panel?
David Bailey: It is possible, but I have no clear evidence that that has happened.
Q8 Chair: Mr Hutcheson, are you confident that none of the data you collect is manipulated either by individual submitters with conflicts of interest or as a consequence of collusion?
Finbarr Hutcheson: Yes, I am very confident that there is no current manipulation going on with LIBOR. We have implemented an awful lot of changes and a very robust surveillance mechanism that oversees all the—
Q9 Chair: You have robust procedures in place and that gives you confidence?
Finbarr Hutcheson: Absolutely, correct. We have a—
Q10 Chair: Do you have those written down?
Finbarr Hutcheson: Yes.
Q11 Chair: Could we see them?
Finbarr Hutcheson: Yes, absolutely. I can forward them after this Committee.
Q12 Chair: Is there anything that you have heard Mr Bailey say that you would disagree with, take issue with or wish to qualify were I to have asked you those questions? If there is a lot, then take your time.
Finbarr Hutcheson: Okay, thank you. No, there are not any elements that I would take issue with. We have seen in the past with LIBOR there is potential on a submitted benchmark for a conflict of interest between those submitting and those who can profit from manipulating that benchmark. I think that the steps we have taken—I am speaking on behalf of ICE Benchmark Administration—with regard to LIBOR, and the FCA and other authorities around the world, to address those conflicts of interest have significantly closed that loop, although I cannot speak to every benchmark, but I think that is—
Q13 Chair: Mr Bailey, were you in the FSA?
David Bailey: I was.
Q14 Chair: Do you share my surprise and dismay that the FSA never thought to look at this?
David Bailey: When allegations of manipulation with respect to LIBOR came to us, we acted very swiftly in taking some of our largest enforcement fines.
Q15 Chair: I am sorry to interrupt, but what we are looking at here is structural weaknesses, where you did not need to have a report but where just by looking at the structure you could see that there were weaknesses. Why didn’t you identify those is my question?
David Bailey: Thank you. Benchmarks, with the exception of LIBOR, are unregulated—and before 1 April 2013 even LIBOR was unregulated—so they sat outside the regulatory perimeter, which made it more difficult for us to spot or have data on what was going on. When allegations and potential misconduct was put in front of us, we acted very swiftly and we have levied four of our five largest fines on LIBOR submitting banks for what they had done.
Q16 John Thurso: Can I ask you both—perhaps start with you, Mr Bailey—a very simple question, which is who uses benchmarks and why?
David Bailey: Thank you. There are, as I said earlier, a variety of benchmarks, for example, in inflation, the Retail Price Index or the Consumer Price Index; there are other benchmarks like LIBOR, which is widely used; then there are more bespoke benchmarks, for example, a benchmark that analyses the price of North Atlantic salmon, the Salmon Index. They are used for a variety of reasons, either to price financial instruments and calculate the payment on financial instruments or measure financial performance, so they are widely used by a variety of members of the public and organisations.
Finbarr Hutcheson: They are a common reference point that are available to two parties generally in a transaction, and it is easily identified what the benchmark is that they are going to refer to and they are referring to a common index. LIBOR is a typical example where it is a rate that a borrower and a lender will use to calculate interest on a loan. It is public; it is widely available; it is hedgeable. They are confident that when they need to identify the LIBOR rate, it will be published by the LIBOR administrator and therefore it is a very convenient mechanism for calculating that interest rate on a typical loan between a borrower and a lender. It is very convenient, I have to say.
Q17 John Thurso: Within that range, you have named some that are pretty straightforward, that are pretty easy to understand and where the contract on either side has a fairly clear meaning. Also within that are some pretty new, exotic, still little-understood benchmarks derived from indices that are not very old. Looking to the ultimate consumer, the small business or the unsophisticated customer of a business, what are the benchmarks that we should be worrying about? Where do those benchmarks expose the unsophisticated consumer and small business to problems? We have obviously seen LIBOR. The majority of small businesses saw that problem obviously through IRP and things of that nature, but where with these other ones is there likely to be a problem for the consumer?
David Bailey: I will start off answering that question, thank you. As I was saying earlier, there are a variety of benchmarks set in a variety of ways, whether transaction-based, whether submission-based, whether auction-based. All have different points of strengths and different weaknesses in their processes. I think the important thing that we want to encourage is that benchmarks and benchmark administrators follow the IOSCO principles and publicly state that they are doing so in a comply-or-explain manner, and then users of those benchmarks can have confidence that they have robust governance and strong oversight.
Q18 John Thurso: Is the integrity of the benchmark more important to unsophisticated customers, in your view?
David Bailey: I think the integrity of a benchmark is always very important, regardless of who the consumer is.
Q19 John Thurso: Mr Hutcheson, can you give me some sort of insight into the range of indices that are produced by your organisation and what they might be used for?
Finbarr Hutcheson: Sure. We do a number within the organisation; we do some equity-based indices and LIBOR. I have to say my experience and knowledge is purely with LIBOR. When we were successful with the Hogg Committee to take responsibility for LIBOR, that was the starting point of setting up an entirely ringfenced entity within the group that purely looks at LIBOR, so myself and my team are responsible for LIBOR only at the moment; we look at no other indices. We have consciously done that, such that we can be focused on the integrity of that index and act in the best interests of that index, rather than be looking at multiple indices simultaneously. We are of course looking to administer other indices going forward, but we decided to start entirely afresh with a new entity, as I say, that is ringfenced from the rest of the group. Sorry, but I am unfamiliar with the other indices.
Q20 John Thurso: Where I was going was that is if you take an index, we are all familiar with passive funds where you are looking at a tracker index—MSCI is probably the most familiar—by a tracker fund, you know what you are buying. It is dead simple. But life is getting more complicated because now funds that are tracking are saying, “That is the wrong weighting if you get a puff through a takeover, so let us do weighting based on dividends or let us do weighting based on earnings,” and things like that. You start to get a complexity in the indices that comes from a degree of judgment, and at every one of those stages, you get to the point where you are introducing a human element. That in turn starts to produce the problems potentially that we saw with LIBOR and the very complexity of some of the modern synthetic credit indices. I am into areas I have not the slightest understanding of; I am not even sure the people who work with them really understand them. How do we, in looking to ensure that where these indices are used to create a benchmark, end up with a benchmark that is robust enough to have strength in a contract?
Finbarr Hutcheson: Sure. If I can draw parallels to how we are operating LIBOR at the moment, an over-arching concern for us that we are trying to address is ensuring that users of the benchmark know exactly how it will behave in any given circumstances. Benchmarks can be complicated—you are right—and the administrator needs to make decisions. To the best of our ability, we will publish exactly how we will behave in certain circumstances. In fact, in about two weeks’ time, we should start a consultation on one element of a policy we are implementing and taking on feedback from all users of the index as to if they think that policy is sensible or we should adjust it going forward. But no matter how well we anticipate what could happen in the future, and that is exactly what we are trying to do, you still need that over-arching regulatory regime and an administrator that in themselves have no conflicts of interest and therefore—the expression I used earlier—really act in the best interests of that index and the integrity of the index itself. You need that kind of backstop of an administrator that is focused on that index and is responsible for that index, and as I said, a regulator that oversees that, and ideally common regulation across jurisdictions.
Q21 Teresa Pearce: Can I take you to the “Wheatley Report”? The “Wheatley Report” recommended a hierarchy of how LIBOR submissions should be calculated, starting with real transaction data at the top, right down to expert judgment. Have you any idea what proportion of LIBOR submissions are still based on judgment?
Finbarr Hutcheson: I do not have any figures to hand on that. What I can tell you is on a daily basis all the submissions we receive from banks—and I think it is 611 on a daily basis—with each and every submission that we receive, the bank indicates to us exactly whether that was based on expert judgment, based on interbank unsecured funding, for example, the sort of transactions we would like them to use, or whether they use other transactions that are related to that market but not directly in that market to derive their submission. We do get data on a daily basis indicating exactly the answer to that question, although I would have to say typical submissions from a bank are based on both transactions and other observations of what is happening in the marketplace and some expert judgment.
Q22 Teresa Pearce: Are you saying that it is not possible for LIBOR submissions to be based on objective transaction data only and there will always be some sort of judgment playing a role?
Finbarr Hutcheson: It is an interesting element. There are certainly flaws if you rely entirely on expert judgment, but one of the benefits you do get is in the event that there are no transactions, you have a fallback methodology there then. The vast majority of submissions that we see on a daily basis have transactions that are used to help the bank derive that and formulate their submission, but you cannot guarantee that there will always be transactions available. One of the things we are working on with the FCA and other authorities globally is how to incorporate more transactions into that calculation, such that hopefully at some point we will be able to get to purely transactions, but I think the expert judgment is currently a part of it and is a constructive part of it. It is important for our role that we scrutinise that expert judgment and we look at the underlying transactions and satisfy ourselves that they are true bona fide transactions, they are accurate and they are what is driving the submission, when possible.
Q23 Teresa Pearce: The first edition of the code of conduct was published by the BBA, I think, in July 2013.
Finbarr Hutcheson: Correct.
Q24 Teresa Pearce: The second edition was published in February 2014, and they are virtually the same.
Finbarr Hutcheson: Yes.
Q25 Teresa Pearce: Did you learn nothing in that interim time? How many more codes of conduct do you think there will be before we get to something that is similar to the “Wheatley Report” recommendations?
Finbarr Hutcheson: The very first version of the code of conduct was implemented by the BBA and it was approximately six months or maybe seven months after the “Wheatley Report”, so it was very much a response to the “Wheatley Report” and it was one of the recommendations that came out of the “Wheatley Report”. We took that document at the time; we made some very minor revisions—I have to say, more logistical revisions—because it was no longer the BBA’s and it became ours. It is on the slate of work for our Oversight Committee that we have put in place to look at any potential changes to the—
Q26 Teresa Pearce: Do you consider that you have implemented enough of the Wheatley reforms or do you think there needs to be more codes of conduct until we get to that point? Have you—
Finbarr Hutcheson: I think the code of conduct will continue to evolve but all of the recommendations from the “Wheatley Report” I would say have been implemented or continue to be worked on to be implemented. What I just referred to in terms of anchoring submissions more and more in transactions and ultimately reducing that expert judgment as much as possible, that is an ongoing initiative that we currently work with the FCA, Bank of England and other authorities globally to continue moving down that path, but that is not a turnkey solution.
Q27 Teresa Pearce: Just a last question: given that before LIBOR became a household word the BBA had responsibility to monitor LIBOR, there was no code of conduct—it was not a regulated activity—so where did the responsibility lie for making sure that it was not a rigged market, in your opinion?
Finbarr Hutcheson: Me specifically?
Teresa Pearce: Or both of you, with all the hindsight you have now.
Finbarr Hutcheson: I guess with hindsight one would say it—
Q28 Teresa Pearce: Do you think it was an accident waiting to happen?
Finbarr Hutcheson: It would be helpful if it had been regulated from an earlier point in time, I think I can say that.
Q29 Teresa Pearce: Anything to add?
David Bailey: There are a couple of points I would make on that, first in direct response to your last question, which would focus on at the time the BBA ran the process— which we have uncovered through our enforcement action. Clearly, there was inappropriate behaviour in a number of banks, and I think all the parties in that process had joint ownership of the benchmark that was produced. It is markets that set benchmarks. Regulation can impose standards and governance on that, but it is markets that should set benchmarks with the regulator putting in place the requisite recommendations or principles that they should follow.
Just coming back to one point you asked Mr Hutcheson about earlier in terms of the use of transaction data, the one point I would like to make on that is that even where you have benchmarks that are entirely transaction-based, such as certain FX benchmarks, they themselves are susceptible to attempts to manipulate them, as we have seen over some of the concerns that were raised on FX benchmarks. So just moving to a state of purely being based on transactions on its own is not sufficient, in our view. You have to have the right governance and controls around that.
Chair: Could I ask colleagues to be brief, please, because we have another session we need to get in. We have a vote at 4 pm, which is problematic.
Q30 Mark Garnier: Mr Bailey, continuing with the transactional thing, how concerned are you about the equity closing prints?
David Bailey: Where there are equity benchmarks, for example, the FTSE 100—
Q31 Mark Garnier: No, I am thinking about closing prices of individual stocks.
David Bailey: Closing prices of individual stocks as in run by an auction process on the FTSE, for example?
Mark Garnier: Or the average bid on the spreads.
David Bailey: Where they are set on a regulated market—for example, the London Stock Exchange—then we have regulatory oversight of that and we look very closely at the procedures that the exchange itself may run. Where they are, for example, benchmarked, such as the FTSE benchmarks, then the FTSE itself is unregulated or we do not regulate the benchmarks that FTSE produce.
Q32 Mark Garnier: This is the index we are talking about, the FTSE 100?
David Bailey: Yes, for example, but the same standards and principles that we put in place through IOSCO and to promote good governance and good systems and controls would apply equally to those benchmarks.
Q33 Mark Garnier: Have you brought any prosecutions against anybody for trying to manipulate closing prints?
David Bailey: Not to the best of my knowledge in front of you today. I can go away and check that.
Q34 Mark Garnier: It would be very useful. I spent 27 years as an equity investment banker and a hedge fund manager before I came to this place and it was common knowledge that people were trying to manipulate closing prices both in the UK market and the US market. I spoke to a couple of sales traders this morning just to find out if my information was out of date, and they said, “Oh yes, that goes on all the time,” as if it is a well-known secret.
David Bailey: If I may, may I go back and I will check out whether we have taken any action?
Q35 Mark Garnier: Have you done any investigations is what I am looking for.
David Bailey: Okay. I will take that back, check that and write to you.
Q36 Chair: Do you know if you have had any investigations?
David Bailey: Again, I will have to get back to you, specifically on equity closing prices.
Q37 Mark Garnier: It is a serious issue. I am sorry to press this point, but this is something that is very, very widely-spoken about among the equity traders across the world. This is not a minor sort of thing. This is a big, big deal, and I just amazed that you have not thought about it. Maybe this is outside the remit of perhaps what you necessarily look at, but if nothing else, this is market abuse, surely.
David Bailey: In terms of it being market abuse, I will have to look at the individual instance to see what was taking place. But as I said, with respect to any investigations we have done, I will take it away, get the information and supply that to the Committee.
Q38 Mark Garnier: One final question, if I may, Chairman, unless you want to have a supplementary. Just getting back to the FX thing, we are nine months into this investigation, a number of people have been in touch with me—and I am sure other members of the Committee—wondering where the regulator is. I am sure you probably cannot give any specific detail, but after nine months, nobody has heard anything—lawyers, banks, people involved in the market. Is anything going to come out? Is there going to be any information coming back from the regulator about this?
David Bailey: You are quite right that I cannot give you any specific information, but we continue to progress our work and as soon as we have something to say, then we will obviously say that.
Q39 Mark Garnier: Any idea roughly when?
David Bailey: I am afraid I cannot give you any timing. It is a complex process.
Q40 Chair: We are all amazed that this equity closing price does not seem to have been looked at all.
David Bailey: I will take that point away, thank you.
Q41 Steve Baker: Mr Bailey, earlier in the session, you seemed a bit reluctant to name any specific benchmarks, so can I ask you, is the FCA going to be helping to write the list of benchmarks that will be brought into the regulatory perimeter under the “Fair and Effective Market Review”?
David Bailey: The review is absolutely being taken forward by the FCA, alongside the Bank of England and Her Majesty’s Treasury, so we will fully support that review and will be part of the group that looks into whether there should be further benchmarks that are recommended to be brought inside the regulatory perimeter, absolutely.
Q42 Steve Baker: Could you give some indication of the total number of benchmarks that you might expect to find you are regulating?
David Bailey: At this stage, I am afraid I cannot prejudice that review. It has only just kicked off. I do not want to say anything now that might prejudice the outcome of that review.
Q43 Steve Baker: You could not even say tens or hundreds?
David Bailey: I am sure the review, which is just kicking off, will look at a variety of benchmarks. As for the final number, I am afraid I cannot put anything on the record today.
Q44 Steve Baker: To what extent are conflicts of interest and other risks to the integrity of benchmarks alleviated by relying on real transactions rather than estimates or guesses?
David Bailey: Different benchmarks are set in different ways, but they all have risks associated with them. As I was saying in an answer earlier, even benchmarks that are based purely on transactions are susceptible to attempts to manipulate the benchmark, and therefore it is important that you have the right systems and controls, the right oversight and that the benchmarks follow the IOSCO principles that we have had a strong hand in promoting internationally.
Q45 Steve Baker: Just turning to the gold fix in particular, that seems to me to be tremendous concentration of power in a small number of corporate hands. I understand it is an auction with limit orders. Have you looked at how realistic it is to get the gold fix to be based on transactions that have not been manipulated?
David Bailey: We have obviously looked at the gold fix. We have taken enforcement action and fined one firm for its activities on one day in that benchmark. The industry itself is looking at potential revisions to that benchmark. The World Gold Council are holding a conference next week to discuss potential ways of reforming or potential ideas to reform that benchmark. We will fully support any reforms following the robust principles we have put out as part of IOSCO.
Q46 Steve Baker: Will you be looking at how central banks participate in the gold market through this mechanism?
David Bailey: To the extent that benchmarks are within the regulatory perimeter, we will look at those benchmarks. There are a number of benchmarks that are produced by national bodies that are unregulated currently and typically fall outside of our specific regulatory remit.
Q47 John Mann: Mr Bailey, good afternoon. Let us talk gold: fixed by five, announced twice a day. How long do those teleconferences last?
David Bailey: I cannot answer that question specifically. I have not listened into one of the calls myself, but it is an auction-based system. As you say, it is run twice a day.
Q48 John Mann: If I was to suggest that some of those teleconferences last over an hour, would that seem unreasonable?
David Bailey: It depends on how long each individual auction would take. My understanding is, however, that it is not just those banks on the call, but other parties can listen in. There is only a certain number of banks that can submit bids and offers into the auction, but other members of the industry can listen in, in order that they can submit bids and offers via the banks that participate in the auction.
Q49 John Mann: So those five banks’ clients can be updated while the fix is going on?
David Bailey: That is my understanding, yes, while the auction is going on.
Q50 John Mann: Have you read the analysis by Dimitri Speck of movements in gold prices from 1993 to 2012?
David Bailey: That particular report I am not familiar with, but my technical team back at the FCA and I have looked at a number of people’s academic research. I am very much looking forward to hearing Mr Thomas, who is on the next panel, who I believe has looked into this issue as well.
Q51 John Mann: What Speck has been able to analyse is that there is a significant movement every day at the time of the fix in the gold price and there has been throughout that period. Does that not give you, as the regulator, significant concern?
David Bailey: We have looked at a number of pieces of research that have come to a variety of conclusions and highlighted a variety of price movements around the fix, some around only the afternoon fix versus the morning fix, which we are factoring into our thinking about both the gold-fixing process but also other benchmarks as well.
Q52 John Mann: But in the afternoon fix, the movement is more significant than the morning fix. If you look at the movements over that more than 20-year period, what you can see unambiguously is that the price drops every day significantly compared to the rest of the day during the fix. Therefore, I put it to you that this market has been manipulated.
David Bailey: Thank you very much. In response, I would say that what that demonstrates is that there are price movements around the fix. There may be a number of reasons for that taking place, one of which could be attempts to manipulate the benchmark, others could be unwinding of hedging of transactions that have expired at the hedge.
Q53 John Mann: But the trend is absolute; it is quantifiable; it has been provided externally—you are the regulator—there is motive and the clients can listen in, the clients can be consulted during the fix, and the fixes can last for more than an hour. That is not what the system in 1919 was meant to be about, but that is what it has become, hasn’t it? This is a flawed and manipulable market that needs to be sorted by yourselves, doesn’t it?
David Bailey: We do not regulate the gold-fixing process. That is not a regulated activity, but in line with other benchmarks, we very much would support and recommend that any processes that are put in place for the fixing of the gold price follow the robust standards we put in place via IOSCO.
Q54 John Mann: But there is no reason whatsoever why gold can be fixed by observing actual trades, no reason, and if it is silver, of course we are talking three banks rather than five, so an even smaller number. You are the regulator. You have caught Barclays out already in relation to this. Why aren’t you coming—be it to this Committee or far greater than us—and saying, “This is no longer relevant in the modern era. It is open to abuse. We have found abuse. This system needs to change.”?
David Bailey: I will come back to something I said earlier, if I may, Mr Mann, which is that it is for the market to set the actual prices and the actual benchmarks, and the important thing from our perspective is it is done by a robust process. As the World Gold Council moves forward in working out any reforms that it wants to put in place for the gold-fixing process, we would support that following the robust standards in the IOSCO principles.
As to whether any further benchmarks should be brought inside the regulatory perimeters, as I was saying to Mr Baker earlier, I do not want to prejudice the outcomes of the “Fair and Effective Markets Review”, which is due to make recommendations in due course as to whether further benchmarks should be brought inside the perimeter.
Chair: Steve Baker has a quick rejoinder.
Q55 Steve Baker: Just thinking about the obvious vulnerability in LIBOR to manipulation—whether it is a guess or a professional judgment, whichever way you put it—do you not think that this whole conversation reflects a profound change of culture from one based on honour and trust to one where there is just an assumption of irresponsibility and self-interest that is only moderated by oversight? Is that what is happening?
David Bailey: I think one of the reasons that the Government has initiated the “Fair and Effective Markets Review”—it is something bank and the FCA fully support and are participating in—is to restore confidence in the behaviour that takes place on trading desks throughout a range of markets. I expect that review to come up with robust recommendations of how to make sure we can have confidence that the behaviour is of the right standard.
Q56 Steve Baker: It is not really what I asked. I asked if you think there has been a profound cultural change from a culture of honour and trust to where you have to assume people have been irresponsible and are being watched by the regulator. Is that cultural change of what is happening?
David Bailey: Concerns over the culture that exists is exactly why the “Fair and Effective Markets Review” has been put in place.
Steve Baker: It sounds like a yes.
Chair: Just the term “gold-price fixing” has its cultural problems, hasn’t it?
Q57 Mr Love: With research showing that the banks collectively have the means, motive and opportunity to manipulate the gold fix, can you leave it to the World Gold Council to sort this out? Surely, there is a responsibility for the regulator here to ensure public trust in what is happening with the gold fix.
David Bailey: That is exactly why we have been taking a leading role in promoting the right regulatory standards for benchmarks and the right standards for benchmarks, both through the IOSCO process—where we have developed 19 principles, which we believe that all benchmarks should follow—but also in the FSB, where we are promoting that work internationally. With respect to specific benchmarks, it is for the market to set the actual benchmark level and it is for the market to put in place the process around that, but that should meet the robust standards.
Chair: We have to move on now, we really have.
Q58 Mr Ruffley: Turning to those 19 principles that are meant to entrench more integrity into the system, going through each of the major benchmarks produced in the UK, how far do those benchmarks in each case comply with those 19 principles?
David Bailey: Those benchmarks were published in July last year. As part of that work, IOSCO asked that all benchmark administrators looked at the principles, and within a 12-month period publish whether they complied, where they did not comply and explain why that was the case. That work is ongoing. IOSCO and the national regulator will look at the output of that and IOSCO have said that within another six months after that, they will decide whether any further action is appropriate, so we are still waiting for the benchmark administrators to conclude their appraisals.
Q59 Mr Ruffley: When is the first phase, because the second phase is six months? The first phase ends when?
David Bailey: The first phase ends at the end of July this year.
Q60 Mr Ruffley: Then it is six months after that?
David Bailey: IOSCO said in its report last year that it would decide within that six months whether any further action was necessary based on the take-up of its principles.
Q61 Mr Ruffley: If in respect of one benchmark half of the principles were breached, what sanctions would IOSCO have? Do they just slap on the wrist and say, “They do not comply with half the principles of benchmark X.”? What can happen?
David Bailey: IOSCO itself does not have any ability to take any enforcement action. It has no statutory responsibilities.
Q62 Mr Ruffley: Do you, the FCA, in relation to the 19 principles that are not being followed?
David Bailey: If it was an IOSCO principle that was not being followed on an unregulated benchmark, then I do not believe we have any statutory ability to take action against the benchmark administrator if it was an unregulated benchmark. One of the things that the—
Q63 Mr Ruffley: Could I just stop you there? Your answer was very helpful. In a situation like that and you being the principal regulator in the UK, the greatest global financial centre on Planet Earth, I am presuming you would want to get the powers so that you could regulate these administrators to ensure that they complied with the principles.
David Bailey: That is something that the “Fair and Effective Markets Review” will be absolutely looking at and deciding whether to bring further benchmarks or to recommend that further benchmarks are brought inside the regulatory perimeter by the Government.
Q64 Mr Ruffley: How far does LIBOR currently comply with the 17 IOSCO principles?
David Bailey: Part of the work that my CEO, Martin Wheatley, is co-chairing under the auspices of the Financial Stability Board is a review of the major interest rate benchmarks, LIBOR, EURIBOR and TIBOR, against those IOSCO principles, and that work is due to report back later in the year.
Q65 Mr Ruffley: By the end of this calendar year?
David Bailey: Yes.
Q66 Mr Ruffley: What about the gold market fix? How does that comply currently as we speak now today with the 17 IOSCO principles?
David Bailey: As an unregulated benchmark, the FCA has not conducted an assessment. However, in line with other benchmark administrators, I would expect that to be a benchmark that publishes whether it complies with the IOSCO principles in line with other benchmark administrators over the course of the next month.
Q67 Mr Ruffley: A final question, Chairman. IOSCO puts the onus on the administrator. There is a view—isn’t there?—that the onus should be on data submitters to ensure that the benchmarks reach the required standard. Shouldn’t that be the point of focus, data submission, rather than the administrators?
David Bailey: With benchmarks, I think it is a shared responsibility, so the submitters should meet high standards, as should the administrators, and the IOSCO principles set high standards for both.
Q68 Chair: Is there anything you want to add while you are here?
Finbarr Hutcheson: No.
Chair: You are going to send us a bit of written documentation we have asked for. Thank you very much for giving evidence to us this afternoon.
We are going to go straight on to the next session. I have been informed that there may be a vote very shortly, in which case I will have to ask colleagues if they can return after the vote. Thank you very much.
Examination of Witnesses
Witnesses: Rhona O’Connell, Head of Metals Research and Forecasts, Thomson Reuters, and Alberto Thomas, Partner, Fideres LLP, gave evidence.
Q69 Chair: Thank you very much for coming to give evidence to us this afternoon. Can I begin by asking you, Rhona O’Connell, whether you think that market participants are aware of these conflicts of interest faced by banks that are part of the fixing process?
Rhona O’Connell: There has been an increasing amount of debate about it over the course of the last 18 months or so, I would say. There is one camp that is—
Q70 Chair: Sorry, let me put it slightly differently. Were they aware of this prior to 18 months ago?
Rhona O’Connell: There is a big debate in the market and a lot of people say that there is no conflict of interest; there is another camp that says that there is. The LIBOR problem has focused everyone’s attention, obviously, and as a result there is an increasing amount of conversation through the markets as to whether there are conflicts of interest or not.
Q71 Chair: How can it be claimed that there is no conflict of interest?
Rhona O’Connell: Would you like me to run through how the fix itself works? That might help to explain.
Q72 Chair: We have a rough idea, but if you think you can illustrate how it can be reasonably claimed that there is no conflict whatsoever, that is something we should have on the record. Is that your view, by the way?
Rhona O’Connell: I cannot say for certain that there is no conflict, because I have not worked with the fix.
Q73 Chair: Then there is scope for a conflict of interest?
Rhona O’Connell: There is scope for it; I cannot possibly say that there is not, because you will find that I have—
Q74 Chair: Have market participants been aware for years that there is a scope for a conflict of interest?
Rhona O’Connell: I would say that anyone with any experience and who thinks deeply about it would be prepared to accept that there is a possibility of a conflict, yes, because a bank may have a proprietary position that would go against all of that.
Q75 Chair: That has been evident for years?
Rhona O’Connell: I would have thought so.
Q76 Chair: In that case, why has it been tolerated for so long?
Rhona O’Connell: That I do not know.
Q77 Chair: Is it back to the question that Steve Baker asked: this was an old boy network, the legacy of an old boy network, where it is based on trust that has long since been displaced by a more cut-throat culture?
Rhona O’Connell: My understanding of it is that when a bank receives an order for gold to be bought or sold through the fix, they are essentially contractually bound to fill the client at that fix. It is very similar to the way the London Metal Exchange works. So a bank will take an order and that has to go straight through to the fixing process. The bank is essentially on risk over the period of time from when the order is received to when the price itself is fixed, because you are required to fill the client at that fixing price. Essentially, as far as the fix is concerned, the fixing members are facilitators. Where it is a potential problem is if they have positions of their own.
Q78 Chair: If they have positions of their own or if they are aware of advance positions?
Rhona O’Connell: Yes.
Q79 Chair: Defenders of the gold fix point out that they have to fill the buy or sell orders at the fixing price. That is okay, but it still leaves the potential to manipulate a benchmark price through real transactions towards a single, very large, strategically-timed buy or sell order in order to fix the market. Isn’t that what was going on at Barclays?
Rhona O’Connell: Up to a point. It is important to bear in mind that it is an auction process, and at any stage during the fixing process, which can take anything from three minutes to much longer—the record is an hour and three-quarters on Black Monday in 1987—any interested participant, be it a client of the fixing bank or a client of another bank who has submitted orders through to the fixing bank can change their order, they can pull it completely or they can put on a new one. If there is a substantial amount of buying going through the fix, which takes the price higher and higher because it is out of balance, ultimately you can only fix if the buy orders and sell orders are pretty much matched to within a tolerance of 50 bars, which is 2,000 ounces. If there is a lot of buying going through the fix, the price will be raised accordingly until there is sufficient selling interest to match.
Q80 Chair: Do you think that the FCA acted in a timely fashion or do you think they could have been on this earlier?
Rhona O’Connell: With respect to?
Chair: Their investigation. Were you here for the earlier questioning?
Rhona O’Connell: Yes, I was.
Chair: I asked the same question effectively of the FCA.
Rhona O’Connell: It has taken a bit of time, to be perfectly honest. My personal opinion, as opposed to an educated opinion, because I am not a regulator, is that when the—
Q81 Chair: You are very well-informed.
Rhona O’Connell: Thank you.
Chair: And close to this market on this very point, so your view is of some value. Do you think that they should have gone into this earlier or do you think they did it in a timely manner?
Rhona O’Connell: I would have thought that once the problems came into public knowledge with respect to LIBOR that everything should come under closer scrutiny.
Q82 Chair: They were slow out of the traps after LIBOR, is that your view?
Rhona O’Connell: I would say that, yes.
Q83 Chair: I have not asked you any questions, Mr Thomas, but again, can I ask you a question similar to the one I asked Rhona, which is—
Alberto Thomas: Absolutely, go ahead. You will not get a corporate answer from me, sir.
Chair: whether there is anything there that you have heard in those exchanges that you want to take issue with and whether there is anything you want to add, rather than my going through—
Alberto Thomas: I would have loved to be on the previous panel, but unfortunately I cannot answer every single question, Chairman. In answer to the previous question, yes, I agree with Ms O’Connell, absolutely, yes. The evidence was there, and as Mr Love said before, in order to manipulate a benchmark, you need to have means, motive and opportunity. The seeds are there not only on gold: we have had them in LIBOR; we have them in FX. I can tell you that we work on all of those and we have now another list of five or 10 that we are going through in due course. The situation is the same across a lot of financial markets, because these markets have all evolved throughout the last century, the information transmission mechanisms have changed and the speed of transmission of information across the markets has changed radically from 1919.
When we look at the LIBOR or FX, the WM/Reuters fixing on FX, or whether we look at the London fix on gold or silver or gas or Baltic exchange or whether we look at ISDAFIX—and I can go on—the problem is that a lot of these markets present similar problems. The means to manipulate them are there. It does not mean that they are manipulated as markets. In a lot of these benchmarks, there are flaws in the design. It is impossible, from my perspective, working on trying to identify whether manipulation has occurred. In a number of markets, like the gold markets, we have no access to transactions that occur in the fix. We only have access to the futures markets. I will just give you an order of magnitude even when we can get access: if I want to do an analysis on the future markets, I pay £500 to access to all the transactions for the last five years microsecond by microsecond. If I want to have access to quotes in the physical gold market, I have to spend about £20,000, £25,000, and I only get quotes; I do not get any volumes or any actual transactional data.
It is very, very hard to identify whether manipulation occurred, but the means is there, the opportunity is there, and if there is a conflict of interest in certain situations, or if on a certain day the dealers who submit the quotations or make the submissions to the panels have an economic interest, then the possibility is there. It does not mean that they do it all the time. As a matter of fact, as we worked across through all the benchmarks I mention, the ratio is approximately similar. In other words, what we identify as a ratio in terms of number of days when we are into that potential possible manipulation to the overall number of days is roughly the same. Therefore, it does not mean that they do it every day; they do it opportunistically, and sometimes they do it up, sometimes they do it down. They are agnostic. That is what we found.
Q84 Mr McFadden: Is this how a market should work, this process of five people on a phone call?
Alberto Thomas: Now we have four. You are absolutely right. The gold market is quite unusual, and if I go through, going back to the one of the questions that has been asked before, does it meet the 17 principles or whichever other sets of principles we want to take—the GFMA is another good set of principles—the fact is that the gold is probably one of the worst.
Q85 Mr McFadden: Why is it one of the worst?
Alberto Thomas: Because it does not tick any of the boxes. If you go, for example, through the list of what those criteria are, it does not have good governance; there is no way to audit what is happening on that call. There is no data transparency, so no one else can question whether the fix is correct or not. It is a tiered market. You have four people or five people on a call and then you have another market that keeps trading in parallel during that period of time. The evidence we found is that there is so-called leakage of information. What happens in the first moment of that call is a very, very good indicator of where the fix is going to be. In other words, nobody should really know what has happened on the call, but the whole market ends up knowing, because, as it was described before, you have those five banks on the call. Each of those five banks have behind them brokers and other banks, which are not members of the panel, and then each of them has clients, and all of them are on the telephone and all of them are being fed information of where the orders are. This information feeds through the market, to the futures market, so effectively you have a massive potential for insider trading in the market and market manipulation. It does not mean it happens every day, but the opportunity is there.
Q86 Mr McFadden: Let me ask you about that. What you have outlined is the potential, the means by which the market could be manipulated. Is it your belief that that potential is being used in the market and is being manipulated?
Alberto Thomas: The analysis we have done has concluded that there is a very strong indication of manipulation, yes.
Q87 Mr McFadden: Tell us about secrecy in this, because you have mentioned that it is quite hard for someone to follow this closely; you have to pay these fees to understand what is going on. What is all the secrecy about? Again, a market should be an open thing, by definition, where a product is bought and sold. Why is there so much secrecy?
Alberto Thomas: Some markets have evolved again over time. The gold market used to be very much discussed around the table in 1919, and that has been exploited somehow. It is no longer the case. There is a possibility to have a lot of these markets, over the counter markets, moving on an exchange and there is clearly a lot of interest by the dealer community to keep control of the information. Information is money.
Also, if you want to be a member of the panel that had a prestige element—maybe it had it, it no longer has it—it had reputational importance for the banks who are sitting on the panel, because they had a very important role in the market and that was bringing business to them, both from a marketing perspective, brand perspective, but also purely for the fact that everyone had to channel their orders. If they wanted to trade at a fix, they had to channel their order through one of these five banks, so that brings business. Also, there is an interest to keep the market as an over-the-counter market, but this is true not only for gold but for a number of other commodity markets as well or other derivative markets, which are still over the counter.
Q88 Mr McFadden: Is the heart of this problem the problem that the Chairman opened with in today’s session, that we have benchmark prices being fixed by people who have an interest in manipulating those prices through their own trading books? Is that part of what is going on? Is that a good way to understand it?
Alberto Thomas: That could be a driver. Ultimately, all of the dealers have a conflict. They are one of the major trade counterparties on the futures, on the COMEX futures in gold, and they are at the same time sitting on the gold-fixing panel, the members of the London gold fix, so there is this conflict of interest. It should be dealt with. There are regulations in place.
Q89 Mr McFadden: Pardon the pun: how can this be fixed? What reform would you like to see? If you were designing an honest market or a more transparently honest market that was less susceptible to manipulation that the one that we are talking about, what would be its key features?
Alberto Thomas: I think there are three elements. One element is what is being done currently, which is introducing better regulation, better governance around the benchmark, and I think the set of principles laid out by IOSCO are very good. There is no problem with that.
The second pillar should really be enforcement, and that takes the shape of two different forms. One is definitely the regulators. They have a very important role, and I am not sure they have the resources, frankly, to enforce in the same way as their US counterparty is doing.
The third aspect that I see all the time is we are involved in disputes around benchmarks—we are leading experts on a number of class actions in the US on this topic—and it is that there is a very big discrepancy between the UK and the US, as you know, from a legal perspective. We have pension funds in the US that have enshrined in their statutes a fiduciary duty—there are pensions here who have lost billions of dollars in these cases potentially, if they are proved right—and they take it upon themselves to act and enforce those rights. Another part of what we do in the UK is to—[Interruption.]
Chair: We will have to adjourn the hearing and we will resume in 10 minutes.
Sitting suspended for a Division in the House.
On resuming—
Chair: Thank you very much indeed, both of you, for your patience. Pat, you had the floor.
Q90 Mr McFadden: Mr Thomas, I think you were just coming to the end of an answer, outlining some key features for remedying some of the weaknesses you see in the current situation. Do you want to just finish your point now?
Alberto Thomas: Thank you, yes. I was saying two pillars, the first one definitely regulation. The second one I was saying is we seem to have a reticence in this country, and in Europe in general, in terms of the investment community—institutional investors, pension funds, insurance companies, the financial sector in general—to enforce their fiduciary rights in these types of cases. That is a big obstacle and that is why we see a lot more litigation in this area going on, on the other side of the Atlantic, than we see in Europe. That is another important point. In particular, apart from the code of conduct on gold, as we touched on gold but we could talk about the other benchmarks as well, I think there is a very important flaw in the gold fix construct, which is the way this auction works.
Earlier, there was a question about the length of the call. On average, it lasts 10 minutes. Sometimes, it last up to two hours, very rarely, but one hour occurs quite often. Now, as I said, there is this leakage system where the information that is discussed on the call gets transmitted to other market participants and influences the futures market, the derivatives market, at the same time, which is a problem. The fact is that no one commits. The participants who say, “I want to sell gold,” or, “I want to buy gold at a fix,” do not commit at any time to transact at any level. Therefore, you have this back and forth during the auction where they can revise their orders and say, “No, at this price I am not in,” or, “At this price, I am in.” The way an auction like the gold one is supposed to work, it should give, at every pricing point around today’s level, what is my interest to buy, what is my interest to sell and these are my orders, not that I can revise those orders back and forth during the call, if that makes sense. It is a very flawed design in terms of the way it works.
Q91 Mr McFadden: You said there were other benchmarks that you had concerns about, too. The regulator has been a little reluctant to talk to us about the potential manipulation of other benchmarks, at least by naming them. Let’s go back to where I began. For our constituents, they expect a market, if that word is properly and plainly understood, to be an honest process of buying and selling, which gets to a price. What other benchmarks, other than the ones that have been in the public domain, and for many people this began with LIBOR, although these things may have gone on before that? We have LIBOR out there. We have gold out there. We have foreign exchange out there. What other things would you suggest a committee like this or politicians and regulators should be interested in? Would it be the oil price? What other benchmarks?
Alberto Thomas: The oil price, yes. There is already a lot of evidence out there and, at least in the US, there is already litigation going on on that. We have found evidence on other interest rate benchmarks that are of big relevance, I think, to individuals in this country because on some occasions everyone has exposure to fixed interest rates through mortgages or other investment products. In particular, I mentioned before the ISDAFIX, which is the equivalent of LIBOR for long-term interest rates. I would recommend that you look into that. That is one.
Q92 Chair: Is there anything you want to add? You have been quiet now for a number of exchanges?
Rhona O’Connell: Thank you, sir, yes. I think one of the most important areas, and Mr Thomas has already touched on this, is transparency. When the bankers or their clients are putting orders through to the fix it is only ever a net position that is declared. You may declare yourself to be a buyer of, for the sake of argument, 10 bars, which is 4,000 ounces. That is a net position when you have aggregated everything from your clients. You could be a gross buyer of 110 bars and a seller of 100, but you are only declaring the 10. There is a lack of transparency and a lack of clarity about the actual volumes that are involved in the fixing process at any one time. That should be addressed.
Chair: This whole subject is one of those stones which, when you turn it over, you find a good deal more and each question leads to an answer that leaves one a little more concerned and that answer does, too, and certainly need further thought.
Q93 Steve Baker: Why is it that the gold market continues as it does? Why does it continue as OTC plus the fix?
Rhona O’Connell: One of the senior dealers in the market said to me recently that the stop price is basically the futures price plus an exchange for physical. I think he was basically saying that he thinks the futures markets are probably more influential than the physical market per se. I can understand that because of the volumes that go through and because of the nature of how a futures market works. If there is a lot of selling pressure, you can hit sell stops and that can have a disproportionate effect on the price as against the underlying physical market that lies behind it in London. To answer your question directly, it has been sustained in this way for many years because, funnily enough, of lack of transparency as much as anything else. If you trade in the OTC market obviously your counterparty knows what you are doing but nobody else does, which is one of the reasons why the central banks use it, because central banks do not stand in the gold market in order to be seen.
Q94 Steve Baker: Who would benefit and who would lose out from gold being traded on a regulated exchange?
Rhona O’Connell: I would have thought the whole market would benefit, personally.
Q95 Steve Baker: You think the whole market would benefit?
Rhona O’Connell: I would have thought so.
Alberto Thomas: It depends from which point of view, short term or long term. I agree that from a perspective of the common good, yes, everyone would benefit. The ones that would probably lose out in the short term are the four or five dealers, yes.
Q96 Steve Baker: Why does a modern exchange not just emerge spontaneously among either the current players in the market or among competitors? Why does it not just naturally move to an exchange basis?
Alberto Thomas: In some cases, it would like to move but it can’t.
Q97 Steve Baker: Why can’t it?
Alberto Thomas: I do not know if you looked at all into the credit default swap market. There have been specific allegations of collusion by the 13 or 16 banks that control the market in order not to license the documentation or the indices or the information to other incumbents who want to trade those products. The allegations are that the Eurex Exchange in Europe and the CME in the US wanted to set up an exchange for credit derivatives and they were not allowed.
Q98 Steve Baker: Could I ask you to comment on the scale of the market? Is that a factor in preventing people from moving to other mechanisms, the fact that only a few people trade gold in the quantities required to participate?
Rhona O’Connell: There are only a few market makers who participate in the fix, but there is any number of organisations or individuals who are involved in the market itself. To go slightly off at a tangent and refer to silver for a second, one of the fixing members currently has 320 counterparties who have contracts with them for whatever tenure it may be based around the silver fix. It would be more for gold.
Q99 Steve Baker: Just reflecting on the Plunkett case for a moment, wasn’t the issue there that the particular individual manipulated the market for the purpose of loading the dice against somebody to whom he had sold a derivative? The problem was not so much the participation in the market, which it seems to me is a quite clearly defined process. It was the fact that the market participant had an ill intent. Wasn’t that the problem?
Rhona O’Connell: Yes. He stood to lose less by exposing himself through the fix than he would have done if the digital option had been triggered.
Q100 Steve Baker: What would you say to people if, say, five organisations or five individuals want to get on the telephone and auction, in a defined way, a lawful commodity? Why should they not just be able to do that? That is what we are talking about. This is an auction to a defined process among consenting adults with lawful goods. Why should they not just be able to get on with it? Why should there be any move at all to move them to a regulated exchange by force?
Rhona O’Connell: To an extent I think you have answered your own question in the sense that you have just referred to the Plunkett case. He bid the market up, then he came away again and then he bid it up again. I am sorry, other way round.
Q101 Steve Baker: Yes. Forgive the slightly loose language, but is it not possible always to have effectively fraudulent behaviour in markets? We do not abolish markets just because some participants are fraudulent.
Rhona O’Connell: No, indeed. Philosophically speaking, any free market is ultimately potentially open to manipulation.
Q102 Steve Baker: Is there not just too much concentrated corporate power in this particular market?
Rhona O’Connell: On the basis, as Mr Thomas was saying earlier, that there is a prestige and there is a certain amount of remuneration that comes through to the fixing members, then you could argue that that is the case, but it is important also to bear in mind that there is any number of market participants who will be taking part in the fixing by virtue of having their orders, not necessarily with the fixing members, but with other bullion banks who then reflect them into the fixing room.
Q103 Steve Baker: To what extent is the leasing of gold relevant to this process? Should it be made transparent? How much of the traded gold is leased?
Rhona O’Connell: That would be very difficult to do. It should be possible. It would take a lot of work. There is a degree of conspiracy theory to the effect that there is a lot of leased gold from central bank holdings and that we do not know where it has gone. There are arguments to the effect that, because gold turnover is so very much larger than the actual underlying amount of metal in the market, there is scope for manipulation on that basis, because it is being leased and nobody knows who has it. That does not carry that much merit, to be perfectly frank, because you can generally trace the path that the metal has taken.
Q104 Steve Baker: Since you mentioned central banks, my experience now is that there is very rarely enough competence anywhere to carry out the conspiracy, but it is a fact that central banks do hold large amounts of gold and do co-operate with one another internationally as well as acting on their own account. Do you think that the central banks are a relevant player in this particular market?
Rhona O’Connell: In terms of the amount of tonnage that they operate with, they are nowhere near as big as they used to be. In the 1990s, in particular, there were several series of transactions driven particularly by Belgium and the Netherlands, latterly Switzerland and the United Kingdom, where substantial amounts of tonnage were put through into the market. They are much smaller now. In fact, they are now net buyers as opposed to net sellers. Where they would stand in the market is to ensure liquidity. Generally speaking, if they are trading, it is because they are doing it for the diversification of their foreign exchanges and managing the fiscus.
Q105 Steve Baker: Just thinking about the tonnage, I understand that last year the Shanghai Gold Exchange saw a 2,100 tonnes go through and I also understand that, excluding Russia and China, the rest of world production went through China. Is the whole gold market just moving east now?
Rhona O’Connell: Yes.
Q106 Steve Baker: What does that mean for London and this whole conversation?
Rhona O’Connell: The Shanghai Gold Exchange has just announced they are going to launch a futures contract in, hopefully, September. There is scope for both that contract and for London and for COMEX and so forth. London is the centre of the market partly through history. It goes back to Moses Mocatta first trading in 1717, I think it was, or 1697, for diamonds. It is basically through history that London has been the centre. There was originally just the one fix at 11 o’clock in the morning. In 1968, that was changed to 10.30 am and 3 pm as the starting points. The rationale behind that is that London sits in the middle of the time zones, so London can feed the Far East at the start of the day and then feed into the United States later on.
Alberto Thomas: If I can add to that, you are absolutely right that gold trades as a currency. It trades throughout 24 hours a day across the globe, but liquidity varies greatly and the peak of liquidity is around the 3 pm London fix.
Q107 Steve Baker: It is a fascinating thing you just said there. I am sitting here with a one-tenth of an ounce gold Britannia in my hand and you just said gold still trades as a currency.
Alberto Thomas: It does.
Q108 Steve Baker: Can you explain what you meant by that?
Alberto Thomas: It is a currency. It is priced as a currency. It is seen by a lot of investors as the ultimate hedge against inflation or as a storage of value. It is a currency, and particularly one of which there is a limited amount around, where central banks cannot create any.
Q109 Steve Baker: The history of this benchmark seems to correspond fairly closely to the monetary history of the world since the end of the classical gold standard. To what extent do you think the situation in which we find ourselves is a legacy of the monetary role of gold?
Rhona O’Connell: I think the fact that we no longer have a gold standard has meant that there has been—shall we say?—scope for laxity in physical discipline because you do not gold backing currencies any longer.
Steve Baker: You are my new favourite in saying that. I am sure that is right. I think I probably want to leave it there, Mr Chairman.
Q110 John Mann: In your view, is there an imperative to sort out some of these markets, not least the gold and silver markets, before London loses out permanently and forever to China? In other words, should we see it as an imperative to be pressing and pushing for the kinds of things you have talked about, transparency and so on—let us call it generally sorting out the market and the reputation—or is it of no concern?
Alberto Thomas: I am happy to answer first.
Rhona O’Connell: Yes, of course.
Alberto Thomas: As I said, we work across Europe. We do work in the US. We work in Asia. I have to say that people are quite shocked when they look at London markets in general by the lack of action in trying to reform the benchmarks and fixing them. It is seen as a weakness. I think we are very often worried about the importance of the financial sector to the UK economy. We should be more worried about the possibility that we will lose the dominance of the financial sector in the UK if we do not make it transparent and suitable to today’s demands by the international investment community, otherwise they will bring in someone else.
Rhona O’Connell: Yes, I agree. The only slightly different complexion that I would put on it is that the rest of the gold market is, to a degree, fragmented. There is the Shanghai Futures Exchange, the Shanghai Gold Exchange, New Delhi, Mumbai, London obviously, New York. There are a lot of exchanges out there and 90% of the physical trade in the gold market is what is known as LOCO London. Essentially, London certainly has the lion’s share of the physical market without a shadow of a doubt, but that does not mean that we should be complacent about it.
Q111 Chair: Mr Thomas, in relation to total volumes, what proportion do you think, on the basis of what you know, are or may be rigged?
Alberto Thomas: In gold specifically, again I have to be a bit careful. I will give you a range. We think in terms of the number of days where we found evidence. We analysed it carefully between 2010 and the end of 2013. We think a range between 10% and 30% of the days and this is consistent with what we have seen in the foreign exchange markets and other markets as well, with a caveat, as I said: we only have the publicly available information.
Q112 Chair: This is not just an isolated odd case, now and again?
Alberto Thomas: Not at all.
Q113 Chair: We are talking about endemic rigging?
Alberto Thomas: The FCA finding and the fine—it is one of the cases that was captured by our models and it is exactly the type of behaviour that we predicted could happen.
Q114 Chair: Which is back to this point when you said there is the opportunity there, you cannot be sure it is there, but once the opportunity is there, can’t you be pretty confident it is there?
Alberto Thomas: That is correct.
Q115 Chair: Once there is an incentive to enable people to rig a market and get away with it, you are very likely, sooner or lately, to get a rigging of it.
Alberto Thomas: Absolutely.
Q116 Mr Love: Mr Thomas, I don’t know whether you were present for the first session—
Alberto Thomas: Yes, I was.
Mr Love: when Mr Bailey from the FCA mentioned a consultation being organised by the Royal Gold Council to devise a replacement. What do you think of that process?
Alberto Thomas: Again, I am not involved in that consultation, so I only have the same information you have. On the surface, what they asked for makes perfect sense and I think, on the surface, the current fixing mechanism satisfies some of those criteria. They need something that is not a reference price. They need a price that producers, central banks or whoever, the end users or investors in physical gold, can transact. They need an actual price at which to execute transactions, so that is one very good criterion. They ask for transparency. They ask for auditable benchmarks. These are all good criteria, and as I said before, they are all captured by the IOSCO principles and I do not think that anybody debates whether those requirements should be embraced or not. The question is how to do it.
I have seen the proposals around the reformation of the silver fixing and they do try to focus, it seems to me, on fixing one part of the problem, which is the transparency of the auction mechanism. That is definitely important and we need to be able to see what happens, because in gold we do not know if any transaction happens at all. We need to have that transparency, but there is a lot more to it, as I said. For example, the timing; there is no reason why we should wait for 10 minutes or an hour or an unlimited amount of time to get the pricing. We should be able to fix that and so there is a lot more than just getting transparency, I think.
Q117 Mr Love: To what extent is it motivated by competition with the London Bullion Market Association?
Alberto Thomas: There is competition with the London Bullion Market Association, but—
Q118 Mr Love: Have they stolen our lead on the London bullion market?
Alberto Thomas: I don’t think so. Eventually, the members of the London Bullion Market Association will have to agree to whichever proposals are put forward. I do not see it that way, personally.
Q119 Mr Love: You mentioned earlier that you thought regulatory scrutiny, bringing it into the regulatory ambit, might be the way forward. Do you think that the FCA and the regulators in this country have an obligation now to step in to sort this problem out? After all, there are quite a lot of people who think that the case that was taken recently is responsible, to some extent, for the difficulties now in the marketplace.
Alberto Thomas: There are two answers to your question. The first is that definitely there is a problem in the current system where the fix seems to be, “Let’s bring in front of the court one trader and he is responsible. It an isolated case and let’s be done with it.” I do not think that is the right approach. It is not an individual case. This is systemic failure. That is part of the answer, but I do not think, again, that regulation in itself, unless you give to the FCA incredibly more resources than they currently have, can investigate day by day the behaviour of every trader. That is just not possible.
The market has a very important role. We are seeing it, for example, in the high-frequency trading scandal, which I don’t think has hit the news probably here as much as in the US, but that has been going on as well and we have seen that the market has fixed itself in a way. The market has decided that they can create an exchange that cannot be tampered with, where high-frequency trading cannot take place, and this has been a market-own initiative. It has been a new regulated exchange that has emerged and through which a high-frequency trader cannot transact because their tactics do not work.
I think that the market should keep people accountable. It should be the shareholders of the bank and it should be the pension funds that pursue the dealers when they lose money. It should not just be the regulators with their stick going around and finding individual traders liable for it. That is a very important part of the problem and it needs to be enforced, absolutely, but it is a much more complicated problem. You need to try to find ways to introduce more competition.
Q120 Mr Love: But does the regulator have a role? In other words, should they be insisting that whatever the market comes up with, whatever the World Gold Council or the London Bullion Market Association comes up with, conforms to the IOSCO principles and delivers what would be public trust in that marketplace to have an objective gold fix?
Alberto Thomas: Absolutely.
Q121 Mr Love: Let me ask you just one final question. Given that we cannot expand the regulatory orbit around every single benchmark, do you think the most pragmatic solution to the problems in the gold market at the present time is a regulatory-led but market-imposed system that conforms to the IOSCO principles?
Alberto Thomas: Sorry, if that proposal is a solution to the problem?
Mr Love: It seems to me that the FCA was happy to allow the World Gold Council to go away and find a solution; then they will come back and comment on it. You seem to be satisfied with that, that the market needs to find a solution. What I was asking you was should there be a more activist role on behalf of the regulator to ensure that what the market comes up will meet with public concern over the—
Alberto Thomas: The job of the regulators is to enforce the fixing. I think the market will find a solution to the shortcomings of the current fixing. I do not think it is the job of the regulators to come up with a new methodology.
Q122 Mr Love: Ms O’Connell, do you want to add anything?
Rhona O’Connell: Thank you, yes. I agree with Mr Thomas in terms of the fact that the market will find its own solution. I did previously declare an interest, but I will iterate it. At Thomson Reuters, we are one of the organisations that are bidding for the administration of the silver fix, so I have a pretty good idea about what is happening there. If that process is a precursor to something similar for the gold fix then it should work out reasonably well because it is very much in the hands of the market.
The LBMA conducted a survey of 400 market participants and the short outcome of this is that the people who are bidding for the fix are essentially in consultation with the market right the way through, trying to put something together that is what the market wants and that is not just the bankers. It is the miners, the refiners, the traders, the people who consume and so on and so forth. The silver market administrators are certainly listening to the market as a whole, so that should serve as a precursor to a similar process for gold.
Q123 Mark Garnier: Mr Thomas, just on the wider issue of the benchmark fixings, I am interested in your thoughts on a couple of things. Somebody has passed me a report by a company called Autonomous that is saying potentially with the FX fixing, which is slightly different from what we are concentrating on here, you could end up with a fine pool of some $35 billion. Do you think that sounds about right?
Alberto Thomas: It is a very large market. I think it is probably ballpark right. Our own estimate is probably slightly lower than that, but—
Q124 Mark Garnier: Much lower?
Alberto Thomas: No.
Q125 Mark Garnier: Do you think the banks see this as a fine by reputation or do they see this simply as a cost of doing business?
Alberto Thomas: I think so, yes.
Mark Garnier: The latter, the cost of doing business?
Chair: Which of those?
Alberto Thomas: They see it as a cost of doing business. They already have set aside money for a lot of these litigation cases that are being tried around the world.
Q126 Mark Garnier: A number of us are on the Banking Standards Commission, around the table at the moment, and we are trying to drive better standards. From what you are suggesting, it sounds like fining banks is not going to drive better standards in banks.
Alberto Thomas: No, I agree. I do not think it will.
Q127 Mark Garnier: That is very depressing.
Alberto Thomas: I personally don’t think so. That is why I think the news is not very reassuring from where we stand because it seems to me that the solution is to fine the banks but the fines we have seen so far are not prohibitive. They are not big enough. If you compare the UK fines to the US fines, already there is a big discrepancy there. That is the first point. The second point, as I said, is the shareholders do not seem to reinforce any rights on the banks. Where are the shareholders? Are they putting any pressure on management to fix things? I do not see it. I don’t see the pension funds saying, “I buy shares in RBS and Barclays. What are you doing? Are you fixing this?” There is no pressure exercised in this country by big shareholders on to the institutions that are the culprits of this behaviour.
Q128 Mark Garnier: Of course, a shareholder will simply vote with their feet rather than doing anything, so they will—assuming the closing print of the equity price—
Alberto Thomas: But the financial community is very closely knit. That is the problem. There is not much competition and people see that suing the bank is something that is going to ruin the business, because everybody depends on banks at some point in time.
Q129 Mark Garnier: Ultimately, of course, it is the consumer—
Chair: We pay.
Mark Garnier: It is the depositor.
Alberto Thomas: Absolutely. They do not have any direct relationship with the bank.
Chair: Just on that point, when you told us about the scale of the rigging on a daily basis—sorry, have you completed your questions?
Q130 Mark Garnier: It might follow on from this. You would have heard that I made these assertions about the equity market and the fixes. Do you have any experience of that at all?
Alberto Thomas: To be honest, we have not yet looked at the equity markets. I have heard about the types of areas you mentioned in the previous session.
Q131 Mark Garnier: That is not a surprise to you?
Alberto Thomas: No, it is not and I think, in due course, we will probably plunge in there, but there are higher priorities and there are other benchmarks or other prices that are much easier to manipulate than equity prices.
Q132 Mark Garnier: Of course, you are absolutely right. The equity element tends to be the smaller caps or the mid caps, as a result of which it is where there is less value to that market. None the less, would you agree that they would be doing it in terms of the incentive would be to try and get better valuations on stops when it could come in terms of when you have bonus valuations for the start or, alternative, it could be where you have derivatives maturity dates so you could have a result of helping on that type of thing? Can you think of any—
Alberto Thomas: There are possibilities and I think, ultimately, it comes down to the same. There seem to be, across all the benchmarks we looked at, a potential for conflict of interest and that could lead to front-running. It could lead to manipulation or lots of this behaviour. The explanations you gave are plausible. I do not know any specific cases that we need to look at yet, but I think that possibly it is another area that deserves further attention.
Q133 Chair: I just wanted to clarify. You gave us an estimate about the number of days that might be rigged. Are you able to give us any idea, on the basis of what you know, about the scale of the deviation from what would be the unrigged price?
Alberto Thomas: In terms of the amount by which—
Chair: Yes. In other words, by how much? We have been discussing—
Alberto Thomas: It varies by days.
Q134 Chair: Customers are ultimately paying for this. By how much are they being ripped off?
Alberto Thomas: It depends. It varies day by day and I would say—
Q135 Chair: We are looking for an average of the variation as a percentage.
Alberto Thomas: Up to $4 per ounce.
Q136 Chair: On a base of when it is varied up to 1,000 or—
Alberto Thomas: Yes, correct.
Q137 Chair: What are we talking about here? Are we talking about—
Alberto Thomas: 0.5% or 0.4%.
Q138 Chair: We are talking about 0.5%?
Alberto Thomas: Up to. That is the maximum. That is what we have seen.
Mark Garnier: About five basis points.
Chair: Or five basis points.
Alberto Thomas: No, it is five—
Chair: 50 basis points.
Alberto Thomas: 40, yes.
Chair: 40 to 50 basis points, okay.
Alberto Thomas: As a maximum. As you have seen on LIBOR, it varied.
Q139 Chair: You have both told us a pretty depressing story, haven’t you? You have told us the market has the potential to be rigged; that it is not remotely competitive enough; that where the potential exists for it to be rigged it is likely to be; that 10% to 30% of the days are likely to have been rigged; that up to 40 basis points may have been extracted at customers’ expense; and that regulators could and should have acted earlier, particularly after LIBOR. Is there any reason, looking at this, that we should not be concluding that this is an appalling story?
Rhona O’Connell: I am not talking the corporate book here, I assure you, but there is a page on Reuters that carries the trying price in the fix and that is real time. Anybody who has a Reuters’ screen—I do not know whether Bloomberg does it as well—will be able to see the price that is being tried during the fix. There is transparency there.
Chair: You have opened another whole set of issues about the relationship between that and the fixes, but we are not going to go there now. Thank you very much for giving evidence this afternoon. I am sorry that the hearing was interrupted. It was extremely interesting to have your evidence.
Oral evidence: Manipulation of Benchmarks, HC 491. 12