Satyajit Das is a leading authority in the field of financial derivatives and treasury management. He regularly leads seminars and teaching forums in Europe, North America, Asia and Australia, and acts as a consultant to a number of financial institutions and corporations from his office in Sydney, Australia.

He has written several seminal works on derivatives including Swap Financing (1989) and Swaps and Financial Derivatives: The Global Reference to Products, Pricing Applications and Markets (1994). This year he'll publish two books on exotic options and structured notes.

Between 1988 and 1994 Das was the treasurer of the TNT Group of Australia, where he had responsibility for corporate finance and risk management. For 10 years before that he worked as a banker for Commonwealth Bank of Australia, Citicorp Investment Bank and Merrill Lynch Capital Markets. In 1987 he was Visiting Fellow at the Centre in Money, Banking and Finance at Macquarie University.

This interview was conducted in May with managing editor Simon Boughey.

Derivatives Strategy: Has the profitability of the derivatives industry been overestimated and are some banks overinvesting?

Satyajit Das: Well, you have to look at it in the context of the financial service industry generally. What happens is that derivatives are perceived-rightly or wrongly-as very profitable, and so individual banks do the numbers and say to themselves, "There are huge margins in the industry and we will come in." What they fail to anticipate is that 50 other banks from around the world have also spotted the same opportunity simultaneously, so there is a tremendous flood of supply into the market sectors and a vicious cycle starts up. There is a tremendous bidding for people and salaries escalate. So your cost structure goes up at the same time that competition squeezes the revenue. That is endemic in the financial industry overall, but in derivatives it has been particularly evident. In the early 1990s people looked at loan margins and concluded that derivatives offered much more interesting returns, relative to returns on capital and loans. So they overinvested. Take very long dated swaps for example. The margins are between 3 and 5 basis points, and it is hard to believe that these sorts of transactions are enormously profitable to the participant.

DS: And yet at the same time we have various banks still spending large amounts of capital on hiring derivatives teams. Do you think, for example, that Deutsche Bank or CIBC will see a return on their investment?

Das: I don't know whether they or any particular person will. It seems to me that the way you make money in this business depends on the market you are in. If you are in a pure commodity product, like interest rate swaps, then the real sources of profitability become cost control. You try to build good client flows by staying very tight on bid­offers and having a very strong distribution system, and then you try to lower your cost and take very little market risk.

The second element of profitability is the value-added stuff, which chiefly derives from innovation. In the last few years a great amount of money and effort has been spent trying to understand clients and their needs better. For example, any company with multiple exposures is really trying to manage the net cash flow. So banks are structuring very elaborate instruments, like multifactor options where the option writer only pays out under certain circumstances. And that is where some of the best margins are, because the pricing is not as transparent.

The last element of profitability is trading revenue. There is no doubt that a significant amount of the earning is coming from proprietary trading and that is now a global trend.

DS: So you are not prepared to say whether you think Deutsche Bank, for example, is barking up the wrong tree?

Das: I don't know; if I were Deutsche Bank or somebody of that stature, I would have a client base I could leverage. I also would have a strong retail base in Europe. And under those circumstances, if I had the right strategy, there should be no reason why I would not be able to recover my investment. But you have to have the right strategy; it is not just the investment.

DS: To help achieve that right strategy banks have spent a great deal of money on systems. Has it been worth it?

Das: I was reading about JP Morgan having spent a billion dollars in technology last year. I am sure that not all of that was in trading systems etc., but that is a vast commitment to make. The banks are asking, "How do we get the maximum bang for our buck?" At the same time technology is not seen as a fashionable area, so you don't pull your best trader to sit for days with a technologist explaining exactly what is required. Consequently, people are so busy managing crises on a day-to-day basis that I am not convinced the systems that they are developing are necessarily the systems they want or need.

DS: Why has it been so difficult for banks to get the right blend of technology for the right price?

Das: There are two functions to which information technology needs to be applied. One is obviously the pricing; the second, in the broadest possible sense, is risk management. But most people still operate in what I would call a mixed system environment, which employs different systems for different products. They are not designed to look at risks or the transactions or the cash flows in a single way, so people are now trying to merge all this information into a unified risk management framework.

Conceptually everybody understands a need for this, but there are real practical difficulties. If I sit with a portfolio of about 10,000 transactions, how do I port that into the new system? If I have 20 operations around the world, how do I now migrate 20 separate dealing rooms onto the same system? It is difficult to achieve this quickly and effectively in a very fast-moving environment.

DS: While we're talking about the efficient use of capital, do you think that the proliferation of triple-A vehicles is a waste of money?

Das: I think you have to ask yourself the question of why they set up the vehicles. When Merrill Lynch set up MLDP, the logic was very straightforward. They were in the bond business and they found that a lot of their business involved derivatives. Since at that time they were a single-A credit, they found it very difficult to win that business, because the sovereign minimum threshold was double-A. Merrill solved that problem with MLDP. Now once that happened there was an uptiering of the credit that certain kinds of counterparties came to expect, and then other people set up all these vehicles. Some of these vehicles have not done a lot of business, some have. So it is hard to see that all of them are necessary. I have always argued that you do not need a special-purpose derivative vehicle unless you have the flow of transactions to justify it. You could always use a third-party guarantee; cash collateral does just as well. But some of the early vehicles, like MLDP and SWAPCO, seem to have built up sufficient volume; they publish their annual reports and they have certainly been very successful.

DS: I was thinking more in terms of Westpac, Tokai or Sumitomo. What about these?

Das: I think you have to go back to the fundamental premise that you have to have a very strong derivatives business underlying it. If you have that, it makes perfectly good sense. If you do not have that, the SPV [special-purpose vehicle] will not magically catapult you into the first category. And there is not unanimous support for special-purpose derivatives vehicles, because some users would prefer a triple-A bank to a triple-A SPV. I think that we will see more of the type of DKB/MLDP piggy-backing that was announced in February.

DS: As profits get squeezed, will we see great consolidation in the derivatives industry?

Das: By the year 2000 the industry will almost certainly consolidate. There will be between five and 10 major players who are truly global in that they cover more than one asset class in 10­12 currencies. I think that you will get a couple of Europeans, one or two Americans-probably one investment bank and one commercial bank-and one or two non-European, non-Americans. I think that you are going to see some fairly major shakeouts, and on top of that you will see some fringe players who will almost be boutiques. They will do Australian dollars only, or they will do equities only, or some sort of differentiation, but I think that the market will consolidate in the next five years, which is logical. The financial services industry generally has overcapacity in it and derivatives are by no means immune from those pressures.

DS: What is happening on the end-user side of things? Are they losing their fear of derivatives generated by the horror stories of P&G, Gibson Greetings and Orange County?

Das: To be frank, I don't think most treasuries ever stopped using derivatives. Nobody stopped using FX forwards, which are technically derivatives. Directors' responsibilities don't flow only one way. Not only will they have problems if they use derivatives and make losses, but they will also have difficulties in defending their actions if instruments were available to manage risks and they did not choose to use them.

I think what has emerged out of the various court cases is that there were fairly aggressive treasuries who took quite interesting, very market-sensitive positions. I think that there is an increasing realization in the corporate community that derivatives themselves are not the problem, it is how you use them that is the problem. So I think the first thing that we have seen is a fairly clear trend toward redefinition of the conditions under which you can use derivatives.

Secondly, I think generally that interest in risk management will continue to grow. The real problem is that most people miss the fact that as you internationalize your business, you are exposing it to international variables like currencies and interest rates. So most corporations of any size are realizing they cannot avoid playing this game, but they know they need to do it with the right tool sets.

DS: Let's talk a little about suitability. How are banks to know if a trade is suitable for a corporate's needs?

Das: From a banker's point of view, the real difficulty is when you have clients who you think you know, who come up and ask you for a price. They do not say, "I want this price because I want to do this or that and this is the overall picture." They say, "We want a price." So you give them the price on that particular structure and the deal is done.

Now it is very difficult to say whether this trade was suitable later on. Over the last 12 months the banks, as a result of things like Proctor & Gamble, have become concerned about this process, and several banks have taken the attitude that for complex, particularly leveraged types of transactions, they will only deal on the basis of the approval of the full board of the corporation.

DS: Would you say that U.S. end-users are generally more sophisticated in their use of derivatives than end-users in other parts of the world?

Das: The pattern of end-user activity around the world is very different. I think that it depends on what you are exposed to. For instance, in my experience, European and often Asian users tend to be very sophisticated about currency derivatives, simply because most of their flows are cross-border. In contrast, in the U.S., the sophistication may well be in other areas, like interest rate risk management. So to say that there is one group that is better at any one particular element is a bit misleading. You tend to get good at what you are forced to do a lot of the time.

DS: Is there always going to be something of a learning gap between the products that are being structured in the banks and the ability of end-users to understand them, as the former's rewards will nearly always attract the most gifted mathematicians etc.?

Das: That is certainly true and that is one of the biggest problems. Having worked in corporate life, I know the rewards there are very different. One of the big issues for corporates is how you set aside a remuneration structure for people in functional specializations that are closer to banking, which may result in your treasurer being paid more than your CEO. Some of the insurance companies that have expanded into financial products have had difficulties, as the remuneration level that they have had to pay has disturbed their, how shall I say, mainstream remuneration expectations. Even in the financial services industry itself, certain people have almost quarantined their derivatives operations in separated vehicles so as to isolate any impact they might have on remuneration practices throughout the institution. So it is a common problem.

DS: Are bankers beginning to look differently at derivatives and the way these instruments fit into their corporate structure?

Das: Yes. What people are realizing is that derivatives are another way of trading the asset. They are in fact an aspect of the asset class, and what we have seen in the last couple of years are very concerted efforts to merge the physical trading of the asset and the derivative trading of the asset class. The most recent one that I can think of is the announcement by UBS that they were going to merge debt and debt derivatives. JP Morgan and BT did this in some areas a few years ago.

DS: What effect will this new perception have on the way that derivatives markets operate, say, in the next five years?

Das: I think that it will change it in several ways. One is that the organization of derivatives desks will change very dramatically. At the moment they are still, to some degree, organized around products. In fact, once you strip every instrument down into almost basic elements, what you get is risk, whether it is interest rate, currency, commodity or equity. So what people will probably do is not have traders of products, but of risks. For example, rather than a trader of interest rate swaps, you would get a curve trader who trades essentially the whole yield curve. The trader would trade interest rate risk in a particular currency, irrespective of what product generates the risk. It won't matter if it is an FRA, a swap, an FX forward, a bond on structured products, etc.

DS: And that spread trader would trade swaps and treasury bonds, for example?

Das: Not necessarily. He would actually price the interest rate risk component, and I think you would have a layer sitting between him and the client. The salesperson may combine sales and structuring. He or she would work with the client to identify the risk to be hedged or the risk profile sought. The salesperson would engineer the appropriate product, whether it is plain vanilla or customized. Whatever the eventual product dealt, the bank (and the dealer) would see it as an amalgam of risks to be mentioned within the new framework. The risk trader would not necessarily see the product; he would see it only as a marginal cash flow position that affects the trader's risk position. And so there would be a complete change in the way that the derivatives market sits and operates. To some extent, you will find a considerable uniformity in how products are looked at. Frankly I think that this is as important a change as derivatives themselves have been in capital markets. People are really struggling to make this transition, which is a very difficult transition to make because first you have to change culturally.

DS: So you are going to have to reorganize the whole way in which you trade and how trading is recorded?

Das: Correct. And this is the reason why you're seeing things like return adjusted performance measurement (RAPM) or any of the other acronyms that have sprung up. To some extent the BIS market risk guidelines force you down this path of risk decomposition, whereby all products are broken down into fairly fundamental aspects of risk. The problem arises because although we are moving towards a risk decomposition approach, we still trade products, or bundles of risk. One answer to this is to have a two-tier system in which traders trade products and risk managers manage the decomposed risk. Alternatively, traders may begin to look more like risk managers themselves and functions now separate will combine. In this way an interest rate swap becomes simply a trade which affects his overall interest rate position, and a currency swap becomes a trade which affects his currency position and his interest rate positions in the two currencies. I can foresee a time when spot trading and currency swap trading are combined into a single function which looks at currency risk irrespective of where it is generated.

DS: What new products are likely to be developed in the near future?

Das: Probably the most important development I have seen in the last two or three years is the move towards credit derivatives. It makes terrific sense because credit is another attribute of an asset that we would like to trade separately. It has expanded very rapidly, and my tip is that within the next two or three years, most institutions will have credit as a separate asset class, alongside fixed income, equity, FX and commodities. The correlation between that and the other asset class performances are not necessarily perfect, so by definition they should get the benefit of diversification.

To date that market has been driven by people realigning their exposures, as when the Japanese banking crisis came upon us. A lot of people who thought they were overexposed to the Japanese banking system wanted to reduce their risk and found they could not get out of transactions, so they tried to find alternative ways of hedging. That was one phase of credit derivatives, but I think that we are going to see several other phases. I am very excited because the volume of bond issuance is over $1 trillion a year. There's a lot of credit risk in the banking system that has to be moved around. Giving it liquidity will dramatically change the way people look at credit. The credit department of the future will look very different. In fact the head derivatives traders will be moved down to the credit department, so if anyone wants to do a deal, they will ring up the credit department, and the credit department will make them a price selling them a credit put. Essentially they'll have a choice of doing the deal or buying the credit put from somewhere else if they can. The credit department itself can trade that risk with other banks in an interbank credit market. While it sounds far-fetched, my belief is that it is not going to be that far away.

DS: What else is ahead?

Das: We are almost certainly going to see a very major market in real estate derivatives, because if you think about all the requirements for a derivative market in terms of high transaction costs and terribly illiquid assets, all of those are satisfied. People are looking at ways of effectively liquifying their portfolios using real estate derivatives. This will grow enormously.

There is also emerging markets. Some of the commercial banks are taking the technology from developed markets and putting them into emerging markets; the margins are better and the risks are greater in terms of liquidity etc. And we have seen quite a lot of expansion; five years ago nobody thought that you could do five-year cross-currency swaps in Thai bahts and Indonesian rupiahs, but you can now.

The last area of growth, which I think is very important, is retail derivatives. Interestingly, in places like the U.K., you have mortgages with almost every sort of derivative component-you can cap it, you can fix it, you can swap it back into floating. This is going to become very important.