A Fast and Frugal Finance: Bridging Contemporary Behavioral Finance and Ecological Rationality

By William P. Forbes, Aloysius Obinna Igboekwu and Shabnam Mousavi

A Fast and Frugal Finance: Bridging Contemporary Behavioural Finance and Ecological Rationality adds psychological reality to classical financial reasoning. It shows how financial professionals can reach better and quicker decisions using the ‘fast and frugal’ framework for decision-making, adding dramatically to time and outcome efficiency, while also retaining accuracy. The book provides the reader with an adaptive toolbox of heuristic tools and classification systems to aid real-world decisions. Throughout, financial applications are presented alongside real-world examples to help readers solve established problems in finance, including stock buying and selling decisions, when faced with not only risk but fundamental uncertainty.

The book concludes by describing potential solutions to financial problems in the forefront of contemporary debates, and calls for taking psychological insights seriously.

• Demonstrates how well-constructed ‘fast and frugal’ models can outperform standard models in time and outcome efficiency
• Focuses on how financial decisions are made in reality, using heuristics, rather than how such decisions should be made
• Discusses how cognition and the decision-making context interact in producing ‘fast and frugal’ choices that follow ecological rationality
• Explores the development of decision-making trees in finance to aid in decision-making

• Language: English
• Publisher: Academic Press
• Release date: Nov 15, 2019
• ISBN: 9780128124963

William P. Forbes

William Forbes is a Teaching Associate at Queen Mary University of London. Forbes has researched and taught upon behavioural finance for nearly twenty years. Previously, he has worked in Exeter, Manchester, Glasgow and Loughborough Universities. He is the author of Behavioural Finance (John Wiley & Son, 2009), and co-author of Corporate Governance in the United Kingdom: Past, Present and Future (Springer, 2014).

Book preview

Altman.

Part 1

A fast-and-frugal approach to finance

Outline

Chapter 1. Introduction

Chapter 2. Fast-and-frugal heuristics

Chapter 3. Adaptive or efficient financial markets?

Chapter 4. Financial regulations and heuristics

Chapter 5. When fast-and-frugal works best

Chapter 1

Introduction

Abstract

This first chapter introduces some of the potential of a fast-and-frugal finance and why such a recasting of standard finance is needed. It does this partly by returning to some of the precursors of Gigerenzer's fast-and-frugal approach, but also by pointing to some of the key themes that motivate Gigerenzer's own work with the ABC Group. These include the distinction between risk and uncertainty, the distinction between classical and ecological rationality, and the less-is-more principle.

Keywords

Fast-and-frugal; Risk/uncertain outcomes; Classical/ecological rationality; Less-is-more

A fast-and-frugal finance might at last allow an understanding of how context and cognition shape the bewildering array of evidence we observe regarding the operation and efficiency of financial markets and their funding of corporations. This has not been a great decade for Finance as an academic subject or profession. Following the global financial crisis, the ensuing Eurozone and LIBOR rigging crises finance academics are commonly portrayed as either knaves or fools. Shiller and Shiller (2011) ask if finance as a subject is hindered by its excessive interest in the techniques of optimising an individual's self-interest. One aspect of expanding finance theory's focus is adopting a fast-and-frugal approach.

Fast-and-frugal reasoning offers a way out of this decline in finance theory's influence and credibility. It does this by reducing one of the central planks of current finance theory, this is the expected utility model of decision-making under uncertainty, or what Gigerenzer calls classical rationality (Gigerenzer and Goldstein (1996)), to a special case of a more general theory of financial decision-making.

In the classical view the decision-maker's calculative powers are infinite and costless. Information, if it is relevant, should always be used in making a decision. Such heroic assumptions fit in with a broader modelling strategy that we can model investors'/agents' behaviour as if they had unlimited calculative powers, were aware of all possible outcomes, etc. The justification for such heroism being that what matters is a model's predictions not its assumptions (Freidman (1953)).

But often researchers confuse statistical tests that explain data, used to estimate the model well, and true predictive performance based on out–of–sample performance Gigerenzer (2004). Indeed the very function of statistical inference in establishing a maintained hypothesis is often hopelessly confused in social science, as argued below by Gigerenzer and Murray (1987).

1.0.1 A fast-and-frugal alternative to behavioural finance

Fast-and-frugal reasoning, developed by Gerd Gigerenzer of the Max-Planck Institute for Human Development in Berlin and his colleagues in the Center for Adaptive Behavior and Cognition, reject the classical view of rational choice in favour of boundedly rational perspective.

Boundedly rational investors satisfice, rather than optimise, in their decision-making. Satisficing is a word of Northumbrian origin suggesting a choice algorithm which required limited information and calculative skill to achieve an acceptable, if not optimal, outcome. The focus is upon finding choices that yield an acceptable, rather than the best, outcome. This is in part because, following Herbert Simon, fast-and-frugal reasoning casts doubt on the notion of a global, or market-wide, best outcome.

Simon's concept of bounded rationality stresses the evolutionary adaption of our minds to the environment in which they operate. As Arrow (2004) has argued we have to be careful that such a definition does not become so broad as to be meaningless for

there is no general criterion for determining which limit on rationality holds in any given context and the building of a complete theory of the economy is a project for the future.

For this reason our application of statistical reasoning must constrain itself to problems of limited domain, what Leonard Savage calls small worlds (Savage (1954)), within which we can adequately specify the bounds on reasonable behaviour that do apply.

Gigerenzer (2004) reminds us that the process of optimisation does not ensure an optimal, or best, outcome since in most applications it simply identifies the maxima/minima of some continuous, twice–differentiable, performance function. This is fine if the performance criteria studied can bear the weight of these assumptions; but simply blackboard economics if they cannot.

Gigerenzer and Selten (2001a) see the main characteristics of the adaptive-toolbox for decision-making as

First it refers to a collection of rules or heuristics rather than a general purpose decision-making algorithm... Second these are fast, frugal and computationally cheap... Third, these heuristics are adapted to particular environments... This ecological rationality... allows for the possibility that heuristics can be fast, frugal, and accurate all at the same time."

(Gigerenzer and Selten (2001a), p. 9)

Thus we can see Gerd Gigerenzer and the fast-and-frugal reasoning tradition as inheriting the baton of Simon's bounded rationality project. As such we agree with van der Sar (2004) (p. 442) that what is needed by finance scholars and practitioners is a credible alternative to the standard model, that embeds a coherent, credible, view of investor psychology. Professor Gigerenzer's fast-and-frugal reasoning may yet offer such an alternative.

1.1 Context and cognition

Many of us will recall being chastised by our parents for how calculators and spreadsheets have made us lazy in conducting the necessary mental arithmetic to do our shopping or budget for the month. In truth our minds have adapted to an environment in which we have access to a calculating machine in our pocket, mobile phone, or the cloud. In Simon's famous phrase

Human rational behaviour is shaped by scissors whose two blades are the structure of the task environments and the computational capabilities of the actor.

(Simon (1990), p. 7)

While this may all seem rather banal and unworthy of comment it stands in marked opposition to the vast majority of finance theory we might think of. Simon (1959) points out nearly all economic models, including those in standard finance, assume agents are rational in the classical sense. Further these agents/investors inhabit a competitive world that swiftly culls the irrational from their midst. Thus

the classical economic theory of markets with perfect competition and rational agents is a deductive theory that requires almost no contact with empirical data once its assumptions are accepted.

(Simon (1959), p. 254.)

Simon points out that there is a wide variety of problems where the assumptions of classical economics might seem a reasonable assumption, for example in heavily traded foreign exchange markets. But, importantly, there are many legitimate areas of inquiry in financial decision-making and investment practice where the assumptions of classical rationality seem honoured only in their breach.

Simon's early work, summarised in Simon (1959), focused upon areas where classical rationality seemed least likely to apply. Two of these were where perfect foresight regarding future outcomes was impossible and so the agent/investor was forced to predict outcomes and where decision-makers faced goal conflicts and perhaps uncertainty, rather than risk, regarding possible outcomes.

Both these deviations from classical rationality are likely to occur in financial markets. The relatively random evolution of asset prices is a central finding of empirical research in finance (Fama (1970)), while the multiple conflicting motives of investors are well documented Statman (2011).

Simon points out that while in experiments with toy gambles between known pay-offs expected utility theory/classical rationality appears to perform adequately as a description of decision-making under uncertainty problems emerge as soon as it is deployed into more realistic contexts of practical application. Simon (1959) points out that while economists took comfort in the affirmation of experimental results behavioural researchers felt inspired to look for alternatives to classical rationality.

Simon doubted whether classical rationality has much to offer in understanding everyday business decisions and investments.

even with new powerful tools and machines, most real-life choices lie beyond the reach of maximising techniques–unless the situations are heroically simplified by drastic approximations. If man, according to this interpretation, makes decisions and choices that have some appearance of rationality, rationality in real life must involve something simpler than the maximisation of utility or profit.

(Simon (1959), pp. 259–260)

It is from this quest for a simpler decision-making calculus that fast-and-frugal reasoning has emerged. A key concept inherited by fast-and-frugal reasoning researchers from Herbert Simon is the notion of satisficing and choices directed at reaching some aspiration level of wealth or well-being. These ideas have later been formalised by those collaborating within the fast-and-frugal tradition (Selten, 1998).

1.1.1 Context, cognition, and the end of accounting

The global financial reporting system constitutes a world-wide, largely invariant system of financial regulation. The International Accounting Standards Committee¹ (IASC) website proudly proclaims that 144 of 149, or 87%, accounting jurisdictions worldwide require the use of International Financial Reporting Standards, or 75% of G20 members (see https://www.ifrs.org/-/media/feature/around-the-world/adoption/use-of-ifrs-around-the-world-overview-sept-2018.pdf?la=en). The IASC clearly state the IASC's goal and the extent of support for it.

The vision of a global set of accounting standards is supported by other organisations within the international regulatory framework, including the Basel Committee on Banking supervision, the Financial Stability Board, the G 20, the International Monetary Fund... and the World Bank

It appears all good men support the adoption of a common accounting framework, regardless of national economic, social or legal setting. This mission has not gone mostly uncontested within the academic community in accounting (Ball (2016)).

Lev and Gu (2016) present a stark case that the fuel financial reports provide for the allocation of investment capital is contaminated and perhaps reaching the point of irrelevance. The primary source of financial reporting increasing irrelevance to a search for corporate value is its failure to transform itself in accordance with the changing dynamics of corporate value drivers

uniformity deprives accounting of a major force for innovation and rejuvenation – the vital experimentation and evolution that come with diversity... the stagnation of the accounting system and the consequent loss of relevance... can in part be attributed to the absence of any experimentation with new information structures or modes of disclosure, which comes from diversity of reporting across countries or regions.

(Lev and Gu (2016), pp. xix–xx)

Some notion of the scale of the loss of relevance of reported accounting performance for measuring stock market performance is given in Chapter 3 of Lev and Gu's book. Correlating key accounting performance metrics with reported stock prices they find a rapid decline in reported earning's and book value's correlation with contemporaneous corporate market values from the 1950's until now. They note the decline became especially marked from the 1980's as elements of intangible value grew to dominate corporate balance sheets.

1.2 Less is more

Consider two criteria often seen as emblematic of classical rationality in decision-making.

1.  more information, if costless, is better than less when deciding what to do,

2.  the desirability of integrating all information into a single predictive value.

To oppose the first appears to make a virtue of ignorance and to oppose the second seems like simply a way of undermining the first principle. Despite this fast-and-frugal reasoning suggests less is more in the sense that, while ignorance may not be bliss, it may be a vital input into an accurate, robust, decision-making process. This Mies van der Rohe, Bauhaus, phrase has become a defining characteristic of the fast-and-frugal reasoning approach.

To see this consider a cue that depends on recognition of a company or its product. If you have heard of a brand of butter you may be more willing to use it believing it is of high quality, as a known brand. So if you know every conceivable brand of butter this knowledge is no longer useful in discriminating between brands on offer. It is better to simply know the leading, most popular, (best?) brands since this allows you to discriminate better between high and low quality offerings of butter products.

Engel and Gigerenzer (2006) describe the power of heuristic-driven reasoning thus

The double grounding of a heuristic in the human brain and in the environment enables simple heuristics to be highly robust in an uncertain world, whereas complex strategies tend to overfit, that is, not generalize so well to new and changing situations. Less can be more.

Engel and Gigerenzer (2006), p. 4

This opportunity led Simon and his followers to try to design simple decision-making schemata for making near-optimal decisions using both limited information and little calculative effort. A world in which simplicity may prevent us from acting stupidly in a manner that over complicating things often can. Such decision-making schemata, or heuristic tools, must be evaluated according to the market terrain in which they are deployed, as opposed to some all-encompassing, a priori, notion of what is rational.

So the rationality of heuristics is ecological not logical as Gigerenzer and Brighton (2009) put it. So to study ecological rationality is to ask into which environment is a given heuristic best deployed. To understand this consider the standard decomposition of a forecast error.

This decomposition is given in many statistical textbooks and stresses the trade-off between biased and volatile predictive methods (Geman et al. (1992)). Fast-and-frugal reasoning methods simply accept some bias in estimation if it offers a sufficiently compensating reducing in forecast variance. Such a bias heavy, but relatively invariant, prediction method gives rise to the claim that less is more for an appropriately constructed fast-and-frugal decision-making heuristic.

The model that explains the most recent data, on stock returns or mergers, may not be the best model to predict its evolution. Partly this is because of the danger of over-fitting. A model that is broad enough to encompass a wide variety of possible scenarios can also associate itself with a large amount of meaningless noise.

So, while avoiding bias, we want our models to be as flexible as possible, not constraining us to a tunnel vision perspective on the market. But to avoid wildly varying predictions we need to make our models as parsimonious as possible too.

The range of applicability of a fast-and-frugal model of financial decision-making

A distinct danger for the heuristics and bias program is that heuristics become applied so broadly they descend into an almost meaningless generality.

Fast-and-frugal models sit at the high bias/low forecast error variance end of the above mentioned forecast error decomposition. The question is when is this the right modelling strategy to adopt? Fast-and-frugal reasoning relies upon the selection, adoption, and abandonment of simple, tightly specified, decision-making heuristics within localised market conditions to which they are well suited.

We might expect minimising forecast error to be a serious problem in fast–moving environments where most data is not relevant to future events. These sorts of conditions seem to occur quite often in Finance, the fortunes of banks after the crash, or the value of the Euro post-Brexit. Often in financial markets it seems like we are entering a new world and past experience can do little to guide us. The financial future thus appears more uncertain than risky, defying a frequentist perspective on the construction of probabilities of possible outcomes. In such a world a fast-and-frugal finance has much to commend it (Mousavi and Gigerenzer, 2017).

Gigerenzer (2004) regards the program of ecological rationality to be to study of

1.  the heuristics people really use in solving any given class of task,

2.  the structure of the task environment,

3.  what aspects of the task environment the selected heuristic can