Robert C. Merton - Managing Financial Institutions in the 21st C - Sloan School MIT 1996

[MUSIC PLAYING]

URBAN: Good evening. I'm Glen Urban, Dean of the MIT Sloan School, and it's my pleasure to welcome you to this inaugural session of the MIT Sloan School Oxford University Distinguished Lecture Series. This is a real partnership between the Oxford University Press and the Sloan School. We're trying to bring distinguished academic speakers here to campus and produce high-impact books based on their lectures. I think our goal here is to really bring leading-edge thinking about the future of business in front of you in terms of in-person things in this series and then to our audience in written form.

Before I introduce Bob, though, I would like to let Ed Barry who is the president of Oxford University Press in the US, have a chance to say just a couple words about the purposes and history of this series. And then, I will finish with the introduction.

BARRY: Thank you very much, Glen, and thank all of you for attending this inaugural lecture in this absolutely beautiful auditorium. We of course hope that the lectures jointly sponsored by Oxford and Sloan will become a major forum addressing fundamental issues in the business world.

Now, I have two reasons for being delighted to be here tonight. The first one is professional, and the second one is personal. As president of Oxford University Press, we committed ourselves about 10 years ago to trying to publish the very best advanced-level business books in the United States, and I think we're really making some good strides-- partly because so many of our authors are being drawn from the Sloan School.

So therefore, on the professional side, I'm really delighted that we're formalizing this relationship between Sloan and Oxford University Press and that we're embarking upon this very exciting venture.

I should say related to that is another reason that we're a little bit proud. It was about 500 years ago that Oxford University Press published its first book, taking advantage of the invention of the printing press and taking advantage of technology. So it seems particularly appropriate that the Press would join with a university whose very name includes the word technology-- and particularly nice to see us doing this at a time when technology is becoming increasingly important in the world.

On a personal level, I've got a special reason for being happy. It was probably 25 years ago-- before my hair turned white-- that I was president of the Free Press, and I was publishing with a great deal of pride the most distinguished sociologist in the United States, Robert K. Merton.

Now, sometime after that, not quite 25 years ago-- a little less than that going back from here-- we were chasing after a younger Robert Merton, Robert C. Merton, trying to convince him that he should sign a book with the Free Press at that time. Well, he didn't get time to do that book for the Free Press, but now we have captured him with the lecture and a book that will follow.

However, the actual introduction here is going to be by Glen, who has known the son almost as long as I have known the father. Glen.

[APPLAUSE]

URBAN: We are indeed fortunate to have Professor Bob Merton here to give this inaugural address. Bob and I do go back a long way. I came here to MIT in 1966, and he was starting his PhD about the same time and finished a PhD with Paul Samuelson in 1970. And then, Bob was on our faculty here at the Sloan School for 18 years. During that time, he and Myron Scholes and Fischer Black invented option theory, and then Bob went on singularly himself to revolutionize the field with continuous time modeling in finance.

Bob is now teaching at Harvard. We think of that as a short-term loan--

[LAUGHTER]

There. But as well, this year he's on leave at a very successful firm that he co-founded called Long-Term Capital. And those of you who know the market know how hot that company is. So Bob is one of these unique individuals who can contribute on the academic level as well as create an extremely successful business.

So it's a real pleasure to welcome Bob to give this first lecture called "The Management of Financial Institutions into the 21st Century." And please do mark your calendars because he'll give two more, April 24th and May 1st, which will complete the trilogy. So Bob, welcome.

[APPLAUSE]

MERTON: Thank you, Glen and Ed. And thank you for coming and inviting me. I'm certainly deeply honored to be here for the inaugural of the Oxford University MIT Sloan Lecture in Business and all the more so-- as Glen has told you-- because I had the great good fortune to spend nearly two decades here at MIT with a wonderful set of colleagues, classmates earlier, and then colleagues on the faculty.

My overall topic for the three lectures is the management of financial service firms into the 21st century. For any business, it is, of course, primal to establish what purpose it will serve before analyzing its organizational structure and management design. And so, tonight, I explore the range of innovative financial products and services that are likely to be offered to households, business firms, corporations, and governments and the influence that these new financial products may have on the economic activities of those customers of the financial institutions.

External customer demands thus identified, I will in the second lecture next week look internally into the financial institution to focus on risk management as the central managerial component necessary in a first-rate, high-quality production process for financial products and services of the future.

In the third lecture, I try my hand at examining the complexities of long-term strategic management of the financial institution in an environment of rapidly changing geopolitical, regulatory, and institutional boundaries and driven by technological and financial innovation. In doing so, I will offer a functional perspective as a means for analyzing the dynamics of institutional change within the financial system and thereby provide a conceptual framework for strategic analysis of the individual financial firm.

As we move through the lectures, you will undoubtedly detect a managerial focus on how things should work instead of on how things may not work. I am not inherently an optimist. I do, however, believe that, in the study of the financial system-- as in the study of the human body-- physiology precedes pathology. I thus reserve for portions of my last lecture the discussion of management and financial pathology. I do so in the last lecture not because pathology is uninteresting, but precisely because it is too interesting.

The manifest objective of these lectures is to analyze the management of financial institutions. Nevertheless, I harbor the hope that some will see a more latent one-- namely, to offer the financial services industry as a possible strategic research site that may cast some light on the management issues of other globally-competitive industries which operate on a large scale and experience rapid and continuing innovation.

I underscored at the outset that the forecasts expressed here are not necessarily uniformly shared among my colleague financial economists. As we all know too well, the historical standard errors in forecasting of the economic future are notoriously large. Thus, the practitioner and researcher alike should therefore approach the ones given here tentatively and assess their limitations carefully in any application.

New financial products and market designs, improved computer telecommunications technologies, and advances in the theory of finance during the past quarter century have led to dramatic and rapid changes in the structure of global financial markets and institutions. The scientific breakthroughs in financial modeling in this same period, some made right here at Sloan-- most notably Fischer Black and Myron Scholes, the Black-Scholes model-- these financial modeling breakthroughs both shaped and were shaped by the extraordinary flow of financial innovation which coincided with those changes.

Now, to put this in perspective, I remind you that the fall of Brenton Woods leading to floating exchange rates, the development of the national mortgage market in the United States, the first oil shock, and the creation of the first listed options exchange accompanied publication of the famous Black-Scholes option pricing model in 1973.

ERISA and the subsequent development of the US pension fund industry, as well as the first money fund with check writing, both occurred in 1974. Now accounts began in 1976. And 25 years ago, total assets of all of US mutual funds were $48 billion. Today, they are more than 50 times larger with one complex alone, Fidelity, managing more than $400 billion in assets. In the same period, average daily trading volume on the New York Stock Exchange grew from 12 million to nearly 300 million.

Now, the cumulative impact has significantly affected all of us as users, producers, or overseers of the financial system. Innovation is a central force driving the financial system towards greater economic efficiency, and considerable economic benefit has accrued from the changes of the past 25 years. Moreover, there has been vast improvements in our understanding of how to use the new financial technologies to manage risk.

Despite all of this, we find today an intense concern among managers, regulators, politicians, the press, and the public over the new activities and risks of financial institutions relative to their traditional risks and activities, such as commercial and real estate loans or LDC debt.

Now, why? Why this concern despite the gains and benefits? Well, one conjecture about the sources of this collective anxiety on the risks of the new activities holds that their implementation has required major changes in the basic institutional hierarchy and in the infrastructure to support it.

As a consequence, the knowledge base required to manage and oversee financial institutions differs considerably from the traditional training and experience of many financial managers as well as governmental regulators. Changes of this sort are threatening. It is difficult to deal with change that is exogenous with respect to our traditional knowledge base and framework and, therefore, comes to seem beyond our control.

Less understanding of the new environment can create a sense of greater risk even when the objective level of risk in the system is unchanged or actually reduced. That knowledge gap may widen since the current pace of financial innovation is anticipated to continue and even accelerate into the next decade. Managing this knowledge gap offers considerable challenge to institutions, but it also offers considerable opportunities to schools of management, such as MIT.

Now, the same points about managers and overseers can be made about households, the ultimate consumers of financial services. One major institutional change over the last 15 years or so has been the aggregation of financial services, especially at the retail level. Another has been deregulation combined by necessity with the reduction in government guarantees of financial performance.

These changes cast significantly more individual responsibility for risk bearing directly onto the household. In the United States, for example, the fully government-guaranteed bank deposit with an interest rate set by regulation, which was once virtually the only form of liquid short-term asset holding of households, now shares significant space with a variety of substitutes, such as money funds-- which are not guaranteed with respect to payment, interest rate, or liquidity.

Where once Social Security and defined benefit corporate pension plans assured households of a no-thought provision for retirement, there is now a major shift toward defined contribution pension plans in which the amounts available at retirement depend exclusively on the returns earned on invested funds during the work years.

Now, in all of this, the household now bears the responsibility for the allocation of the funds among investment alternatives. Even residential mortgages, which had almost always set at a fixed rate in the US, have now shifted towards variable rates which subject the household to direct interest rate risk exposure.

Thus, like financial managers and regulators, households are now being called upon to make important financial decisions involving risk that they have not had to make in the past and may not be trained to make in the present. And in this sense, the evolving structure of the financial system has a dysfunctional aspect. The successful financial institutions in the impending future will be those that can address this dysfunctional aspect while still fully exploiting the functional benefits of new financial technology.

To help forecast managerial demands, including the type of research and training needed to run financial institutions as we enter the next century, I turn now to the future product and service demands of the business and consider, to begin with, the household sector of customers.

Now, as I have noted to you, the retail investors in the most developed financial systems-- United States, in particular-- have experienced a secular trend of disaggregation in financial services. There are those who see this trend continuing and furthermore that the existing products, such as mutual funds, will be transported into technologically less developed systems. Perhaps that is what will be the case but I also believe that the trend will reverse back towards more aggregated products.

If you'll permit me a bit of an analogy to the idea of consumers, when I was growing up, I had in my living room a hi-fi that was in a box to listen to music. It had the record player, the amplifier, everything in a box. We bought a box-- a very handsome box, by the way-- and we listened to music.

And then evolved the components explosion , and you had hi-fi shops where you had amplifiers, preamplifiers, woofers, subwoofers, all these things all over the shelves. And it was wonderful. I'm not suggesting that this was a mistake. But it also had its dysfunctional aspect, which meant that most of us who are either not trained nor particularly interested in building systems would walk in and say what we would like to have, in fact, is not the component parts-- not the woofer, the tweeters, or whatever. We want to have something that delivers to us the music product at a price that we could afford in the environment that we're in.

Now, what happened? What has happened? We've seen a trend back to the box. Of course, it's not the same box that I grew up. It's a much better one as the result of technology. But nevertheless, it's a box. Perhaps the best known is the so-called now, I guess, media room, where the whole room was made of blocks. But the idea is, at the consumer level, to bring back and move toward the direction of aggregated finished products rather than leave the consumer with thousands and thousands of mutual funds, untold other individual financial products, and ask the consumer-- who is ill prepared to do so either by training or desire, and certainly is not the unit that is best served by bearing what we call basis risk or shortfall risk-- by asking them to assemble these in some way that approach anything optimal certainly doesn't seem the most likely event as how things should work.

So in thinking about the new financial products, they will be aggregated. They should be easy to understand. A case would be, for example, someone who is looking who had a two-year-old child who was concerned about tuition for that child going to, perhaps, MIT or Harvard-- but maybe some other university here in the United States-- in 16 years. Under most cases-- I know there are exceptions, but in most cases-- how is that planned achieved? You go to a financial planner. They say put money in a growth fund or some mix of growth funds. And if you put enough in, in 16 years you should have enough to pay for tuition. And if you look at historical returns, you can even make an estimate of how much you have to put in order to make tuition.

But what happens if the ex post performance of those collection of funds over the subsequent 16 years does not match expectations-- and in fact deviates considerably from it? Well, if it turns out that they earn a lot more than is needed for tuition, you might say, well, that's great. Except, of course, there could be some degree of regret because that says that the parents didn't have to put so much away, and they could have consumed it earlier-- something that cannot be reversed now. You can't go back in time as far, as I know.

But there's, of course, the other side. There's a shortfall. Who bears that risk? Under the decentralized, disaggregated products risk, the advisory-type business in this area, the individual household does. I would think that the move should be toward aggregation, towards a product that simply says you put so much in, your child's tuition is paid for. And let the professionals, let the institutions who are supposed to provide these services undertake the risk bearing the analysis and what we might call the basis of shortfall risk.

Now, an organizing theme for thinking about these consumer products in the future I think can be found in MIT's own Franco Modigliani's work on the lifecycle. Those of you who have the good fortune to still take his class know well what I talk about from the master, and so I wouldn't pretend to even talk about the lifecycle here to you. But I will say that I find the point that it's common to every time and place-- different countries, different periods of time-- the lifecycle feature is a common theme and thus highly transportable, and thus highly usable in a global environment with rapidly changing technology as a way of framing the kinds of products that we would like to design and offer.

Now, at this point in time, probably the most important in terms of impact in the next years will be the retirement phase of the lifecycle. I need hardly mention here the baby boomers, but also Germany, Japan, and quite another-- the other more developed economies in which we see a very strong sort of graying of the population and a need to address the retirement segment more effectively.

Just to give you some numbers, in Germany in 1990, 20% of the population were age 60 and over. By 2020, it's forecast that that percentage will be 30%, and by 2030 35%. Now, we all know how accurate demographic forecasts are and so forth, but even if they're off by quite a bit that's an extraordinary change for that country. And in any event, if it does reach 35%, It's a good bet that that's going to be close to half the voting population or something akin to that-- no, I guess it would be larger, smaller-- but a very substantial amount of the voting population.

I mention this because, if you look at Germany-- and I'm not a German expert-- you see that they have 60% marginal tax rates. They do not have any funding of either social security or corporate pension plans. And in fact, the tax incentives are against funding for corporate pension plans. There is mutual guarantees across the corporations of pensions. But as I understand it, those corporations could choose to opt out of the system by developing what we would call here in the States the equivalent of defined contribution plans. For all of those who are students of adverse selection and moral hazard, what do you think that suggests as to who are the ones to opt out of that system and who will stay if they still have the choice?

There is a historical unacceptability of inflation, and you add to that strains on EMU leadership if the EMU comes to be, and the question, although it's claimed that they have covered the costs of bringing in the East, whether they've included in those costs all the retirement and pension benefits as extended.

The parallel numbers for the US, 16.6%-- oh, I neglected to mention standard retirement age in Germany is 60, so that cut off is, by the way, not an irrelevant one in terms of percentages. In the United States, 16.6% to 28% in the period. In Japan, 17% to 33%.

Now, some see this more than just a crisis. They see disaster. Well, perhaps. One can also see in that a challenge and opportunity for financial services, for financial institutions to do their part to help meet this challenge of the graying of the population, meet this part of the lifecycle retirement worldwide, and profit by it. And not only is the aging making those opportunities operative, but as I'm sure you're all aware, there is an appearance of a trend away from the government component of providing for retirement-- what we would broadly call social security-- towards privatization.

So you have two events going on. Both the size of the population that's going to need retirement services globally-- I shouldn't say globally, in the more developed countries. This is not true in Latin America and in other parts, but certainly in the countries I mentioned and in most of Europe. So you have the aging, but you also have the fact that privatization is taking place of the retirement provisions. And with, that you have an even greater chunk being transferred directly to financial services firms to deliver.

If existing institutions, such as insurance companies and banks, cannot or do not change so as to address this challenge, I believe other new types of institutions will arise that will. But further discussion of the character of those institutions and the dynamics of institutional change are topics for my third lecture, so I'll move on from this one.

Now, all of this about the household products and services creates a kind of paradox. If the forecast is for consumer products to become more user friendly, simpler to understand what they do, and more customized and tailored to meet individual profiles, then paradoxically making the products simple creates considerably more complexity for the producer of those products. And thus, financial engineering creativity and the technological base to implement that creativity with reliability and at low cost will be a key competitive element in financial services.

Let me give you an example of what I'm talking about. Consider, again in the retirement phase, a life annuity-- that is, you receive payments for the rest of your life after retirement. So you're assured of receiving the funding for as long as you're alive. But instead of a standard annuity, at least standard in the United States, which pays a fixed dollar amount each period for the rest of your life, imagine that annuity is indexed-- but not indexed solely for inflation, thereby trying to provide a protection against inflation but nothing else, but also indexed against the standard of living. So that, if the standard of living in the environment in which you reside increases, the annuity payments increase in such a way that, in principle, you could not only enjoy the standard of living that you had, but in fact keep up in a relative sense to those around you.

Now, I'm not saying that's the annuity that everyone would pick, but it seemed to me like it sounds like a fairly simple and sensible one for a lot of people-- allowing you to maintain the standard that you had during the life you were working and allowing you to keep up with changes in the standard of living. And I underscore to you, that's much more important-- particularly for many of the people in the group here-- because I suspect you will be working longer than some anticipate now, but you'll also be living in retirement much longer.

And so, during your period after retirement, the exposure to significant standard of living changes may be non-trivial and more important.

But let's stick with that product, pretty simple to describe to a user. What does it take to create that product? Well, relative to the fixed money rate annuity, which basically with a little mortality risk any insurance company can create using readily-available fixed incomes instruments-- it's not quite that simple, but close to it-- this other product will require substantial engineering and innovation and technology in order to deal with the basis risks of the issuing institution associated not only with inflation, but with changes in the standard of living.

Complexity in the production process with the ends being simplicity for the customer will require elaborate and highly-quantitative risk management systems within the institution. That's the topic of next week's lecture. So let me move to the next class of customers to institutions, the corporate customer, and on the managing of risk outside financial institutions.

Now, we know that non-financial firms now use derivative securities and other contractual agreements to hedge interest rates, currency and commodity price risks. With improved lower cost technology and learning curve experience, both on the side of corporate users and on the side of financial firms who provide the advice or the instruments-- with this learning curve experience, this practice is likely to expand.

Eventually, this alternative to equity capital-- by that, I mean hedging-- as a control for risk may have a major influence on corporate structures as more firms can use this hedging and therefore free it up to move from publicly-traded equity to privately, closely-held equity. By that, I mean equity is kind of an all-purpose bucket. It's the cushion for risks that were unanticipated in the firm-- all risks because just pop. It's a cushion.

Hedging, whether it's commodity prices or interest rates or inflation rates or whatever-- degree temperatures in the state of Minnesota-- hedging is the opposite of equity in this dimension in a sense they're very specialized. They're targeted risk management. And so it takes quite different skill and a quite a different set of tools to implement that form of managing risk than the equity alternative.

But with those skills and the lower cost provision for hedging, then some of the alternative, the benefits of moving from large blocks of publicly-held equity to closely-held corporations, businesses where the equity is closely held, is made more feasible. So that's an example of an influence on structure outside of financial institutions.

The other major area for the applications or financial innovation products in the future for corporate customers is in strategic management of the firm. A favorite example of mine-- and some of you I recognize in the audience have heard it many times, but if it's a good example it's hard to not give it again and some of you at least may not have heard.

One of the ones I like is that of the synthetic refinery, where you imagine a firm with oil fields-- you can do this with gas, of course, but let's keep it with oil fields-- and also a distribution system-- let's say for heating oil and gasoline-- but missing the centerpiece, the refining component, that transforms the crude into finished products to be distributed. And let's suppose that this firm does a strategic analysis and decides based on the risks and the need to coordinate that it's decided to have vertical integration. They ought to introduce the refining component, the middle piece.

Now, in the past, that would have meant either buying a refinery-- existing one-- or building one, and that is indeed one possible solution and may even be the best solution in some cases. But now, there is an alternative, and the alternative is to enter into financial contracts in which, for example, you deliver crude oil in quantities and places at times and in return receive-- perhaps with a delay, not unlike that of physical production-- some mix of gasoline, heating oil, and so forth in return in a financial contract.

And you can jazz it up to have it be an option-type contract. So at the time of the delivery of the crude, you have the choice as to what mix of refined products you will receive just as you would with a refinery with multiple outputs from the input. Functionally, you're bringing in crude and you're getting back refined products. So in that sense, you've achieved the economic purpose of integration of the process.

But in many cases, that will be a far superior solution to the actual acquisition of the physical refinery. Now, we could go into issues of EPA. If you're in the United States, refineries are not some of the most pleasant things to have around, and there's a tendency to have them elsewhere. There's the expertise of going to environments and to countries, to areas to run these refineries where you have no expertise as an institution or a firm, and so forth. There's also a flexibility of supply, since a refinery is typically in one place-- that you can have subdivision of refining capability.

Now, the counter-party to this contract with this firm could be a specialty firm in refining, who specialize in that, but it also could be a financial firm which is in itself intermediating or laying it off with a host of other institutions-- perhaps in some cases, just simply using the standard futures markets as a form of hedging the contractual agreement.

Now, if I could give you a second example, one that is unfolding at this very time that involves the production and distribution of electric power. As documented in a delightful and informative case study by my colleague at HBS, Peter Tufano, the Tennessee Valley Authority, the TVA, in December of 1994 began a process of accepting bids to have qualified producers and perhaps some financial firms bid to sell to TVA long dated call options to buy power at a pre-specified price and to buy put options from TVA, giving them the counter-party the right to sell TVA power at a fixed price.

TVA, as I remind some of you if you don't know, produces and transmits power to a grid. It was developed in the '30s as a power-generating-- well, the TVA projects.

The scale of the bids was the power output equivalent of two to two and a half nuclear power plants, which being at MIT I have to be very careful because I'm sure there are those who will correct me, but what I understand is it's about the equivalent of 30 conventional power plants in terms of output. Thus, when this process is completed, TVA will have the economic equivalent of those plants without actually owning or building, and hence a synthetic power plant.

So they will have contracts which gives them the right to get power from the system from others, from the counter-party. These are very long dated contracts, not short-term options as you see typically traded on stocks and so forth. They have also sold puts to counter-parties. So those of you are facile with these things, you can see they've done something like a range forward contract or a synthetic forward contract, but it's more complex than that as I understand it because they're not matched and so forth.

But the essential feature is they made a strategic decision to create robust and low-cost cost supply of power by entering in these contractuals as a competing alternative to building the power plants. This process went on through most of 1995, and it makes feasible-- as with the refining case-- the economic integration of generation of power transmission and distribution, but without having to have the physical managerial integration and take advantage of the division of expertise that is necessary with actual integration.

It's been reported by the TVA but with no official connector-- it just came out in the newspaper after these bids were completed-- that they the TVA has made a decision not to build two nuclear power plants. Perhaps that's coincidental, but at least it would make a good story to suggest that they've used the financial contract that they've been engaged in as a substitute and have found it to be a superior substitute.

Now, to complement these custom-tailored contracts-- these are bilateral ranges between TVA and their counter-parties. They are not publicly known, so unfortunately I don't know what their particular terms are-- but to complement that, just about two weeks ago, futures contracts on electric power have begun trading on the New York Mercantile Exchange. And some are predicting that as much as $1 trillion of these contracts will be trading by the year 2000 and that the further effects on other businesses that depend critically on power is likely to be significant.

I should also say-- and there's another set of cases, little bit earlier-- that much the same story applies to natural gas. And indeed, there's a company called Enron that is essentially turned itself into what looks to me like an investment bank of sorts. But instead of dealing with money, it uses gas as the unit.

Now, a major requirement for the successful application of these contracting alternatives will be defined effective organizational structures for insuring contract performance, both for contracts issued to households and for contracts issued to corporations. And this will require global clarification and revisions of the treatment of such contractual agreements in bankruptcy. Simply put, if we're going to have these contracts, especially custom contracts, used with customers, used with corporations, and used effectively, it must be that what's promised is on the whole delivered. And it must be clear.

And I make an aside on that, that default on these customer contracts has significantly greater efficiency losses than defaults on debt of the same institution to investors. The reason being is investors expect that their returns from investing in a company-- whether it's debt or equity-- will be influenced by the future fortunes of that company. So that's part of the process of being an investor. But when you are a customer, corp customer or a household customer, you do not want the payments on your contracts to be contingent on the performance of the issuing institution.

The company that wants refining wants to get the crude in and get back the refined products, and it doesn't want to have it getting those conditional on which institution actually issued the contract. That's not part of the state contingency that's desired. And so there's enormous efficiency gains and growth gains by using them.

The need for assurances on contracts' performance is likely to stimulate further development of financial guarantee business for financial institutions, and such institutions will have to improve further the efficiency of collateral management as assurance for performance.

Now, I move to the third category of customer, government. First things that are needed by the government in this dimension will be education and education and training for oversight. You imagine what this is like for regulators, for supervisors to look at all of this complexity and sometimes oversee it. And clearly, they need to understand it.

One of the things we may see happening is that the solution is that it gets farmed out-- that, instead of government or government agencies coming in and doing all the checking, evaluation, there will be external firms-- specialized firms-- not unlike those of public accounting firms that we are all familiar with that will be created, perhaps called risk accounting or some other better type of term, but specialized firms that will go in and do evaluations of the appropriate institutions to see, indeed, whether they are complying with the relevant standards. That may be a far more efficient and effective way to undertake this, but we'll see.

In the direction, however, it's not just merely as the outside overseer, but just the activities of government itself in the financial sector that will make use of this technology. Monetary policy. As we all know, much of the at least theoretical effect of this-- theoretical reasons for effectiveness of monetary policy-- turn on frictions. You need some kinds of costs or frictions in the financial system in order to have the monetary policy bite.

Now, we have seen an extraordinary reduction in costs in the financial transactions-- principally through derivatives, but also in even the more traditional areas. The ability to contract and contract at low cost and the existence of the large number of markets have all combined together with, again, technology to reduce costs dramatically. So it's not at all clear that those approaches to implementing monetary policies that were entirely effective-- if they were-- in a relatively high-friction environment will also be effective in a low-friction environment.

Those of you who know Mount Washington up here, you know you can drive up there in the summer in a car with ordinary tires. That's one friction level. Try it again in the winter. You might find it's a quite different one. And so, what's a successful technology in high friction may not be in the low.

Third area, fiscal policy. More and more countries have to deal with the reality that they're depending on capital, their flows both back and forward with global investors. And so, specialized types of taxes, such as withholding taxes on debt and so forth, perhaps designed to generate revenues, are in fact just introducing largely frictions and cost.

Jaime Dermody and [? Tyrell ?] Rockafellar-- I don't know how many people in here know the name Rockefeller. It's not spelled the Rockefellers in finance, but perhaps in mathematics-- convex analysis, optimization, very eminent mathematician-- teamed together and wrote a little book, a little monograph called The Mathematics of Debt Instruments Taxation. And this was apparently written in collaboration with some lawyers who preferred to remain unnamed and uncredited.

And the title of the book was Mathematics of Debt Instruments Taxation in the United States. And basically, what they did is to try to take the entire tax code the United States on debt, taxation of debt. And as they put it, the first 70 pages is redefining a lot of terms that don't have a lot of consistency to them, but the last 12 or so have nothing but mathematical equations that purport to fully represent that whole debt structure with all the inequalities.

Now, for my point here, it doesn't matter whether they fully effectively done it. I think they probably have done a pretty decent job, but there's no doubt that somebody with what we have to do a better job if they haven't done a good enough job. And what is the implication of that? Tax legislation typically based on the case method dealing with specific debt contracts that are familiar are formulated in such a way that they did not contemplate very different kinds of contracts under the heading of debt. And as a result, there undoubtedly will be unintended consequences of that tax structure as the result of, if you like, the exploitation of using this mathematical structure that fully captures it and then applying optimization techniques which effectively minimize the tax costs or maximize tax benefits.

But this same technology could be used by the government to design more functionally consistent and effective tax rules. And thus, those who are going to be engaged in fiscal policy will have to come facile with these elements of analysis and security design.

The last element in government is the stabilization of currency and interest rates central bankers are now discussing using derivative markets as more efficient than the cash market for implementing both on interest rates and currency stabilization policies. And indeed, I wouldn't be surprised to see them doing it in the future in a major way.

Now, I'm looking at the time. I don't remember when I started, but I can see we're getting there, and it's getting a little warm. So I'd like, though, however, to point out one major last thought as it deals with this. Of course, this is not to say that financial technology and the institutions that provide this technology can provide a solution for all the world's economic ills. I've been sounding here like a cheerleader to some degree, like this is a solution.

In general, we know they involve trade-offs. I'll give you an example, talking about retirees. Retirees during the accumulation period certainly are providers of capital. They will better be able to take care of themselves in retirement if they earn high returns on that capital. So hence, we'd like higher returns for retirees.

But that's not the only policy objective or economic objective in our societies. We also have interest in more jobs, jobs, and growth. Well, the users of that capital-- entrepreneurs and business firms-- will clearly be able to invest more if the cost of capital is lower. So for that purpose, we'd like to see low required returns. So the users of capital want low returns. The providers of capital want high returns. And from a public policy point of view, I think both could, if not sympathize, empathize with both parties.

So these, at this level, are in apparent conflict, and therefore it must be resolved as a trade-off. But sometimes, both can be improved, make both better off. And how is that? Well, we tend to neglect it in our basic economics courses. You have the providers of capital-- let's keep it with the retirees-- and we have the users of capital-- let's say the entrepreneurs. And we just presume they get brought together in some frictionless, costless fashion.

But as we all know, it's a very complex distribution system that allows that capital to get from providers to users. And in certain environments, those costs could be quite substantial. The spread cost-- all I mean by that is retirees, the providers, get 8% on their investment, entrepreneurs and business firms have to pay 10% on their investment. And the spread, that 2%, is the cost of getting the two together, and that service is one of the things that financial institutions and financial markets and the financial system performs.

If we can find ways to reduce that spread, to tighten it up, to reduce the cost of that distribution, then it is indeed possible to have both parties do better-- from 8% to 8 and 1/2%, for example, from 10% down to 9 and 1/2% on the other side, a 1% spread. And to the extent that financial technology can be employed to reduce that spread, it could have a large and profound effect.

I'll give you one number. For the Republic of Italy, if you could find a way to reduce their interest costs by 1%, 100 basis points, you would give back to the Republic in the form of reduced costs 1 and 1/4% of the GDP of Italy-- at least the measured GDP. It's a huge number.

There is a great study out there to be done-- I'm not aware that one has been done-- of the economic benefits of the US mortgage market just in terms of the sheer size. I'm not to talk about that, but I throw that out to be thought of. These are very large and very real savings and efficiency and contribution that can be achieved.

If you want to see a nice case on the cost savings of using contractual agreements versus the cash market, again, another colleague of mine, Andre Perold at Harvard Business School, has written a so-called BEA case, in which he shows particularly for foreign investors-- whether US and other countries, or other countries and US, or why do we even have to talk about the US, one country to the other-- where those costs can be quite traumatic by using the cash market. But by implementation of contractual agreements such as swaps and other alternative forms of getting the same exposure, costs can be reduced even for institutions as much as 150 to 200 basis points.

Now, let me say one other example of that, and that is the idea of creating liquidity for emerging markets. And that liquidity could be created by effectively having those entities who don't need liquidity, such as pension funds, buy shares in emerging stock markets which are known to have very large spreads and put them in a box. Buy them once with the idea of never selling them again. Put them in a box. We'll call it a depository box.

And then, create a derivative market, whether it's a futures market or some other form, which trades in the index for that emerging country's stock market in which the natural, quote, "short side," that who's willing to issue the contract, is this pension fund which is fully hedged with the cash securities in the depository. The counter-party on the other side are all the investors who, for liquidity or other needs, could not otherwise invest in that emerging market because of the lack of liquidity. In effect, you're manufacturing liquidity by going to entities that don't need it and intermediating it to entities that do.

In the process, one would hope to see a substantial reduction in the spreads by spreading risk across a larger base, by opening up the possibilities to invest in emerging countries to investors who otherwise couldn't because of liquidity constraints.

Now, this form of creating liquidity is one of the earliest examples of financial engineering and derivatives. You may recall that people used to, at least as I understand it-- I wasn't there-- but they used to carry gold around on the trade routes, and they found that was very costly, hard to carry, easy to steal in some sense, and-- amazing to me, I didn't realize it-- hard to verify that it really was gold.

So someone got the idea, well, why don't we just deposit this goal which is very costly and expensive to carry around in a box? And we'll issue claims against it, and people will carry the claims around. And that was the early depository banks, which, by the way, worked very well in Europe and in a number of other places, and one could argue certainly helped trade. So one of the earliest derivatives were bank notes, bank deposits issued against these golds. This is a very old tradition.

Well, I've run on much longer than I have right to. You've been very generous to me I would just close by saying-- where to look for clues to future products? I would say look to finance theory rather than finance history. Practice is moving toward theory as much as theory is changing to better reflect practice.

I hope to exemplify that to those of you come back next week when I talk about the development of [INAUDIBLE] securities, something you all know is fundamental to understand the theory of risk-- has been since 1953-- but of which all of us who've studied it realized it was totally impractical to have in the real world. But indeed, [INAUDIBLE] securities do exist in the real world, and their structure is used is a key element of the production and risk management of financial products within sophisticated institutions.

So those of you who would like to come back next week, I'll talk about that. Thank you very much. It's a great pleasure. And thank you all for giving this wonderful building. It's a delight. It's a treat to be up here. Thank you.