Thursday, August 20, 2009

Markets: Incomplete, Inefficient, Self-Fulfilling – But Irreplaceable

A major review of “The state of Economics” was published by The Economist of 18 July 2009 (printed version; online version 16 July). It consisted of a general introduction on What went wrong with economics – And how the discipline should change to avoid the mistakes of the past, and two articles on “The turmoil among macroeconomists”, The other-worldly philosophers and on the foundations of financial economics: Efficiency and beyond, (all three pieces unsigned).

The review has both the merit and the demerit of being non-partisan, trying to be impartial and, therefore, being doomed to inconclusiveness. For instance, while it begins by defining Robert Lucas as “one of the greatest macroeconomists of his generation” it reports on many of the telling criticisms to which Lucas has been subjected. On The Economist‘s Blog (16 July), maxreuter soberly but most effectively commented on Lucas:

“Surely you jest. A more accurate description would be ‘successful economist’ or perhaps ‘leading economist’ of his generation. Remember, this is the gentleman who claims that there is no such thing as ‘involuntary unemployment’ – that all unemployment is purely voluntary. No doubt he believes that what happened during the Great Depression, and what is going on now is really an epidemic of laziness.”

“As the article points out: ‘...economists missed the origins of the crisis; failed to appreciate its worst symptoms; and cannot now agree about the cure. In other words, economists misread the economy on the way up, misread it on the way down and now mistake the right way out’. To refer to an economist whose theories were in large part responsible for much of the above as 'great' is a somewhat unorthodox use of the word indeed.”

“Finally a memorable quote from the 'great' Mr Lucas: ‘...the central problem of depression-prevention has been solved, for all practical purposes...’ – Robert Lucas, Presidential Address to the American Economic Association 2003. Perhaps we should leave it to history to judge the greatness of Mr Lucas and the usefulness, if any, of his contributions to economics.”

In spite of inconclusiveness, by and large The Economist’s review is useful and timely. There is no point in trying to summarise it, as it is already dense and condensed, it is available online and has been subjected to various commentaries in the press and on the blogs.

Three comments are offered here instead, on the importance of markets incompleteness and sequentiality, on the implausibility of the so-called Efficient Markets Hypothesis (EMH), and the frequent self-fulfilling nature of expectations. While these arguments amount to a strong criticism of any market system and its efficiency, in the end it must be emphasised that markets – for all their incompleteness, inefficiency, self-fulfilling nature – are the irreplaceable engines that keep an economic system moving. When these engines are running, a government can try and steer the economic machine. This machine may get out of control on its own, or the government itself may drive it off the road, or crush it. But when the markets/engines are not running, the whole economic system grinds to a halt. However, financial markets are different and do not deserve the same positive appreciation and the consequent free reins.


The Economist‘s review rightly criticises those economists who conveniently “assume that markets are ‘complete’ – that a price exists today, for every good, at every date, in every contingency”. These are the followers of the modern general equilibrium approach developed after Léon Walras by Kenneth Arrow and Gérard Debreu (Econometrica, vol. XXII, 265-90; Debreu’s Theory of Value 1960, etc).

An irrefutable criticism, as futures markets – apart from a handful of currencies, and standardised raw materials over a short time horizon of 3-6 months – are not the rule but the exception, not to speak of contingent markets. But this is the least of those economists’ worries, for they can and do argue – as Christopher Bliss did when I raised this issue with him a long time ago – that markets have a cost, and therefore only those markets will be activated whose expected benefits exceed their costs. Thus in spite of missing markets we remain in the best of all possible worlds.

As a matter of fact the incompleteness of markets cannot be dismissed, for there is one commodity – labour services – for which a forward market could be contemplated only in a society of serfs or slaves, not in a society of free individuals. An irrevocable commitment to deliver one’s labour services in the future to a given master, or a given firm, at a price agreed in the past, would involve a feudal/slave tie inconceivable in a capitalist economy characterised by wage labour (a tenured job involves an option to sell one’s labour at a predetermined wage, but not an obligation to deliver it). Therefore market incompleteness is not a question of costs exceeding benefits of missing markets (unless we tautologically define an institutionally impossible market as infinitely costly), but of the incompatibility between capitalism and feudal institutions in labour relations. It is no accident that Gérard Debreu never speaks of capitalism, but of an unspecified exchange economy. It is not (only) a question of realism, but of which economic system we are talking. One thing are the “parables” often indulged in by neo-classical economists, another thing the science fiction of imaginary planets on which with absolute certainty nobody has ever set foot or ever will.

The lack of future markets for labour services is also why lower wages may not deliver higher labour employment – because those lower wages cannot be guaranteed to continue in the future once unemployment falls – and why a social contract between capital, labour and government may be a superior arrangement with respect to labour market flexibility.

Sequentiality of markets

Nevertheless, the real problem is not so much markets incompleteness but their sequentiality, which is largely ignored. For markets to deliver an efficient resource allocation they should open for the very short time, ideally an instant, that it takes for all economic agents (including representatives of future generations, but let this pass) to express their demands for and supplies of all goods, on the basis of their individual preferences and original “endowments”, thus determining inter-temporal, contingent, equilibrium prices and quantities. Then, once an intertemporal equilibrium is reached for all states of the possible world, markets should shut for ever while all transactors and their successors execute without fail all their transactions to kingdom come. (We leave aside here the thorny issues of existence, uniqueness and stability of such equilibria; complications with externalities, increasing returns, public goods; and the only too often forgotten departures from perfect competition, not because these issues are negligible or unlikely but because they are too large and complex to handle here, and even in their absence there is more than enough to shake anybody’s complacency).

Fortunately for us, and unfortunately for market efficiency, our markets do not function as in this unknown and grotesque economic system. Markets open, shut and reopen incessantly, indeed in the global economy today they seldom shut, for almost invariably you can transact anything at any time of the day and the night somewhere or other on the globe.

In the world as we know it, in order to secure a good’s availability tomorrow, we do not have to express our demand for it today, even if in principle the good could be transacted today in any number of future markets. Therefore all the time, beside taking into account current prices of current goods, we act on the basis not of today’s spot prices of future goods, but of our expectations of their spot prices tomorrow (and of the quantities associated with them). This is the ultimate foundation of the keynesian theory of unemployment due to lack of effective demand; of liquidity preference and of the associated need for fiscal policy. If today’s savers had to express today a demand for future goods, current investment would be activated for their future supply and a generalised excess capacity could only be structural, i.e. due to mismatching of the structure of capacity and of demand. As things are now, whoever saves creates unemployment unless his savings are being spent simultaneously by someone else. It is odd that even critics like Joseph Stiglitz should concentrate on markets incompleteness (as well as asymmetry of information, see his Whither Socialism?, Cambridge Mass, MIT Press, 1994) and neglect their sequentiality.

The trouble is that the Arrow-Debreu model is the only rigorous theory that would support the claim of an efficient market economy. From this viewpoint we can conclude that an efficient market economy is a utopia, in the literal sense that it does not exist, it has never existed and will never be capable of existing anywhere. What had began as an attempt to theorise the efficiency of markets ended up, by the successive tightening of all the conditions necessary to such efficiency, conclusively demonstrating their inefficiency.

Paradoxically the most effective critique of this kind of general equilibrium theory has come from within that same theoretical tradition, from Jacques Drèze and his work on temporary equilibrium. This is a Hicksian concept (from Value and Capital), infinitely more destructive and promising, at the same time, than anything produced by neo-marxian or neo-ricardian or neo-keynesian economists, so much so that it leads straight to keynesian conclusions.

The Efficient Market Hypothesis

In the 1960s Eugene F. Fama at Chicago University and Paul A. Samuelson at MIT independently put forward the Efficient Market Hypothesis, i.e. the proposition that “prices fully reflect all available information” (Fama 1965). The title of Samuelson’s 1965 article, ‘Proof that Properly Anticipated Prices Fluctuate Randomly’ says it all: if markets are informationally efficient, in the sense of prices incorporating the information and expectations of all market participants, price changes cannot be forecast, and viceversa. Everybody will exploit the slightest informational advantage in profitable transactions. As the old story goes, if you see something looking like a $100 bill on the pavement you should leave it where it is, for if it really was a $100 bill someone else would have picked it up already.

Lucas (1978) buttressed this hypothesis with his notion of “rational” expectations, efficiently embodying the sum total of economic information (although really we should call them “successful” expectations, for there is nothing rational or irrational about them). These are the foundations of the denial of the Phillips trade-off between unemployment and inflation; of the presumed ineffectiveness of government policy; and of the theory of Central Bank Independence with sole responsibility for inflation targeting.

This Panglossian view of the efficiency of markets is reminiscent of the parallel claim, by Soviet planners, that their central planning was always necessarily optimal, because if they had known any better they would have made it better. Admittedly it should be easier for countless, decentralised market transactors to recognise mutually advantageous improvements through bilateral exchanges, than for a single central planning agency to spot and implement planning improvements unilaterally; but if plan construction was decentralised, as Oskar Lange had proposed back in 1937, mutatis mutandis the Efficient Market Hypothesis and the Optimum Planning Hypothesis would be equally plausible (or, rather, equally implausible).

The Economist‘s review covers some of the criticisms raised in the literature to this construct. “In 1980 Sanford Grossman and Joseph Stiglitz [AER 70, 393-408], pointed out a paradox. If prices reflect all information, then there is no gain from going to the trouble of gathering it, so no one will. A little inefficiency is necessary to give informed investors an incentive to drive prices towards efficiency.” This is a confusion between the properties of the end-result and those of the process by which that end-result is obtained.

The Economist‘s criticisms of EMH include also the neglect of “institutional frictions” in markets, such as the presence of some less well-informed “noise traders”. But The Economist‘s strongest criticisms rest on the challenge to markets’ inherent rationality raised by behavioural economics in the past decade.

The behavioural approach

Andrew W. Lo (2007), a contributor to the EMH who now seeks its synthesis with behavioural economics, i.e. the Adaptive Markets Hypothesis, characterises the behavioural approach thus:

“Human decision-making under uncertainty” exhibits systematic biases “several of which lead to undesirable outcomes for an individual’s economic welfare – for example, overconfidence (Fischoff and Slovic, 1980; Barber and Odean, 2001; Gervais and Odean, 2001), overreaction (DeBondt and Thaler, 1985), loss aversion (Kahneman and Tversky, 1979; Shefrin and Statman, 1985; Odean, 1998), herding (Huberman and Regev, 2001), psychological accounting (Tversky and Kahneman, 1981), miscalibration of probabilities (Lichtenstein, Fischoff and Phillips, 1982), hyperbolic discounting (Laibson, 1997), and regret (Bell, 1982). These critics of the EMH argue that investors are often – if not always – irrational, exhibiting predictable and financially ruinous behaviour” (see Lo’s references list for bibliographical details).

The latest contribution to this approach is Animal Spirits, by George Akerlof and Robert Shiller, entitled after what Maynard Keynes regarded as the ultimate urge to practice enterprise and to invest[1]. More power to their elbows. But ultimately the strongest criticism of EMH is epistemological, for that hypothesis is based on a peculiar theory of knowledge.

The EMH suffers from what we could call the X-Files Syndrome, “The Truth Is Out There”. Out there is The Truth, some of which is naked and visible to all market participants; some of which is covered in such a way as to be visible only to some of them – either with total certainty or with an attached probability which is the same for all – and the rest of which is unknown but is known with certainty to be unknown. Information available therefore is somewhat limited, but what information is available is true – or likely to be true with equal probability for all those who share it – otherwise its privileged use would not necessarily give an advantage.

Instead of which, in the world as we know it, market participants have different beliefs about the probability of truth of the information they possess; the same “quantum of information” may be viewed as a gold nugget by some and as utter rubbish by others (for instance, the time sequence of numbers generated by a fair roulette to date, with a view to predict future numbers). Otherwise it would not be possible to even talk of “noise traders”; whoever uses this concept should explain what can be relied upon to contain the noise level below a critical, deafening number of decibels, while a bubble inflates up to the point when it bursts. I guess one can belong to the behavioural school without even realising it.

Self fulfilling expectations

If he were to re-write The General Theory – Maynard Keynes wrote in the Introduction to one of its later editions – he would distinguish between economic agents acting on the basis of wrong expectations and of right expectations. He also wrote: “For if we consistently act on the optimistic hypothesis, this hypothesis will tend to be realised; whilst by acting on the pessimistic hypothesis we can keep ourselves forever in the pit of want” (Preface, Essays in Persuasion, 1931). Joan Robinson used to quote Shakespeare to express the possibility of self-fulfilling expectations: “[… for there is nothing either good or bad, but] thinking makes it so”(Hamlet to Rosencrantz, Act 2, scene 2).

The Economist‘s review states: “… Nor can economists now agree on the best way to resolve the crisis. They mostly overestimated the power of routine monetary policy (ie, central-bank purchases of government bills) to restore prosperity. Some now dismiss the power of fiscal policy (ie, government sales of its securities) to do the same. Others advocate it with passionate intensity.” But in a recent, inspiring article (Economics is in crisis: it is time for a profound revamp, FT 21 July 2009). Paul De Grauwe explains beautifully the self-fulfilling nature not just of expectations but of dominant economic theories.

“Take government budget deficits, which now exceed 10 per cent of gross domestic product in countries such as the US and the UK. One camp of macroeconomists claims that, if not quickly reversed, such deficits will lead to rising interest rates and a crowding out of private investment. Instead of stimulating the economy, the deficits will lead to a new recession coupled with a surge in inflation. Wrong, says the other camp. There is no danger of inflation. These large deficits are necessary to avoid deflation. A clampdown on deficits would intensify the deflationary forces in the economy and would lead to a new and more intense recession.”

“Or take monetary policy. One camp warns that the build-up of massive amounts of liquidity is the surest road to hyperinflation and advises central banks to prepare an ‘exit strategy’ [see our post on Exit Wounds”, 3 August 2009]. Nonsense, the other camp retorts. The build-up of liquidity just reflects the fact that banks are hoarding funds to improve their balance sheets. They sit on this pile of cash but do not use it to increase credit. Once the economy picks up, central banks can withdraw the liquidity as fast as they injected it. The risk of inflation is zero.”

“Does it matter that economists disagree so much? It does. Take the issue of government deficits. If you want to forecast the long-term interest rate, it matters a great deal in which of the two camps you believe. If you believe the first one, you will fear future inflation and you will sell long-term government bonds. As a result, bond prices will drop and rates will rise. You will have made a reality of the fears of the first camp. But if you believe the story told by the second camp, you will happily buy long-term government bonds, allowing the government to spend without a surge in rates, thereby contributing to a recovery that the second camp predicts will follow from high budget deficits.” …

“This conflict matters not only for market participants, but also for policymakers.” … “The cacophony of analysis helps to explain why policymakers react in different ways to the same crisis and why it is so difficult for them to come up with co-ordinated action.”… “How to resolve this crisis in macro-economics? The field must be revamped fundamentally” (De Grauwe, cited).

Back to the drawing board, then: “We need a new science of macroeconomics. A science that starts from the assumption that individuals have severe cognitive limitations; that they do not understand much about the complexities of the world in which they live. This lack of understanding creates biased beliefs and collective movements of euphoria when agents underestimate risk, followed by collective depression in which perceptions of risk are dramatically increased. These collective movements turn uncorrelated risks into highly correlated ones. What Keynes called ‘animal spirits’ are fundamental forces driving macroeconomic fluctuations” (De Grauwe, cited).

Markets as indispensable homeostatic mechanisms

Well, markets are incomplete, sequential, inefficient, self-fulfilling… So what? All of these considerations taken together do not eliminate the absolute need for markets in resource allocation, for they are irreplaceable homeostatic, self-regulating mechanisms that adjust prices to excess demand (Walras), quantity produced to excess price over cost (Marshall), and actual capital stock to desired capital through investment decisions. Axel Lejonhufvud stressed long ago the markets’ self-regulating role (which should not be confused with self-regulation of their own functioning). More generally, markets also adjust production capacity and production of inputs to those of outputs (Dick Goodwin on the Multiplier as a Matrix).

Sometimes the self-adjustment is too fast, sometimes it is too slow - as it happens with all homeostatic mechanisms, like thermostats - but it is there; the alternative is manual control, i.e. central planning. As I tell my students, there may be special circumstances in which manual control is superior to an automatic mechanism. In Star Wars, when Luke Skywalker targets the heart of the Empire, he disables automatic controls and goes manual, and succeeds. But he only had one target; there were only two alternatives, hit or miss; and … the Force was with him. In Central Eastern Europe, on the contrary, the inability to introduce markets in almost forty years of attempted reforms – primarily because of persistent endemic excess demand, but also for the leadership’s fear of loss of political power and control – was the ultimate economic cause of collapse of centrally planned economies in 1989-91.

We could paraphrase Wiston Churchill’s dictum about democracy, and say that the market economy is the worst economic system except all the others that have been tried. But markets alone do not, on their own, characterise an economic system. We are engaged now, as never before, in an intellectual struggle over what ownership mix, what rules of the game, what share of government expenditure, what social and re-distributive policies, what global governance institutions should prevail in the modern market economy. While economic arguments are used, the choices involved are essentially political, and attempts to control or replace markets as the economy’s engines are always grabs for political power and control.


… that the markets' homeostatic properties praised above demonstrably do not apply to most credit and financial markets, where products may multiply instead of reducing and spreading risk; securities may go toxic; transactions are more likely to be fraudulent because of lack of transparency; there is a corporative system of rewards masquerading as a market for managerial skills; leveraged bets in derivatives markets can inflict massive damage on innocent bystanders; banking pyramids proliferate; profits are privatised and massive losses are eventually socialised. Here the primacy of markets must give way to effective regulation, first national then global.

[1] “Most, probably, of our decisions to do something positive, the full consequences of which will be drawn out over many days to come, can only be taken as the result of animal spirits - a spontaneous urge to action rather than inaction, and not as the outcome of a weighted average of quantitative benefits multiplied by quantitative probabilities.” (The General Theory of Employment, Interest and Money, 1936, pp.161-162). Akerlof and Shiller take “animal spirits” to imply irrationality, but the urge to action rather than inaction could well be a rational survival strategy, instead of the neglect of reason.


Francesco Bogliacino said...


as usual your posts are very deep and dense, so it is difficult to enter into all details. More or less you know my position, so I will just stress one point again.

On thing is theory, which serves as an instrument to validate our decision-making process, another thing is policy.

You are absolutely right about theory: what I called principle of non-transparency (but I will buy your X-files version) implies heterogeneous beliefs, which means that we are not sure to agree on the support and so we cannot in principle create markets for all possible contingencies. Let alone OBVIOUS transaction costs problems (for what I know free information violates even laws of physics) and institutional rigidities. Moreover, increasing the number of markets is not for free: e.g. it reduces the dimensionality of the space where the law of demand is satisfied (leave aside what excess demands do, since there literally "anything goes"). I agree with sequentiality, and so on.

I am not so sure I can accept the keynesian argument as it stands: if I save, I am not creating unemployment per se. Unless I take away money from the system, there will be a financial market. My "sequential view" is that people can transact at non-equilibrium prices, or prices can show some frictions, which can create unemployment. But these are details.

The delicate point is about policy. Put cognitive biases, put satisficing behaviour, put whatever you want, but what does guarantee that daddy can do better? This is the point, I think: how can politicians do better? My suggestion is that the problem of policy should be re-scaled at a lower dimension, where you may at least measure what is going on and have a decent theory of what is a causality effect; elsewhere, you have to use institutional settings to avoid the emergence of systemic problems. But maybe I am simply too much pessimist...

D. Mario Nuti said...

Thanks, Francesco. Are you too much of a pessimist? Why, ”A pessimist is an optimist who is well informed” - they used to say in the ex-Soviet Union.

I agree with you even where you think you disagree with me. You say: “if I save, I am not creating unemployment per se. Unless I take away money from the system, there will be a financial market.” I said “unless [my] savings are being spent simultaneously by someone else”, which is what financial markets can and should do but not necessarily always do: Keynes believed that money could be a bottomless pit into which purchasing power simply disappeared from circulation.

“This is the point, I think: how can politicians do better?” I had called them central planners, but it is the same thing. I said markets are irreplaceable - except financial markets need regulation.

And yes, trading “at false prices” on the way towards equilibrium, as Edgeworth pointed out, is likely to alter the final equilibrium through its impact not on efficiency but on income distribution (unless there is also consuming and investing at false prices on the way to equilibrium). I had left distribution out of the picture altogether; nobody ever claimed, or should ever have claimed, that markets can deliver a fair/just/equitable/acceptable/optimal distribution in any meaningful sense.

D. Mario Nuti said...

James K. Galbraith (University of Texas at Austin) sent me this comment via e-mail:

“All I can say is that you have great patience. I wonder how long (or even, how much) it is still necessary to point out the irrelevance of Arrow-Debreu. I long ago decided I would spend my short life on other topics!”

My patience derives from the fact that this model actually sets out the impossible conditions for market efficiency, thus providing the best foundations for irrefutable criticism of the market system.

Francesco Bogliacino said...

sorry, I relized that I wrote a silly thing (it happens when you don't re-ready :-)): in my intervention the correct phrases should be "increasing the number of heterogeneous agents is not for free".

Anyway, yes we agree on everything, I got your point..