My reflections start from the observation of a rather curious phenomenon: the stock market, the most abstract and advanced field of the alleged highly rational sector of the economy, is very often wrapped in a mysterious and strangely irrational aura. Stock prices, it is said, move in an unpredictable way and for obscure reasons;1 the stock market is described as "an oracle, issuing mysterious and meaningless pronouncements,"2 in whose interpretation luck and intuition count more than rational calculation and the assessment of probabilities. The most widespread procedure seems to be the attempt to make sense of events that are in themselves purely random, meaning the price of securities turn out to be the product of self-fulfilling prophecies and non-consequential reasoning. These valuations are, as such, related more to the mutual influence of investors' expectations than to the actual state of economic factors.
Francesco Cesarini and Paolo Gualtieri, La borsa. Il Mulino, Bologna, 2020.
Robert J. Shiller. Irrational Exuberance. Princeton University Press, 2000. p.99.
Beyond the apparent formalism, in the financial markets there seems to be a sort of "magical thinking"3 at work, which refers to the "original magic" of the creation of money, whereby spending money to buy money can increase the amount of money available, and therefore, increase the volume of money circulating.4 It is not by chance that people speak of ‘mantra’ for stereotyped and continuously repeated statements in the markets, which seem to be effective due to sheer repetition,5 and in general there is a proliferation of terms that directly recall the procedures and practices of divination.
Robert J. Shiller. Irrational Exuberance. Princeton University Press, 2000. p.143.
John Kenneth Galbraith. A short History of Financial Euphoria: Financial Genius is Before the Fall. Whittle Direct Books, 1991, p.56 it.tr.; Niklas Luhmann, Vertrauen: Ein Mechanismus der Reduktion sozialer Komplexität. Enke, Stuttgart. 1973.
Mario Sarcinelli. “Introduzione”, in M.Tivegna /G. Chiofi, News e dinamica dei tassi di cambio. Il Mulino, 2000. pp.13-19.
This is not a recent attitude: already in the mid-nineteenth century, a financial commentator stated of the stock market that: "the most rational way to decide, in our opinion, in assessing the convenience of being bullish bulls or bearish bears, is to close our eyes, flip a coin and follow the response.”6 It is also a tendency that research in the field is not immune from, something which has been labelled as "no more rigorous than the reading of tea leaves."7 The parallelism between speculation and gambling is commonplace since the beginning of stock markets from the mid-seventeenth century onwards, with a continuous exchange of people and strategies from financial exchanges to dice or card games and vice versa.
Quoted in: Edward Chancellor. Un mondo di bolle. La speculazione finanziaria dalle origini alla new economy. Carocci, 2000. p.180.
Robert J. Shiller. Irrational Exuberance. Princeton University Press, 2000., p.xiv, representing many similar formulations.
How can we explain this curious attitude, which questions the irrationality of the ‘market of markets’ using terms and procedures that belong to divinatory logic, i.e. to a totally different order of rationality? The divinatory way of thinking was not merely arbitrariness and superstition, but on the contrary a highly refined system to eliminate arbitrariness, offering decision criteria in situations of uncertainty.8 Why do Western societies, which generally reject divination so vehemently as superstition, continue to believe in the rationality of the markets in the economic sphere?
Elena Esposito. Soziales Vergessen. Formen und Medien des Gedächtnisses der Gesellschaft. Suhrkamp, 2002. The "solution" was then to obtain indications from chance (the study of the entrails of animals or of the flight of birds), to propose self-fulfilling prophecies (Oedipus) and circular reasoning of the “post hoc ergo propter hoc” type (the indications of oracles), to use obscure and enigmatic formulations that remained valid even when events contradicted them (the interpretation was wrong) and in general to rely on forms of circular constraints that are diametrically opposed to the linear logic of current rationality.
An almost undisputed premise of any observation of financial markets is the assumption that they have a positive function in the economy. Financial instruments are assumed to be useful to the "real economy" because they make it possible to distribute and select investments in the most logical way, contributing to the formation of that "efficient market", which is considered an essential prerequisite for the proper functioning of the economy.9 A market is defined as efficient when all information is available to all operators. The prices of securities should then faithfully reflect information and, therefore, the state of the fundamental economic variables. The stock exchange contributes to transparency, hence to the publicity and dissemination of information, which in turn is the prerequisite for the self-regulating mechanisms (first and foremost competition) that are supposed to guarantee the rationality of the economy.
Francesco Cesarini and Paolo Gualtieri, La borsa. Il Mulino, Bologna, 2020. pp.9, 119.
The financial markets, however, which are ostensibly the starting point of the virtuous circle that guarantees the rational functioning of the economy, show a series of phenomena that are not very rational and quite inexplicable. These are well-known trends which should not occur in an efficient market. If the market is efficient and all information is available to all operators and used appropriately, this must also apply to information on the irrationality of the markets, which should also be exploited, thus cancelling the potential for gain. But this does not happen: the markets continue to reproduce phenomena such as the ‘return to the mean’ (whereby securities that have lost or gained a lot tend to downsize), the weekend effect (whereby the values of securities are higher on Friday afternoon than on Monday morning) or the revaluation in securities in the days following the expiry of public sale or exchange offers – even though everyone knows or can know it. Then we are dealing with a double contradiction to the hypothesis of the rationality of markets: their most abstract sectors, financial markets, seem not to follow rational criteria and do it in a way that conflicts with the ideas of efficient use of information that are their premise.
Therefore, it is irrational that these markets are irrational. In the case of speculative bubbles, irrationality seems to be fomented by the very availability of information. The faster the communication, the faster the "irrational euphoria" spreads during the "inflating" phase and the panic contagion when the bubble deflates – both equally irrational.
Economists usually trace back these phenomena to certain factors, which are described but not explained – as if they were nuisances, annoying but ultimately non-essential, that may hinder but not seriously challenge the basic assumptions about the rational functioning of markets. For example, it has been observed that in all episodes of speculative bubbles, there is a widespread feeling of facing something absolutely new. This perceived exception would confirm the norm of the functioning of the economy but would justify a particular behaviour. For example, technological innovations such as telegraph, telematics or probability calculus itself, or the introduction of new types of bonds or other financial instruments. The assumption is, of course, the well-known shortness of memory of the economy, which not only systematically forgets the conditions and reasons for previous transactions (pecunia non olet), but also seems to quickly forget previous mistakes, collapses and erroneous speculations. With each new bubble, the same trends recur.
Well-known research in psychology on the limits of the rationality of individual behaviour shows that individuals, while adhering in abstract terms to the criteria of rational decision-making, in practice follow a series of heuristic principles that in fact contradict them.10 These principles are: illusion of representativeness on the basis of similarities or non-existent correlations; reliance on supposed internal coherences and compensations (regression towards the mean); and overestimation of intuition and the possibility of control.11 These are all tendencies that are particularly evident in financial decisions.
For more information see here as a reference to a huge debate: Daniel Kahneman, Paul Slovic and Amos Tversky (Eds.). Judgement under uncertainty: Heuristics and biases. Cambridge U.P., Cambridge-New York. 1982.
Ibid.
Even more serious, however, seem to be the induced conditions that can trigger perverse spirals of self-feeding irrationality during financial decision-making. The crucial point is the apparently irrepressible tendency of investors to orient themselves to the future and to other investors rather than to the current state of economic factors: expectations and emulations. In more abstract terms, the most severe difficulties seem to arise when one abandons the unambiguous reference to material data (the "real" economy as expressed in the fundamentals: company performance, profitability, profits, dividends and the like) to move to the temporal and the social dimension of the financial market: to suppositions about the future and the behaviour of other economic agents. Because the choices of each agent depend on what she assumes to be the behaviour of others (who naturally do the same) and because the future to which one is oriented in the present clearly depends on current decisions, but above all because each of these dimensions are intertwined, this leads to vertiginous forms of circularity. In other words, one’s prediction about the future, however correct, becomes false if others do the same, this is 'the dilemma of competition'.
For a sociologist, this indeterminacy is the starting point and the basic condition of social behaviour but to economists appears as an annoying source of irrationality. It manifests as the source of the "gregarious behaviour" and feedback loops that lead to a dangerous polarisation in the markets.12 In other words, such behaviours lead to uniformities in the interpretation of phenomena that enormously amplify market movements and tend to feed on themselves. Instead of referring ‘correctly’ to the trend of economic factors, agents tend to refer to the behaviour of other agents which is itself amplified by nebulous expectations about the future and circularly of other agents, until the market seems to lose all control and surrender to arbitrariness. As a result, the univocity of the world is lost in an uncontrollable fragmentation of perspectives reflected in each other and in the vagueness of the expectations about the future.
Charles P. Kindleberger. Manias, Panic and Crushes: A History of Financial Crisis. Basic Books, 1978.
So described, the complexity of financial markets seems to leave little room for planning and control; one might as well rely on chance, as shown by the references to divination mentioned above. This kind of attitude, however, is not without costs. If taken seriously, it amounts to a more or less explicit rejection of the basic assumptions of economic theory, and therefore also of its conclusions. One cannot limit oneself to labelling financial markets as islands of irrationality in an economy that, as a whole, is supposed to be rational and controllable – as many theories keep doing. In the last decades, finance has assumed an increasingly prominent and visible role in the economy and in the corresponding research, with computational models becoming increasingly refined and increasingly complicated. However, this sophistication has not been accompanied by the development of theoretical models capable of explaining, if not predicting, the performance of financial markets.13
A fairly undisputed remark. See, e.g., Galbraith 1991, p.103 it.tr.: "Nothing in economic life is so misunderstood as the big speculative episode."
Facing this situation, it becomes interesting to extend to the field of economics the transition, now consolidated in sociology, from first-order observation (observation of objects) to second-order observation (observation of observers, who themselves can observe objects or other observers).14 This is a move actually already present in the practice of economic operations. For example, since the sixteenth century, economic operations have abandoned the doctrine of the ‘right price’ and made the variation of prices depend not so much on the characteristics of the objects but rather on the cost of money, on political decisions, or on other mechanisms that give up the assumption of a perfect state of nature. This is also why prices can vary. What is observed in the economy is not exactly prices themselves, but rather, the possibilities that prices change, either to take advantage of opportunities or to protect against possible harm.15
Standard reference is von Foerster 1981. For an extension of the concept into sociology and an analysis of its consequences see Luhmann 1990, pp.68ff.
Niklas Luhmann. Die Wirtschaft der Gesellschaft. Suhrkamp, Frankfurt a.M.,1988. p.31.
Both of these possibilities derive from the behaviour of other investors. The period of the increasing relevance of the financial markets deliberately corresponds with the decision in 1971 to suspend the convertibility of the dollar (the reference currency for the foreign exchange market) into gold, thus giving up the remaining link to factors that do not depend purely on the internal dynamics of the markets.16 This does not mean that the external environment is irrelevant because ultimately, it is a matter of gaining or losing ownership of goods. However, the characteristics of these objects only indirectly affect the performance of the economy, whose constraints depend increasingly on the movements of the economy itself as a whole. The overall availability of money can also expand in ways that have nothing to do with the availability of goods – such as the development of new credit instruments or the increase in personal credit outside banks. In the current economy, the central position is occupied by banks and not by industry, by the circulation of money and not by the production of goods.
Chancellor 2000, p.226, quotes a financial trader's statement that "Information parity has replaced gold parity as the basis of the financial market."
What then, does the formation of prices on financial markets depend on if they are so disconnected from the reference to products? It is now relatively undisputed that the determining factor is expectations. It is not data, but the future as analysts and investors believe they can predict it. And since in the absence of certainty about the future, the opinions of individuals tend to influence each other, what one reacts to is primarily what one assumes to be the reaction of others. This leads us, as Keynes observed, to “devote our intelligences to anticipating what average opinion expects the average opinion to be.”17 The result should be the normal functioning of the markets, with prices reflecting the tendencies and orientations of the agents, who derive from this mutual observation the criteria directing their operations. Criteria that are circular and contingent, but by no means arbitrary.18 In the case of speculative bubbles, the mechanism becomes jammed because it immediately jumps to a third-order observation in which we "guess how the average opinion imagines the average opinion to be."19 The observers' observation immediately turns back on itself, in a sort of uncontrolled acceleration that leaves no time to produce and process information. Expectations are constituted in a form of synchronous “social contagion,”20 in which the expectations of each agent are constituted simultaneously with the expectations of the others to which they are meant to refer. This cyclical operation folds on itself quickly and prices no longer reflect the expectations of operators but only themselves. The self-referential order that usually governs the markets becomes mere noise, in which specific agents (the so-called noise traders) can also operate, obtaining gains by directly influencing market prices and speculating on the following expectations:21 the movement goes from prices to expectations, and not vice versa.
John Maynard Keynes. The General Theory of Employment, Interest and Money. Macmillan, 1936. p.156. – in reference to Americans' particular interest in speculation and the stock market. Not surprisingly, there is an affinity between public opinion and the market, both concepts that emerged during the eighteenth century, indicating the need, in a more complex society, for self-referential forms of observation: the standard reference in this regard is Habermas 1962.
The fundamental problem of analyzing the economy, according to Shackle, is "how should we organize our knowledge about the ways in which people organize their knowledge?": cf. Loasby 1999, p.8. See Shackle 1990.
Chancellor 2000, p.323.
Michel Aglietta and Andrè Orléan. La violence de la monnaie. P.U.F., 1982.
Andrea Beltratti. Teoria della finanza. Laterza, 1996. p.136.
In a self-referential economic dynamic these "accidents" cannot be avoided and are produced periodically due to more or less random events. What can be done to prevent this is to activate asymmetrisation mechanisms that hinder the implosion of observation and the uncontrolled acceleration of reactions to reactions. These include various devices aimed at curbing speculation and making it more controllable (particularly in the case of financial derivatives). The need to intervene from outside with appropriate regulations, however, confirms that speculation is nothing else than the other face of investment and a constitutive element of financial market operations. Therefore, one cannot rely on the (presumed) internal rationality of the markets to keep it under control. The inevitability of "bubbles" also shows that the dynamics of an economy based on reflexive expectations depends first of all on the use of time and on refined management of the mutual dependence of operators. We see that circular factors have little to do with a linear rationality based on decisions informed by stable and shared parameters and criteria.
From the perspective of second-order observation, the entire dynamic of the economy, and in particular the movements of financial markets, appear as a gigantic apparatus for managing contingency – i.e., uncertainty, which is not an inevitable evil but a fundamental resource. The uncertainty of the future, multiplied by the uncertainty of the behaviour of others (for whom the same applies) becomes the prerequisite for producing money and wealth, even if not for everyone. If the future were known, or if the behaviour of others were known, opportunities for profit would be available to everyone, thus mutually eliding each other. Investment risk is the source of possible profit, and it comes from managing uncertainty about the choices of other traders in the face of a future that is equally uncertain for everyone. If this is the case, information about fundamentals and real economy data is certainly not sufficient to guide economic decisions. Profit opportunities do not derive from how things are but from how others believe things will evolve, because their behaviour and confidence in certain economic operations will depend on the ‘market sentiment’ that we mentioned earlier.
For such an arrangement combining social and temporal uncertainty, our society lacks interpretative categories, that is, unless one resorts to a divinatory rationality which works precisely with circularity and simultaneity. Obviously, the social and semantic assumptions taking place in financial and divinatory predictions are profoundly different. The parallelism is limited to the management of information, which in both cases seems to lack external selection criteria and requires the use of self-referential procedures. In divinatory frameworks, characterised by the omnipresence of the divinity and the lack of a clear separation between sacred and profane spheres, every sign and every event (the colour and shape of smoke, the direction of the flight of birds, the configuration of the entrails of sacrificial animals) could in certain circumstances convey divine intention.22 The problem is not a lack but an excess of information, which becomes as such non-interpretable. We could call this a form of "information overload" not dissimilar to the typical syndrome of our technical and computerised societies.23 In divinatory systems, this excess is controlled by isolating well-circumscribed and regulated procedures such as the responses of oracles–the shapes of livers or the configurations on turtle shells–about which specific interpretative techniques had been developed. Information is thus obtained by narrowing the scope of possible informativeness and on the basis of the intervention of the interpreter who, for example, interrogated the oracle or started the sacrifice.
Gerdien Jonker. The Topography of Remembrance: The Dead, Tradition and Collective Memory in Mesopotamia. Brill, 1995.p.183.
Luhmann speaks of divinatory societies as the first form of "information society": cf. Luhmann 1997, p.1092ff.
What does this have to do with our issue? One can think that the complex computational techniques used in finance recall this kind of rationality, more than a type of scientific rationality in the current sense. This is revealed by the recourse to ambiguous and elusive categories such as market sentiment. Computational techniques could act as a hook to limit arbitrariness and force a world of excessive (and therefore unusable) information into a format that can be processed. The information obtained is again self-referential and contingent, but – as we have seen – it is the only kind of information that can be useful under conditions of second-order observation, where agents observe other observers and their behaviour rather then directly observing the world. Of course, what must be given up is the assumption of an underlying divine intention, which, on the other hand, is imperceptible and has no heuristic role.
In the face of the evident incongruities of the classical model of rationality and its role within financial markets, one might think to give up its promises and look towards alternative models referring directly to second order observation, circularity and the use of uncertainty. From this point of view, the increasing references to magic, oracles, divination and mysteries that can be observed in recent descriptions of financial markets are not so surprising. Divination offers procedures to cope with the “unkowledged insufficiency“24 of the human perspective without shifting to pure irrationality. The juxtaposition of finance and divination might apper provocative, but divinatory procedures were highly controlled - they were techniques in the full sense of the word. Their distance from the established rationality of our society does not imply irrationality, but rather hints to a different form of rationality.
—
Elena Esposito is Professor of Sociology at the University Bielefeld and the University of Bologna. A leading figure in sociological systems theory, she has published extensively on the theory of society, media theory, memory theory and the sociology of financial markets. Her current research on algorithmic prediction is supported by a five-year Advanced Grant from the European Research Council. Her latest book “Artificial Communication: How Algorithms Produce Social Intelligence” will be published in 2022 by MIT Press.
Niklas Luhmann. Die Gesellschaft der Gesellschaft. Suhrkamp, 1997. p. 239.