Scenari > Numeri > Rischio
 
  
     
   
 
 
Can Science Help Solve The Economic Crisis?
By Mike Brown, Stuart Kauffman, Zoe-Vonna Palmrose and Lee Smolin
 
 
 
1. The crisis and regulation [*]

The main cause of the financial crisis is instability in the financial sector including the firms, institutions and markets which comprise it. To understand this instability, we have to begin with the legitimate primary purposes of the financial markets. One is to provide capital, as equity and debt, to the goods and services economy to allow it to thrive and grow. A second is to provide a stable repository for our collective savings. And a third is to responsibly provide appropriate credit to individuals. These legitimate functions have been hijacked by speculative behavior that was unchecked by regulatory structures. The consequences of this threaten to disrupt the productive efforts of millions of ordinary people who go to work every day to make stuff and provide services to one another.

In the decades leading to this crisis, the shift in our economic thinking from the long-term view on Main Street to short-term speculation and gratification on Wall Street have not only brought us to the brink of economic collapse, but have also compromised a sufficient flow of capital to important long-term initiatives—economic sustainability, renewal of infrastructure, abatement of climate change, and development of alternative energy sources—all important to a vibrant and sustainable economy.

This has happened before in history—in Rome, in Spain, Holland and England. More recently, in America, there were smaller crises before to the present one, perhaps early warnings—Black Monday, Long Term Capital Management, the dotcom bomb, and others. Now that we live in a global economy, we cannot afford the next crisis, an order of magnitude larger, in which the world's governments themselves will have to be bailed out. Rather, we can only hope that these governments are collectively up to the task this time.

There is honor and service to society in inventing and building companies and products that make life better for people, which should be justly rewarded. There is honor in architecting balanced financial regulation, to which we should dedicate careful attention. There is honor for the important financial sector when it functions as it should for the collective good, and this too should also be justly rewarded. Reasoned risk-taking by knowledgeable investors plays an important role in capital markets in providing support for initially risk innovations. But there is no honor in abusing our regulatory and financial systems for reckless speculation (i.e. gambling) that has no productive value for our collective future and that of the generations who will follow.

Nonetheless, blame will not get us out of this situation. We need to understand how and why the crisis happened and why warnings over the last years were not understood or heeded. We need to use this knowledge to stop this crisis and get the economy functioning again. In the longer term, we need to redesign and reregulate the financial system so that it performs its necessary functions without leading to periodic crises of global scale.

Two basic assumptions must guide any thinking as we undertake these tasks. First, economies, financial institutions and markets cannot function without a context of rules and laws, which regulate them. In a market, each participant tries to do the best they can for themselves. In a properly architected and regulated market this contributes to the public good. There is simply no place for an ideological discussion about regulation. Stable systems in nature such as individual organisms and ecosystems are regulated. So must ours be. The only relevant question is do the regulations work or not, where work means that stable markets allow an orderly flow of capital to and from the goods and services economy and the people who comprise it.

Second, mathematics, physics and computers already play a major and necessary role in our economic affairs. People with training in mathematical sciences play a big role on Wall Street designing and valuing complex investment instruments, and running sophisticated trading strategies. There is no going back to the era before banks and funds depended on quantitative analysis and big computer programs, and the scientifically trained people to run them. Along with economists with whom they work, other scientists and computer scientists now have a profound responsibility to see that their skills, the principles which they have found effective, and the tools they have wrought, are all used well and wisely.


2. The crisis is at least partly due to shortcomings a certain theory of economics

"Those of us who have looked to the self-interest of lending institutions to protect shareholder's equity (myself especially) are in a state of shocked disbelief. … It was the failure to properly price such risky assets that precipitated the crisis. In recent decades, a vast risk management and pricing system has evolved, combining the best insights of mathematicians and finance experts supported by major advances in computer and communications technology. A Nobel Prize was awarded for the discovery of the pricing model that underpins much of the advance in derivatives markets. This modern risk management paradigm held sway for decades. The whole intellectual edifice, however, collapsed in the summer of last year because the data inputted into the risk management models generally covered only the past two decades, a period of euphoria."

— Testimony of Dr. Alan Greenspan, US House of Representatives Committee on Government Oversight and Reform, October 23, 2008

When economists and other scientists study a complex system they begin by asking about what assumptions have been used previously in understanding it, and how well they have done compared to data. So if we approach the crisis in this way, we have to begin by asking about the principles and assumptions that have been used to construct and justify the complex financial instruments whose use contributed to the present instability. We want to know how these theoretical ideas have been tested, and whether or not the present crisis is evidence that the ideas that the financial system have been built on may need to be improved.

In fact, there is an economic theory that shapes much of our thinking, as well as the practices of investment banks and the decisions of economic policy makers. This is called neoclassical economics. It is based on the following assumptions:

i. Most of the time markets are in or close to stable equilibrium.

ii. Participants in markets act rationally to maximize fixed and known preferences described by definite and time independent utility functions.

iii. Participants in markets have perfect knowledge of the information driving the markets as well as all other participants.

iv. Prices are set by a deterministic process of joint maximization of the preferences of all involved in a trade.

v. Fluctuations in prices are small, random and uncorrelated.

vi. There is perfect liquidity so all prices are well defined, and all markets clear.

vii. There is no important difference between markets comprising a few individuals and those comprising millions, so simple models suffice to elucidate the principles that govern markets.

The neoclassical paradigm based on these ideas has had some undisputed successes. At the same time, it appears to have led to the adoption of practices and recommendations, which are at least partly at the root of the present crisis. These included the ideas that,

i. Regulation is limited or unnecessary because markets find and stay close to stable equilibrium where they operate most efficiently, leading to maximally stable economic growth, whereas regulation only leads to slower growth. But we face a potentially precipitous decline in economic growth and prosperity in the wake of some deregulation.

ii. Everything has a value or price, at all times, that can be uniquely determined by some definite objective process. This includes contracts that refer to prices of fluctuating variables at future times. There is experience with futures contracts, which have prices which are set daily by their being actively traded. But we are now seeing these values evaporate.

iii. This trading experience may be generalized to a claim that complex financial instruments which oblige actions to be taken at future times based on conditions not known till then, still have definite values and prices even if they are never or rarely traded. But part of the crisis is due to the fact that the balance sheets of banks and companies holding these contracts cannot be computed because they include instruments whose prices have been revealed as simply hypothetical and are now proving to be indeterminate [1].

iv. Stability can be increased by inventing and trading abstract complex financial instruments rather than principal contracts like stocks and mortgages. Examples are derivatives. Although these predate the birth of Christ and have been a factor in every economy of scale since, our markets have recently been flooded with a host of new ones which cleverly combine functions of different prices at different times into financial instruments whose values are purported to fluctuate less than the values of stocks making them up. The theory behind the possibility of combining fluctuating variables into variables that fluctuate less is critically dependent on the above assumptions, especially that the fluctuations are small, random and uncorrelated. But these assumptions have been shown to be false.

v. It has been argued that these innovative instruments should not be regulated even as much as stock trading because they function as insurance to increase stability. This was based on another false assumption that any mathematical function of the values of stocks at different times has a fixed and determinate value at any time.

vi. Because price determination is a definite process of maximization of known preferences in an environment of perfect knowledge, and because all values are definite, it can be in some instances automated and carried out by computers programmed to trade under specified conditions. But some markets thus operated have failed to function or clear trades.

Before we look more deeply into possible difficulties with the neoclassical paradigm, we have to also emphasize that it has been so influential because it does give important insight into how markets work in some circumstances. Nor has it ignored the possibility that markets can have instabilities. For example, there is a long list of well known market failures (principal agent problems, moral hazard, public goods, menu costs, lemon markets, adverse selection, rent-seeking behavior, incomplete knowledge, incomplete markets, multiple equilibria etc. …) So we do not want to ignore the successes of this paradigm or overlook the role that these well known understandings may play in understanding the current crisis. But we also want to ask if there are alternative ideas, principles and methods of modeling economic systems which might also provide the basis for wise advice and policy.

As a result, in part, at least of belief in the neoclassical paradigm, a very technical approach to trading has come to dominate markets based on complex financial instruments and strategies that require mathematical scientists and computers to carry out. Beginning as small speculative efforts, these now dominate markets. In most markets including equities and credit, the value of derivative contracts now exceeds by an order of magnitude the total value of underlying contracts, which must be traded to fulfill those derived from them.

When physicists made the atomic bomb they realized what they had conceived and immediately felt a sober responsibility to help make the world safe from their invention. At this time there is a responsibility for those with the knowledge and skills to understand the financial instruments involved in this crisis to help first to resolve this crisis and to next turn their attention to the design and regulation of a stable market system. This will involve economists, mathematicians, physicists, biologists, computer scientists and others working together to make a more stable economic system.

In our view, the current crisis does suggest there are weakness in the paradigm of neoclassical economics. In particular:

i. The big markets in the economy appear not to be in equilibrium. Not now, and perhaps also not normally. The fluctuations in the values of stocks, currencies, and commodities are often not random and uncorrelated and, as we have seen recently, they need not be small. Some other paradigm is needed to describe the workings of real markets.

ii. More generally, the theory of competitive general equilibrium is based on assumptions that appear to be too idealized. These include the assumption that at equilibrium prices are set so that all markets will clear no matter how the future unfolds and the assumption that each agent has a view on the value of all possible dated contingent goods [2].

iii. Participants in markets do not have fixed preferences, but the theory of competitive general equilibrium assumes that they do. Preferences change in time unpredictably due to changing tastes and circumstances as well as in response to innovations which introduce new products and eliminate the needs for old products, and we should acknowledge this. The unforeseeable aspects of innovations renders risk assessments problematic.

iv. There has been an unjustified extrapolation from simple models of markets with two participants and two goods (or something similar) to real markets with millions of participants and thousands of goods, a mistake we should not repeat.

v. Participants in markets do not have perfect knowledge, indeed their knowledge and beliefs about the market conditions are sometimes false or unreliable and different participants have different knowledge and beliefs. We should acknowledge this freedom because it is not the case, as sometimes assumed, that the lack of perfect knowledge by traders averages out as noise.

vi. This has the effect that swings in belief can crash markets and hurt people even when much of the machinery of the goods and services economy is healthy and prepared to function well with an orderly availability of capital.

vii. Increasing returns can lead to path dependence in the economy so that the evolution of an economy will depend on historical contingencies. This makes prediction and risk assessment difficult.

viii. There appears to be a basic lack of appreciation of the importance of relative scales. This is because of the misapplication of the neoclassical paradigm that the markets operate near equilibrium. Financial instruments such as derivatives indeed can do little harm except to those who use them, so long as they represent only a fraction of a market. However, an extremely dangerous situation emerges when their use grows to the point where so much equity is pledged in the resulting contracts that a movement in the markets in a non-random direction can introduce an instability in which the contracts are called but cannot possibly be fulfilled. Any meaningful discussion about whether a novel financial instrument requires regulation must involve the scale of its use.

3) Does science, including economics, offer an alternative basis for conceptualizing a theory of economic markets?

The answer is yes. For one thing, a critique of the neoclassical paradigm has been developing from within economics as well as from the study of complex systems for the last twenty years. To this can be added other insights about how to describe markets which depart from neoclassical assumptions. These can be combined to yield a new scientific conceptualization of economic systems. This needs to be developed before it can yield precise detailed models of the economy of sufficient complexity to be reliable. But it offers a lot of promise.

Key components of this view and the methodologies that underlie it include the following.

1) The economy is a physical system, involving flows of goods, information and energy, hence it might be useful to model an economy as a system in physics. However, while there is a concept of equilibrium in neoclassical economic theory, the concept of equilibrium in physics is not applicable to economies because it applies only to particular kinds of systems called closed systems.

These are closed off from the outside world and have fixed unchanging amounts of the goods that compose them. They have fixed amounts of energy, which cannot be added to or subtracted from the outside. Markets are not describable as closed systems, so the notion of their being in physical equilibrium cannot be applied.

2) Instead, markets are examples of systems physicists call open systems. They do not have fixed amounts of goods or currencies. They are situated inside larger open systems including the biosphere and energy and materials flows through them from the larger system that contains it.

3) Flows of energy and goods through an open system can drive its selforganization to meta-stable states. These states are not like equilibrium in that fluctuations around them are neither small, nor random, nor uncorrelated.

4) Instead of being in equilibrium, markets can be understood as self-organized critical systems. This is a theoretical construct that can be usefully applied to real markets. It leads one to expect that in steady states markets are approximately scale invariant. This implies a prediction for how certain quantities will be distributed in an economy, which includes wealth, income, sizes of firms, populations of cities, and total values of currencies. The prediction is that these distributions are power laws and the prediction is observed in the real economy.

5) Complexity matters. In neoclassical economics many conclusions are drawn from studying situations with two traders and two goods (or similar simplifications). The conclusions from modeling these and other simple systems are then applied to real economies with millions of participants and thousands of goods. The new paradigm finds that the qualitative features of real economies cannot be correctly captured by such simple models, because basic features of how they work depend on their size and complexity.

6) Heterogeneity matters. In the real world participants have very incomplete knowledge of markets and different players know different things. Different players also have different strategies which persistently co-evolve and change as the market, partially engendered by those changing strategies, itself changes. This renders risk assessment in financial instruments difficult. This diversity cannot be modeled by neoclassical economics, but it can be modeled by new techniques such as complex adaptive systems.

7) Economic growth is driven by the development of cycles of materials, goods, energy and money, which are analogous to cycles which comprise ecological systems. These cycles are understood as basic components of self-organized systems that are far from equilibrium. Neoclassical economics studies flows that do not generally close into cycles and hence miss the key issues regarding stability and instability of an economy.

8) There is a lot of experience modeling ecological systems, which are similar to economies, in that they are open systems which self-organize due to flows of energy and material. This gives us a methodological basis for theorizing an economy in a way in which the role of energy is intrinsic and issues such as what we do with the inevitable waste products of industrial processes necessarily arise.

9) Time scales matter. Different functional cash to cash cycles have very different time scales. Instabilities can be easily introduced by too strongly coupling processes on different time scales.

10) Complex systems can function in different phases, analogous to the different phases of matter: solid, liquid, and gas. Some of these are more hospitable to us than others. Transitions between phases can be abrupt and disruptive. Regulators of economic systems would do well to keep track of measures that indicate proximity to phase transitions and act to avoid them.

11) There is a natural language for describing and quantifying departures from the conditions of equilibrium described by neoclassical theory. These are important because these are conditions in which the assumptions that go into the design of many complex financial instruments such as derivatives cease to be reliable. This physics, but their general importance to economics was discovered only recently by Pia Malaney and Eric Weinstein [3]. This gives us a concept called curvature that measures how far a market is from equilibrium, how much markets fail to clear, how large arbitrage opportunities are, and the effects of changing preferences.

12) Economic markets can be described as networks of traders and transactions evolving in time. There is a well developed theory of such networks, which provides a useful methodology and language for economic modeling. For example, the notion of a small worlds network has been found to apply to economic systems as well as other networks in present use such as the internet. Computer scientists have a lot of experience developing and using models of complex systems based on such networks.

13) The number of distinct goods and services, the kinds of companies, and the numbers of ways to make a living, have all grown dramatically over time. This growth is driven by innovation and it in turn is a major cause of real economic growth. Models such as neoclassical economics work with fixed baskets of goods and services and thereby miss the key driver of growth.

14) There is a limit to the accuracy of future predictions, because the innovations that drive the increasing diversity of the economy and economic growth generally cannot be reliably predicted. But one can make an economy more or less hospitable to them.

These critiques of neoclassical economics point to the possibility of a new paradigm in economic theory and modeling. Given that some of these points have been argued and studied for decades, by economists and others, it is pertinent to ask whether they have led to economic models, which have proved accurate in predicting the behavior of real markets. There have been some successes, such as a model of the NASDAQ stock market [4]. At the same time it cannot be said that the complexity point of view has led to a well developed economic theory that presently stands as a full alternative to the neoclassical paradigm. For example, there has not been, to our knowledge, a large scale economic model built with these ideas which is presently up to the challenge of modeling the current complexities of the worlds financial markets. So there remains much to be done to develop and test models based on these ideas and see how well they do applied to the real economy.

Nor do these new ideas necessarily invalidate the successes of neoclassical economics. The conclusion we draw from this is that much more work has to be done. The good news is that both the successful aspects of the neoclassical paradigm and the newer ideas based on complex systems offer much scope for development of economics as a science. What is needed is an open-minded development of economic theory, as in any area of science, based on the development of detailed models, through which the applicability of different principles and hypothesis can be compared with real world data.

The real success of the American economy has been in its functioning well as an incubator for innovation in real goods and services. This has given us expertise and technology, which has now led to the invention of financial instruments whose use, in a failed theoretical context, largely unregulated or comprehended, threatens to undo all the economic progress of the last decades. The question in front of us is whether the same spirit of innovation can be applied to the principles of economic theory that governs financial markets so as to base the design and regulation of those markets on correct and verifiable principles and models.


4) What is to be done?

"… since the early 1980s …the way we train people to think … in main-stream economics…has left them almost unable anymore to distinguish the surface from the underlying reality. Not only was it the age of Reagan and the beginning of market fundamentalism that came in the early 80s, and the rational expectations revolution and all the rest, but a fundamental break in how we actually train our students to think. …Because the new kind of economic modeling, that won all the Nobel prizes said: you don't have to understand the deep picture…You don't really have to know underlying mechanisms in the economy because the prices reflect the underlying mechanism."

—Jeffrey Sacks, remarks at Earth Institute (Columbia University), October 20, 2008.

Classical economic theory was a product of the enlightenment, invented by philosophers who wanted to contribute to the growth of liberal democracy. They taught us how to construct societies conducive to human dignity based on a balance between cooperation and the freedom to pursue life, liberty and happiness. The highpoint of the enlightenment was the mutual influence of Newton, Locke and Montesquieu, who in turn influenced the founding fathers to adopt principles of government they believed came from observations of nature. It is in the spirit of their shared values and idealism that we today call for a renewal of the enlightenment approach to rationally understanding and governing human societies.

In the last century, science has developed new tools with which to understand complex and evolving systems such as the economy. We ask economists, other scientists and policy makers to work together to develop a new approach to conceptualizing and modeling the economy that is reliable enough to serve as a guide for building and regulating stable markets.

How are we to bring about such a transformation in economic theory? Let us separate the discussion into the immediate, near and long-terms.

The economic crisis has to be stabilized immediately. This has to be carried out pragmatically, without undue ideology, and without reliance on the failed ideas and assumptions which led to the crisis. Complexity science can help here. For example, it is wrong to speak of "restoring the markets to equilibrium", because the markets have never been in equilibrium. We are already way ahead if we speak of "restoring the markets to a stable, self-organized critical state."

In the near-term, Eric Weinstein has spoken about an "economic Manhattan project". This means getting a group of good scientists together, some who know a lot about economics and finance, and others, who have proved themselves in other areas of science but bring fresh minds and perspectives to the challenge, to focus on developing a scientific conceptualization of economic theory and modeling that is increasingly reliable.

To accomplish this, research has to be done to develop a new paradigm for economic theory and modeling markets. This research has to be carried out as the science it is, which is to say that assumptions must be tested against the real world, alternative theories must be developed and refined, and these must be compared with one another and tested again.

In all of this work economists, accountants and financial mathematicians should join forces with complexity theorists and other scientists with the goal of remaking economic theory and modeling so that it can offer reliable guidance for the organization and regulation of stable financial markets. The research has to be carried out in an interdisciplinary and open spirit.

The goal of this research is a new scientific conceptualization of economics and economic modeling which can provide reliable advice in constructing, running and regulating financial institutions and markets so that they serve the purpose of growth and stability in the standard of living of all people. Financial firms and markets must innovate to serve their larger purpose in providing efficient capital flows for the growth and innovation of real goods and services as well as a safe repository for our collective savings. Innovation in methods of speculation that are unrelated to stable production of goods and services and the efficient flow of capital within the system must be recognized and discouraged.

If this research succeeds then a discussion of whether a given financial instrument should be allowed or how it should be regulated should not be a matter of opinion or ideology. It should be based on detailed modeling and data taken from real world experiment and treated with the scrutiny brought to the introduction of a new airplane.

In the longterm, there needs to be an independent, non-partisan methodology for economic and financial modeling which involves globally agreed upon standards, as in the world of climate modeling. As in that world, one can imagine an international commission of economic scientists who develop, test and benchmark economic models against each other, and against past data, so that there is a reliable understanding of what the best models are and how reliable they are for studying different kinds of problems and predicting the impacts of proposed new economic and financial regulation. This will allow new proposals for innovative financial instruments or changes in trading rules or accounting rules to be tested in an open environment using best practices to understand their results.

An economy involves finding balance between long- and short-term objectives, acceptable distributions of wealth, and rewards for innovation and risk taking. Different governments may embrace different social philosophies and may seek to establish economic and financial regulations to obtain somewhat different desired results. The role of an independent, non-partisan scientific conceptualization of economics should be to provide these policy makers with a notion as to the likelihood that new economic and financial regulations they are considering will have the results they desire and that these will not involve unintended consequences to others.

One can also imagine that in the long term the software used to model markets and economies should be expected to be open, so that it can be critiqued and improved on by experts, whether they work in business, academia or government, and irrespective of the country in which they work.

Perhaps at this uncertain point it is good to end with a vision of what a broader scientific approach to modeling economic markets might lead to. As a result of increases in airplane safety and rigorous application of risk management procedures there have been fewer and fewer crashes per mile flown over the last decades. This has not damped innovation—new airplanes are regularly introduced incorporating improvements in engines, constructions and instrumentation. What we can aim for is a financial system similarly designed and regulated according to sound and tested scientific principles so that crashes and crises once a decade or so become a thing of the past, and in which innovation of goods and services thrives in a stable economic environment.

_______________

[*] This essay is based on ongoing collaborations and discussions with Jim Herriot, Jaron Lanier, Bruce Sawhill and Eric Weinstein.

[1] This is because the markets for such instruments are inactive as well as because of uncertainties around otherwise estimating the amount and timing of expected future cash flows from the rights embodied in such instruments.

[2] Such as a bushel of wheat to be delivered in May if it rains in Nebraska for a week in April.

[3] The role of gauge theories in financial mathematics was developed also by K. Ilinksi, Physics of Finance: Gauge Modeling in Non-Equilibrium Pricing).

[4 ] V. Darley and A.V. Outkin. "A NASDAQ Market Simulation: Insights on a Major Market from the Science of Complex Adaptive Systems." Singapore: World Scientific Publishing Co. Pte. Ltd., 2007.