On Incentives

"Never penalize those who work for us for mistakes or reward them for being right about markets. It will go to their heads, is counterproductive and, in any event, material compensation will not correlate with their ability to predict the future next time."

Site Search

Writings & Speeches

Testimony of Eugene H. Rotberg
Before the
U.S. House of Representatives
Committee on Banking, Finance and Urban Affairs

Thank you for inviting me to present testimony at this hearing. Regretably, I am unable to attend the hearing because of a previous commitment outside Washington, D.C. In my absence, I request that the statement below be incorporated in the record of these proceedings.

Let me quote from selected parts of my testimony given to your committee over four years ago on June 23, 1994. Though the subject was different, “derivatives”, and today it is “Hedge Funds”, the problem is the same - leverage, the use of collateral and the availability of information.

“My name is Gene Rotberg. Let me first express my appreciation for being asked to testify with respect to matters dealing with the derivatives markets. I ask to be incorporated into this record remarks I gave recently to the National Association of Corporate Treasurers entitled, “The Only Perfect Hedge is in a Japanese Garden.”

A lot has already been written and reported about derivatives: a minority staff report from this committee, Congressional hearings, a GAO study, a Group of Thirty report and commentaries by virtually every accounting, banking and securities association. There have been press reports of losses by dealers and corporations, lawsuits, investigations and attention by every relevant regulatory agency. For purposes here, let me try to focus on why the subject matter has and will likely cause a great deal of continuing stress. I believe it is a peculiar combination of five unique and potentially dangerous circumstances.

First derivatives can be used to leverage risk— interest rate, currency rate, share prices— without putting up a lot of money. That simply means that during a period of volatility, losses or gains are magnified manyfold. And often the leverage is asymmetrical; that is, the potential gains are limited, while the losses may be multiples of the maximum gain. (emphasis added)

Second, current accounting conventions mask error, risk and mistake. They are not designed as risk management tools. They have tax consequences, which may be one of the reasons why it has been so difficult to develop a comprehensive set of conventions which also can be used for risk management purposes.

The truth is we do not, generally, mark derivatives to market. Many derivatives are unmarkable. In certain transactions, mistakes can be hidden because accounting conventions do not record them, either because they are ad hoc or there is no market, or they are off balance sheet. There is, too often, little reality testing. We continue to pretend that a rolling loan gathers no loss. We pretend that if a triggering event occurs in a different time period, the loss can be delayed. And when losses can be ignored, greater risks are taken. The latest FASB proposed draft on derivative accounting is a beginning, but the draft is deficient because it will not, yet, put the users under the pressure involuntarily of admitting to failure, risk and error. I think the response to the latest FASB draft will illustrate the point.

Third, senior managers are rarely as informed as traders, and legislation is not likely to make them so. Typically, senior management is usually unaware of the technical operations of financial engineering. Worse, they are often afraid to ask, out of concern of admitting to their lack of mastery over the subject matters, and I think we also must admit to the fact that there is a good deal of underlying hostility to financial superstars, mathematicians, physicists. Senior management often believes the financial engineers are too young; too overpaid, they have too much control; they are too smart; they know what to hide and, too often, how to hide what they are doing and why they are doing it. Management is not trained in the intricacies of convexity or volatility. As a result, reports are inadequate, supervision thin. Risk management leaves a lot to be desired. Worse, most of us have great difficulty in admitting to those who report to us that we do not know nearly as much as they. That is a recipe for potential disaster.

Fourth, many products, particularly over-the-counter derivatives and aspects of the mortgaged-backed market are idiosyncratic, ad hoc, unpublicized, illiquid. That means they are difficult, if not impossible, to price or value. It means that if held as collateral, there may be no buyers in the event of a forced sale, or the spreads between buyers and sellers may be so wide that even hedges are ineffective.  That means that a bank dealer which holds such instruments may have to sell short instead, say, plain vanilla U.S. Government bonds in very large amounts to protect itself. That complicates the Federal Reserve responsibilities.  (emphasis added)

*    *   *  

With respect to H.R. 4503, I have the following suggestions:

  1. Page 6, line 23. Add margin requirements as a subject to be covered. (emphasis added) 
  2. Page 7, line 21. The language is quite soft. You might wish to consider providing the regulatory authorities the “power to implement” regulations, not merely make recommendations (to whom?) with respect to the subject matters on Pages 8 and 9
  3. Pages 10 and 11. All references to “revenue gains and losses” might explicitly include the phrase, “whether or not realized.”
  4. Pages 29 and 30. The term “speculation” should be avoided in the Bill. Any financial decision is “speculative.” Alternatively, you might wish to consider using the term “leveraged” where appropriate.
  5. Finally, I would suggest that the study of margin and collateral be the responsibility of the Federal Reserve Board.  (emphasis added)

*    *    *

This brings me to my final point and, to my own mind, the most important. We have enough essays, surveys, studies, green books, Basle guidelines, international studies about credit risk, basis risk, legal risk, event risk, operational risk. They are all fine and so will be future ones— whether mandated by legislation or done voluntarily. But they all read like a cross between graduate school theses, at best, and a public policy consultant’s think-piece. We are writing essays without really knowing, in a systematic fashion, how the market works.  We need far more precise day-to-day market information on who does what; how is it financed; how do bankers and dealers pass on their risks; how is leverage actually accomplished, etc.   (emphasis added)

The time is now, I suggest, for a “Special Study,” under oath, with subpoena power, conducted independently, reporting directly to Congress, with such commentary by the Federal Reserve, Treasury, Comptroller of the Currency, SEC, CFTC, and anyone else who would like to comment on the ultimate analyses and conclusions of the Study.

The Chairman of the SEC and Chairwoman-Designate of the CFTC should designate a Director of the Study and then let that Director staff the Study – with subpoena power. We will not find out how the market really works without such a Study. Why subpoena power? Let me remind this Committee that at the time of the Salomon Brothers affair almost three years ago, which had many of the elements of the subject now being looked at, no securities firm would voluntarily testify about their operations in the REPO and government securities market. Nor will they do so fully and frankly about derivatives in response to a letter from the Secretary of Treasury or the Chairman of the Federal Reserve.

They will under oath. And that is the way to develop a body of knowledge in this particular area. The alternative is to rely on Grand Juries, SEC investigations after the fact, class action lawsuits and surveys.

Three years ago, in Senate hearings on the operations of the government securities market in connection with the Salomon Brothers affair, I testified:

 “Finally, I would urge a major inquiry— not an adversarial investigation— into the operations of the securities markets (including the government securities markets and those of derivative products and financing) similar to the Special Study of Securities Markets conducted in the early 1960s which reported directly to Congress.”

I can only repeat the same recommendation here, but this time note, merely by way of example, five matters, almost chosen at random, which have not yet really been publicized, and which are indicative of what we don’t know about— except in the most superficial and uncoordinated fashion.

*    *    *

  1. The effects of illiquid collateral, particularly in the mortgaged-backed market, and its effect on the U.S. government bond market when small changes in interest rates are magnified when the collateral can’t be sold and, instead, the U.S. government bond market absorbs the selling pressure as financial intermediaries seek to protect themselves.
  2. Equity swap positions of banks. To what extent are banks, through the use of derivative products, taking substantial positions in the stock markets domestically and/or in foreign stock markets with the explicit currency risk?
  3. The practice and implications of end-of-month or quarterly cleaning up of derivative portfolios in order to avoid disclosure.
  4. The use of derivatives in the FOREX market and its implications for public policy, government intervention and the maintenance of stable exchange rates. These matters get too close to the edge of propriety or legality to expect voluntary disclosure to form letters.

Does this all mean that there is great systemic risk? No. Or that major banks or corporations are likely to tumble in a domino effect? No. Will some be badly hurt? Yes. Are some S&Ls, securities dealers and corporations taking imprudent risks? Yes. It means mostly, though, that regulators are not up to date because they do not have up-to-date quality information about what is really going on in the market— and when they do get it, it is after the fact, ad hoc, in a criminal investigatory setting, which rarely predicts the next financial crisis.

*    *    *

I also incorporated in the record of June 23, 1994 hearings the main risks that were on the horizon. Let me quote again:

  1. Liquidity Risk. You think you are precisely hedged, but the product is so esoteric and idiosyncratic that you cannot sell it because there is simply no market for the product. This, typically, will happen on the asset side, where you may want to either capture a profit or minimize a loss or sell collateral, and you can find no buyers. This is typical in the OTC derivative market or parts of the mortgage-backed securities market.
  2. Credit Risk. Your counterparty has lost money and fails. You were on the right side of the market, unfortunately, your counterparty was on the wrong side. Or, your counter party would ordinarily be just fine, but its counterparties, strangers to you, default.
  3. Legal Risk. The laws in Asia and Western Europe are not nearly as clear as those in the United States (and even these are not without doubt). You believe that you are totally netted with a particular counterparty; that you had a net zero position and, in the event of default and bankruptcy, you would be protected. It turns out that the netting rules outside the United States are ambiguous, and you may have to get in line with other creditors or depositors.
  4. Event Risk. A war takes place; an earthquake occurs; a flood of a magnitude not seen in a hundred years washes over the land; a cartel falls apart; oil prices quadruple; tax laws change, and the market in which you were speculating, or even hedged, moves in a magnitude not only unforeseen, but totally outside past models. They always do. You are in trouble.
  5. Basis Risk. You thought you were hedged. You believed that investment A hedged instrument B. You were long in one, short in the other. They, in fact, moved in the same direction. The three-year Treasury note in which you were long deteriorated in price, but unhappily, the five-year note, in which you had a short position, increased in price. You lost both ways. Again, the only perfect hedge is in a Japanese garden.
  6. Leverage Risk.  You are so leveraged that even a small market movement will prompt a margin call or liquidation. The security which is out of line will move back to its normal position on the yield curve, but someone out there, for one reason or another, has chosen to put pressure on a particular coupon, a particular security, at a particular point on the yield curve, and while over the next week or two it will surely come back into line, in the meantime, you must liquidate. Then your loss becomes visible.
  7. Operational Risk. Back-office systems, yours or someone else’s, fall apart; credit monitoring systems break down; documentation is flawed; transcription and recording mistakes are made; settlements are delayed; systems do not capture fully the nature of the transaction— the computer program doesn’t yet cover that kind of transaction (they are working on it). And, it is all quite expensive to put in place and keep it up to date. And, most important, there is no natural constituency to support the financial and resource expenditures that are needed, particularly if you are not supposed to be a profit center and are trying to keep quiet the risks you are taking.”

Little has changed from that testimony. I can only repeat, yet again, my primary recommendation: An independent study, authorized by and reporting to Congress should address the forgoing matters, conducted with subpoena power. It should not be an essay, survey or boiler plate description of the “market.” The truth is there is very little expertise within the federal regulatory supervisory agencies as to precisely how the financing and leveraging is actually done in practice. That must be determined by a dedicated group with full access to the most sophisticated and active players in the market. Then, after the facts and operational practices and procedures are fully understood, the study can then address some very difficult policy issues, e.g., 1) how to monitor or constrain activities in a global cross border market with different (or non-existent) regulatory structures; 2) how to share information about credit exposure without running afoul of anti-trust, privacy or competitive considerations; and 3) how to monitor and constrain “leverage.”

But these difficult issues cannot be addressed until there is first a more thorough understanding as to precisely how the market works in practice., what are the incentives and who are the players.

There is little doubt that despite the rhetoric, the SEC, the Federal Reserve Board, the Comptroller of the Currency, the CFTC, the self regulatory agencies— all have different responsibilities and agendas. That is but one of the reasons why the study should be independent from any particular agency or entity and should report its findings directly to Congress with such commentary from the interested government agencies as they wish to make. The last time this was done, as I noted above, was during the Special Study of the Securities Markets in the early 1960s. It is time for another Special Study.