Adapting Financial Rationality: Is a New Paradigm Emerging?

Adapting financial rationality: Is a New Paradigm Emerging?

Mona Soufian1, William Forbes2, Robert Hudson3

1Newcastle Business School, Northumbria University

2School of Business and Economics, Loughborough University

3Newcastle University Business School, Newcastle University

Abstract

We discuss the implications of an alternative to the efficient market hypothesis (EMH) the adaptive market hypothesis (AMH). The AMH may give a theoretical basis for a new financial paradigm which can better model such phenomena as the recent financial crisis. The AMH regards the financial market order as evolving, tentative and defined by creative destruction in which trading strategies are introduced, mutate to survive or face abandonment. The concept of investor rationality is less helpful than the distinction between investment strategies which are more or less well adapted to the prevailing market environment in which they are deployed. We outline how a more systematic and grounded basis for behavioural finance can be developed in line with the later approach. Based on this we develop testable hypotheses allowing the AMH to be distinguished from the EMH. Finally we discuss how the AMH can aid our understanding of important issues in finance. A crucial feature is that in the But in this survival of richest, as opposed to fittest, implied by the AMH there is much room for misallocation of resources as price and value uncouple. In this evolution of the financial market order the regulatory State features as a further market in which the vote market verifies orand disrupts settled market conditions.

Acknowledgements

We would like to acknowledge very insightful and helpful comments by the editor and three anonymous referees on an earlier draft of this paper which have considerably improved the work.

1. Introduction

The recent global financial crisis has dealt a huge and largely unanticipated shock to the world economy. The Queen of England surely expressed the thoughts of much of the world’s population in November 2008 when she asked in a visit to the London School of Economics why no one had seen the crisis coming.We believe the fundamental answer to her question lies in the dominance of the neoclassical financial paradigm. A few far-sighted people hadseen problems aheadbut were largely ignored[i]. The idea that markets rationally price assets and risk was so entrenched amongst influential academics, practitioners, regulators and politiciansthat dissenting views were completely marginalised.

Recent years have seen an almost continuous succession of financial crises including the emerging markets crisis of the late 1990s, LTCM, the bursting of the ‘dot com’ bubble, the accounting scandals at Enron and Worldcom, cumulating in the near collapse of the global banking system in 2008 and on going problems with sovereign debt. As Hyman Minksy (1986) has taught us financial crises are a recurring theme of economic history. What is so disturbing is the escalating frequency and intensity of the crises we now observe. Galbraith (1990, p viii) states the case thus“Recurrent speculative insanity and the associated financial deprivation and larger devastation are, I am persuaded, inherent in the system. Perhaps it is better this can be recognised and accepted.” Recently Ferguson (2012) has portrayed the 2008 Crisis as a critical point in a “great degeneration” of Western capitalist economises as they enter a “stationary state” characterised by the rule of law being replaced by the rule of lawyers and an intense rent-seeking amongst market participants for shares of a pie that has ceased to grow or has even entered decline.

Mainstream finance theory has clearly failed to anticipate, or even convincingly explain, the recent crises. Indeed to a large extent it might be considered to have caused them by giving intellectual authority to the ideal of unrestrained financial markets and dogmatically suppressing dissenting views. There seems a vital need to address this situation with new research programmes to better model reality. In the terms of Kuhn’s seminal work on the structure of scientific revolutions the crises are anomalies; that is a failure of the current paradigm to take into account observed phenomena (Kuhn, 1962). An accumulation of anomalies eventually leads to a change of paradigm. The current crisis may act as the breakpoint enabling serious consideration of different theoretical approaches.

Perhaps we can finally now accept the cyclical nature of financial crises and move on to explaining the function they serve. Periods of trauma and destruction may intensify the speed at which differentiation based on evolutionary fitness proceeds. This suggests seeking to build financial institutions that can withstand further crises may be a misplaced effort. It may be better to try to put in place bank resolution procedures that are both easily triggered and avoid capable of protecting the taxpayer’s purse. Taleb (2012) identifies a category of trader/entrepreneurs who thrive of volatile, if not destructive times. Such innovators are “antifragile” in the sense that they enter their own in periods when pressures to survive are there most intense.

As one would expect a crisis of the magnitude we are experiencing has given rise to enormous debate. There have been hundreds of popular and academic articles and books on the subject[ii]. Different authors have emphasised different perspectives. Much of the popular coverage has personalised the issues. Often individuals have become scapegoats for behaviour they personify, for example Richard Fuld, Sean Fitzpatrick and Fred Goodwin, the CEOs who presided over the demise of Lehman Brothers, Anglo-Irish, and the need for the government rescue of RBS, respectively.A simple assertion that we have a flawed and greedy banking culture is now commonplace as a result.

Whilst it is surely prudent to rapidly address particular flaws in financial practices and regulations much of the post crisis response has been very piecemeal and ad-hoc in nature. This type of response is inevitable given the evident lack of an appropriate and credible theoretical basis to inform policy. The deficit in theory has been recognised even in some essentially practical and hard-headed assessments of the crisis.This is one of the primary insights of the UK's Turner Review into the failure of regulatory authorities to head off the burgeoning securitised debt crisis (Turner, 2009). Turner concludes (Turner, 2009, pp 85) “the conventional wisdom relating to the global financial system – that risks had been diversified – was widely accepted and was wrong”. If the ability to diversifyas a risk reduction strategy now looks tarnished in the face of systemic risk then the very fundamentals of received professional wisdom are in doubt. In a related review of the UK’s equity market’s John Kay explicitly demures from a view (Kay, 2012), that “public policy should proceed as if these ‘irrational’ behaviours did not exist: such an approach would not be consistent with the fundamental goals of performing high performance companies.”This insight echoes Jean-Claude Trichet’s (Governor of ECB) view that many of the economic models used by advisors during the financial crisis (Trichet, 2010) “seemed incapable of explaining what is happening in the economy in a convincing manner.” Thus the EMH seems to have been discarded as the basis of sound public policy and the race is now on to gain acceptance of a credible alternative. The AMH could become such an alternative source of guidance.

In this paper we question the core assumption of the EMH which is the ‘rationality’ of economic agents[iii]. FollowingGigerenzer et al (2003) we argue that true evolved rationality emerges when the response when investors’ cognition is a good match to the demands of the environment in which they find themselves trading. The distinction here is between the investment strategy and its cognitive and external context, as opposed to a proposed statistical property which is conjectured to prevail regardless of context or cognition. Based on the above, in this paper we discuss directionsfor future research which offer some hope to build a more persuasive and useful theorisation of financial decision-making. Initially we propose replacement of the concept of the Efficient Markets Hypothesis (EMH) that financial markets always act to set prices ‘rationally’by an understanding that prices change as investors’ constantly adapt their behaviour as markets evolve their own internal order.The latter process is known as the Adaptive Markets Hypothesis (AMH) and was initially proposed by Lo (2004, 2005). The AMH recognises the importance of behavioural finance and was partly designed to offer a way to reconcile this emergent literature with the mainstream. Our second, and complimentary, proposal is to work towards a more systematic and theoretically grounded basis for behavioural finance, building on the work of Herbert Simon on bounded rationality, and the research programme of Gigerenzer on heuristics. At the moment behavioural finance is somewhat fragmented from a theoretical point of view and can be criticised as often being anarbitrary catalogue of observed departures from rationality without a unifying theoretical vision to explain those anomalies.

Rebuilding financial theory is a huge task and it would be naive to over-commit to a specific way forward at this stage. However, the research agenda we propose is far from exclusive in scope. The AMH ismuch less theoretically restrictive than the existing paradigm in that it recognises the possibility of awide range of responses by players in the markets rather than taking a particular view of investor rationality as axiomatic. An advantage of the approach is that it may reconcile understandings from various research traditions, including neo-classical economics, behavioural finance and psychology. It can also accommodate different modelling approaches. This is important as the rapid development of computing is introducing radically new research tools such as the direct modelling of economic agents (agent-based modelling).These can be characterised as a ‘bottom-up’ approaches asthey focus on the behaviour of the individual agents in the market and then aggregate this behaviour to deduce the implications for the overall market.This is, of course, a largely meaningless activity within the neo-classical world as the agents are axiomatically assumed to be both homogeneous and rational, according to the normative construction of the “representative agent” model.In contrastto the agent-based approach most research to date, both neo-classical and behavioural, takes amodelling approach that can be characterised as being ‘top-down’ with an emphasis on observing movements in the prices of financial instruments rather than the underlying behaviour of market participants. In general, the motivation of the participants cannot be directly observed. This creates difficulties for practitioners of behavioural finance as motivations can only be deduced from indirect evidence. From a methodological point of view it would be premature todefinitely favour the new agent-based approaches over much more established methods but equally it would be foolish to reject them out of hand. They certainly permit the testing of hypotheses that relate much more directly to the behaviour and motivations of individual market participants.

While the AMH seems like a radical new departure to standard finance theorists evolutionary theory has a history of use within transaction cost perspectives on both the development of corporate organisational form and governance structures operating within any chosen form (Nelson and Winter, 1982). This evolutionary theory of economic change shifted the analytical focus from managers maximising a given objective of profits, or sales,etc, to the selection of “routines” appropriate to a fluctuating and uncertain environment. Rather than invoking calculus to solve for implicit maxima the evolutionary perspective favours processes, such asseems to parameterisemarkov-switching processes, for rotating across alternative “routines”, selectedto fit the trading environment the company faces. Such routines include “well specified technical routines for producing things, … research and development or advertising and business strategies about product diversification and overseas investment.” (Nelson and Winter, 1982, p 14). In tranquil times old established routines are likely to remain unchallenged and comfortably embedded. But in more turbulent periods, rapid technological, or regulatory, change induces threats to survival requiring major revisions to the set of routines adopted. In such critical periods, as Nelson and Winter (1982, p 58) state“it is more natural to represent large scale motivational forces as a kind of persistent pressure on decisions, a pressure to which the response is sluggish, halting and sometimes inconsistent…. an … evolutionary purging of motives that diverge excessively from survival requirements.” However, within these bounds, imposed by the need to survive, established, comfortable, if sub-optimal, routines abound. So while within the Nelson and Winter (1982) schema environmental changes initiate switches in prevailing routines. Realised learning (RL) methods characterise routine rotations as emerging from the diffusionof individual successful adaptions through a broader investment community.

The evolutionary perspective on organisational form reflects a much older interest in the analogy between economic and biological processes which dates back to Bernard Madiville’s satirical sonnet “The Fable of the Bees” (1714) where the Bee hive mimics the market by allowing the struggle for individual survival (or private vices) to produce a perfect, if brutal, social order (of questionable public virtue). Indeed Charles Darwin was much inspired in conceiving of “The Origin of Species” by Robert Malthus’s “Essay on the Principle of Population” suggesting Economics and the life sciences shared a common analytical frame in much the same way as Econophysics marries finance and the natural sciences now (Ferguson, 2012 p 63).

In the next section of the paper, Section 2, we initially outline thebackground to current mainstream theory with its emphasis on perfect rationality and follow this with a discussion of an alternative approach which assumes that rationality is bounded and the resulting implications of this for our proposed conceptual approach. The remainder of the paper is structured as follows: Section 3 reviews the Adaptive Expectations Hypothesis (AMH) followed by an examination of the role of bounded rationality in recent theoretical developments and considers how the AMH may be tested using ‘top-down’ approaches. Section 4 examines how a more systematic and theoretically grounded basis for behavioural finance can be developed.The internal and external bounds on financial decision-making are examined in sub-section 4.1. Heuristics and adaptive behaviour are examined in sub-section 4.2, followed by discussions on satisficing and investor heterogeneity in sub-section 4.3. Finally, sub-section 4.4 outlines a tentative modelling strategy for an evolutionary perspective on financial innovation based on agent-based modelling. The various sub-sections of Section 4 imply testable hypotheses, allowing us to distinguish the predictive power of the AMH from the EMH using the ‘bottom-up’ methods associated with agent- based modelling. Section 5 considers some practical examples of how the AMH and EMH differ by considering some current areas of finance research. The paper concludes with overall remarks and suggestions for future work in Section 6.

2. Background

2.1Mainstream Theory and Perfect Rationality

The classical assumptions of Finance theory are broadly that individuals are rational, seek to maximise the expected utility of their wealth, are risk averse and follow the tenets of subjective probability. Capital markets in turn are perfect and generate financial returns which are not predictable. Despite broad critiques, not least in this journal, (see, for example, Hudson et al, 1999; Keasey and Hudson, 2007; Hudson and Maioli, 2010; Shiller, 2000; Clarkeson, 2009; Krugman, 2009 and Akerlof and Shiller, 2009) this mainstream approach has remained very dominant. There are reasons for this rigidity, all the elite finance departments and academic journals are overwhelmingly dominated by scholars steeped in the mainstream approach (see Whitley, 1986 and Fox, 2010 for accounts of the rise of this dominance). In addition, the mainstream approach is very closely allied to the philosophical belief that free markets are the best way to allocate resources. This was almost a point of patriotic faith in the US during the Cold War era when mainstream finance was developed and certainly an easy and powerful argument to make after the collapse of the Soviet Union. However, it may be that, like reason itself, markets are a good servant but a poor master. Below we discuss the concept of rationality as employed in mainstream theory and then outline how the concept has become so associated with free market policies.

The standard notion of rationality employed by economists is that of purposive rationality. There are no value judgements of the desired ends simply whether the methods used to achieve them are optimal. This notion lends itself to logical axioms which can be used to determine rationality. For example,in his seminal multi-million copy selling text book, Paul Samuelson, often considered the most influential post war economist, codified four principles of rationality: completeness, transitivity, non-satiation and convexity (Samuelson, 1948, p45)[iv]. From this viewpoint the human being is seen as autility maximising, calculating machine (or “Laplacean Demon”) which raises obvious difficulties once we reflect upon studies of actual human behaviour or merely our own social observation. Even Kenneth Arrow, a Nobel Prize winning economist whose work is synonymous with the logical analysis of economic issues,admitted that human beings could not be rational in this sense (Arrow, 1986).