Department of Economics - University of Massachusetts at Amherst Abstract
This paper evaluates the economic theories developed by Keynes in the light of recent research in behavioral and experimental economics. We found that many of the ideas set forth by Keynes in his economic works, especiallyin the The General Theory, have a sound behavioral foundation and fit broadly the actual behavior of economic agents in the real world. As a consequence, his views about the psychological processes causing, among others, the instability of capitalism, the speculative nature of financial markets, and the necessity of government intervention, are plausible and underscored by evidence concerning individual decision-making under uncertainty.
Introduction Along with other classical economists, like Adam Smith in the Theory of Moral Sentiments, and Irving Fisher in The Theory of Interest, Keynes’s work constantly emphasized the importance of psychological factors in human decision-making, and how these factors could change the analysis of economic issues. In his major philosophical work, A Treatise on Probability, he touched upon several concepts that would change the classic frequentist view of the judgment of probabilities, stressing the necessity of explicitly considering psychology to improve probability theory. By the same token, in The General Theory, ideas like wage rigidity, animal spirits, money illusion, conventions, and uncertainty suggest that the behavioral assumptions of neoclassical economics, with the imposition of rationality as the decisive criterion, i.e., obeying some specific axioms, do not conform to how people actually behave.
That Keynes paid substantial attention to the role of psychological factors when constructing his economic theories is a well known fact in the history of macroeconomic thought. George Akerlof, for example, remarked that the current development in behavioral macroeconomics has its roots and it is, to a certain extent, a continuation of Keynes’ project. He argues that (2002, p. 411):
“That dream was the development of a behavioral macroeconomics in the original spirit of John Maynard Keynes' General Theory (1936). Macroeconomics would then no longer suffer from the "ad hockery" of the neoclassical synthesis, which had overridden the emphasis in The General Theory on the role of psychological and sociological factors, such as cognitive bias, reciprocity, fairness, herding, and social status. My dream was to strengthen macroeconomic theory by incorporating assumptions honed to the observation of such behavior.”
To the best of our knowledge no single study to this date has explicitly laid bare the links between psychology and the economics of Keynes, providing textual evidence on his insights concerning the behavior of economic agents that could be directly connected with behavioral studies. This is probably due to the relatively recent appearance of behavioral and experimental economics, the confusing treatment Keynes gave to psychology, the exclusive focus on aggregate relationships in the traditional studies of Keynes and, obviously, the fetters imposed on economic analyses by the behavioral postulates of neoclassical economics. For instance, although Akerlof recognizes the importance of Keynes as an inspiration for the current studies of behavioral macroeconomics, he does not show how Keynes addressed those issues, or how behavioral economics would tackle them.
This paper is an attempt to fill this gap, seeking to find in the works of Keynes themselves the hints and suggestions about what a realistic approach to behavior under uncertainty might be. We claim that there is strong evidence that Keynes was deeply conscious about the necessity to incorporate realistic behavioral assumptions in macroeconomic models that deals with judgment under uncertainty. Moreover, we found that indeed his research program is broadly compatible with and finds support in most of the recent findings of behavioral and experimental economics, with important consequences for macroeconomic theory and policy. A large number of experimental works has been done in areas that Keynes considered important in his General Theory. The most important psychological findings discussed in this paper regarding the economics of Keynes are the status quo bias, money illusion, money wage rigidity, gift exchange, overconfidence, conformity, the generalized use of heuristics and their corresponding biases under situations of uncertainty, and prospect theory.
The work is organized as follows: after this introduction, we will briefly discuss the different interpretations and controversies surrounding Keynes’s ideas in order to contextualize and advance our interpretation. The following section will provide a comprehensive comparison of the main behavioral tenets of Keynesian theory with the most updated findings of the behavioral and experimental economics. The next section will consider the consequences for macroeconomic theory. The final section concludes the essay.
Controversies and Interpretations of Keynes’s Approach to Economics Since the General Theory, and even before, the theories put forth by Keynes have always been surrounded by controversies, which is not surprising considering the far-reaching impacts of his work. The IS-LM interpretation, the consumption function, the role of expectations, the real balances effect, the fundamental uncertainty, the liquidity preference, the influence of the Treatise on Probability on the General Theory, to mention just a few, have all received a large deal of attention, both in heterodox and in orthodox circles. To propose an interpretation of Keynes’s economic theories in terms of individual behavior, emphasizing the psychological dimension is no less controversial, for several reasons.
The first objection is obvious. In stressing the role of psychology we must consider the role of individuals in Keynes’s macroeconomic theories. There are no detailed methodological discussions in the works of Keynes in general, and in The General Theory in particular, that would enable one to claim that Keynes clearly avowed his commitment to the methodological individualism. Winslow (2003) claims that Keynes rejected atomism and embraced an organic approach. However, he argued that it is still possible to consider Keynes as an individualist (p. 156, fn 5, italics in the original):
“Atomic individualism needs to be distinguished from individualism per se. Much writing on methodology, for example, on so-called ‘methodological individualism’, implicitly and mistakenly identifies individualism with atomic individualism. Keynes, though he abandoned atomic individualism, remained philosophically an ‘individualist’ in the sense of ‘Paley’s dictum that “although we speak of communities as of sentient beings and ascribe to them happiness and misery, desires, interests and passions, nothing really exists or feels but individuals”’.
Anna Carabelli (2003, p. 218), also supports this view:
“For Keynes, then, the material, or the object of economics, were the beliefs, the opinions of economic agents. Intentionality, motives and human agency, on this view, are the material of economics.”
There remains the question, obviously, of the aggregate behavior of the economy, which cannot be reduced to a sum of individual behaviors. We will not deal with this issue here, but will accept Winslow’s and Carabelli’s characterization as a valid distinction.
The second objection is about the rationality of individual behavior. In terms of judgments under uncertainty, this is usually addressed in terms of the rationality of conventional behavior. According to Keynes, in situations of uncertainty, economic agents use conventions as useful guides to action, supported by their higher or lower degree of confidence (or weight of argument) in those conventions (Crotty, 1994). David Dequech (1999) analyzes the different uses of the concept ‘convention’ in the post-keynesian economics literature. These range from something that structures individual expectations, to individual or collective rules-of-the-thumb that lead to a convergence of beliefs. Dequech then discusses the different arguments employed to defend the rationality of conventional behavior. However, we believe that this discussion is a deadlock. Even if one considers that rationality for Keynes means ‘reasonable’ (Meeks, 2003), these debates have an intrinsic normative bias, trying to adjudicate between different behaviors in terms of what is a value judgment.
In this regard, Michele Baddeley (1999), discussing Herbert Simon, distinguishes between substantive and procedural rationality. The first is the rationality employed by actors in neoclassical models and the second the reasonable rationality used by individuals in the real world. It is worth noticing that there is a direct relationship between the concept of rationality employed in developing theories and the role ascribed to psychology. Baddeley argues that (op. cit., pp.197-198):
“Simon (1979) comments that the substantive rationality approaches underlying these orthodox concepts ‘freed economics from any dependence upon psychology (…) Keynes’s emphasis on the subjective determinants of investment, the limits to quantification and the role of conventional behavior, fits broadly into a procedural description of rationality. In contrast to the orthodox analyses based on substantive rationality assumptions, Keynes argues that scientific theories should be able to cope with real-world situations and should not force the facts to conform with theoretical assumptions.”
We take Simon’s distinction as a suitable one to the purposes of this paper, without further getting involved into these debates.1
The third potential opposition concerns the very realm of our interpretation, namely, the explicit importance of psychology in Keynes’s works. To refute this objection, we will consider three approaches. First, we will take the comments of Keynesian scholars on the issues of psychology and behavior in Keynes’ work. Second, we will propose an exegesis of Keynes’s texts. Finally, in the next section we will compare some passages in Keynes’s works that overtly treats of individual behavior with the most important findings of behavioral and experimental economics up to this date.
The fundamental importance attributed by Keynes to psychological elements in his economic theory has not passed unnoticed in the literature. For instance, the Nobel laureate James Tobin (1980, p. 28) utters that:
“More serious, perhaps, was his [Keynes’] insistence that the marginal efficiency of capital is as much psychological as technological.”
Winslow (2003) argues that the behavioral postulates of neoclassical economics (op. cit., p. 143):
“(…) has so entrenched the set of philosophical and psychological foundations from which Keynes escaped.”
And about Keynes’s theories (ibidem):
“Both its philosophical and its psychological foundations differ radically from those now dominant. He combined these very different philosophical foundations with very different psychological foundations.”
Sheila Dow (2003, p. 210) argues that:
“The General Theory incorporates key features of human behavior under uncertainty.”
And that (op. cit., p. 212):
“His reservation about mathematical formalism, his reference to psychology and social convention as essential elements of behaviour – all follow from his understanding of the real, social world as complex and evolving and incapable of yielding much in the way of certain knowledge.”
Carabelli claims that (op. cit., p. 223):
“(…) Keynes allowed a role for both psychological and subjective influences on individual judgment (...)”
And, quoting Keynes, she further emphasizes her point (op. cit., p. 218):
‘“Economics deals with motives, expectations, and psychological uncertainty. One has to be constantly on guard against treating the material as constant and homogeneous. It is as though the fall of the apple to the ground depended on the apple’s motives, on whether it is worth while falling to the ground, and whether the ground wanted the apple to fall, and on mistaken calculations on the part of the apple as to how far it was from the centre of the earth’.”
Baddeley claims that (op. cit., p. 198):
“In Keynes’ approach, psychological forces play a crucial role in his analysis, and Keynes argues that it is psychology of actual behavior which vitiates classical and neoclassical analyses (…) According to Keynes, when there is no basis for rational belief, behavior is dictated by psychological motivations and non-rational forces, such as animal spirits and conventions”
And (op. cit., p. 209):
“His contribution to the understanding of the psychology of investor behavior is nonetheless crucial…If beliefs are irrationally based, or if there is no rational basis for belief, action may be the outcome of purely psychological motivations.”
Davis (2003, p. 199-200), advancing his own interpretation of convention, considers that Keynes’s treatment of conventions was explicitly in terms of psychological propensities:
“ (…) these psychological propensities manifest themselves in varying degrees in different individuals, and thus it is more useful and more informative to say that Keynes’s interest in conventions was ultimately directed toward explaining how conventions act to structure different individuals’ psychological propensities and attitudes in relation to one another, or alternatively how conventions relate the degrees to which psychological propensities and attitudes operate across different individuals.’
Gerrard (2003, p. 242) claims that Keynes sought to incorporate psychological elements in his approach to probability:
“He sought to encompass his earlier, more rationalistic and academic thought within a more psychological and practical framework.”
He also maintains that psychology was necessary to push Keynes approach toward a more advanced level in his theory of probability (op. cit., p. 243):
“Keynes considered Ramsey to have clarified the limitations of the logical theory of probability and to have pointed the way towards the next area of enquiry, namely the psychological and practical aspects of human behavior under uncertainty.”
In his view, psychology was also an important dimension in the building of an alternative economic theory (op. cit., p. 238):
“According to Keynes, then, the doctrine of mathematical expectation is inadequate as an explanation of human behaviour under conditions of uncertainty. He argued for the need to develop a more general theory of behaviour under uncertainty.”
Gerrard summarizes his interpretation suggesting that (op. cit., p. 239-240, his emphasis):
“The basic argument is that Keynes has provided the outline of a theory of the effects of fundamental uncertainty on economic behavior, in stark contrast to the characteristic assumption of mainstream economic theory that agents possess perfect or near perfect knowledge of the consequences of their actions.”
It is clear that, according to these readings of Keynes’s works, he placed a large emphasis on psychological matters. And this is not an overstatement. Keynes himself refers several times to the psychological aspects underlying his theory. Sentences such as “psychological laws”, “psychological effect”, “psychological propensities”, “psychological influences”, “psychological characteristics”, “psychology of the community”, “psychological motives”, are extensively deployed in the General Theory, mainly regarding the influences on the marginal propensity to consume and the multiplier. In chapter 18, where Keynes summarizes his general theory of employment, there is explicit reference to ‘fundamental psychological laws’ and the fundamental role these laws played in stabilizing the economic system. Some passages from the Quarterly Journal of Economics article and from the General Theory show the importance Keynes attributed to behavioral issues:
From the QJE (1937, pp. 215 and 222, respectively):
“Perhaps the reader feels that this general, philosophical disquisition on the behavior of mankind is somewhat remote from the economic theory under discussion. But I think not. Tho (sic) this is how we behave in the market place, the theory we devise in the study of how we behave in the market place should not itself submit to market-place idols.”
“The hypothesis of a calculable future leads to a wrong interpretation of the principles of behavior which the need for action compels us to adopt, and to an underestimation of the concealed factors of utter doubt, precariousness, hope and fear.”
From the General Theory there are some quotes that undeniably suggest the existence of profound behavioral elements in Keynes theory, and that psychology is a major factor in his macroeconomic theory (op. cit., pp. 217, 246-247, 250, and 251, respectively):
“(...)a marginal efficiency which is at least equal to the rate of interest for a period equal to the life of the capital, as determined by psychological and institutional conditions.”
“Thus we can sometimes regard our ultimate independent variables as consisting of (1) the three fundamental psychological factors, namely, the psychological propensity to consume, the psychological attitude to liquidity and the psychological expectation of future yield from capital-assets (…)”
“Now, since these facts of experience do not follow of logical necessity, one must suppose that the environment and the psychological propensities of the modern world must be of as such character as to produce these results. It is, therefore, useful to consider what hypothetical psychological propensities would lead to a stable system; and, then, whether these propensities can be plausibly ascribed, on our general knowledge of contemporary human nature, to the world in which we live.”
“Our first condition of stability…is highly plausible as a psychological characteristic of human nature.”
However, recognizing the importance of psychological factors in Keynes work is not enough. Keynes himself did not undertake a deep analysis of these psychological factors, perhaps assuming those to be so evident - or so complex - features of the real world that a detailed treatment was not necessary - or very difficult. Keynes is also ambiguous about some issues. In chapter 15 of the General Theory, for instance, he considers the interest rate to be a ‘highly psychological phenomenon’, just to assert a few paragraphs below that the interest rate is ‘a highly conventional phenomenon’. This lack of a more rigorous treatment, understandable in terms of the problems he was dealing with, led to a large number of disparated interpretations. Baddeley remarks that (op. cit., p. 198):
“In both A Treatise on Probability and The General Theory, Keynes treats psychology as, in some sense, the contrary of rationality. However, Keynes does not distinguish adequately between mass psychology and conventional behavior and at times he seems to treat them as distinct forces.”
And that (op. cit., p. 199):
‘This fuzziness in Keynes’ ideas about rational versus conventional versus psychological forces has led to the development of divergent interpretations of his analysis.”
If the conceptual treatment of psychology and conventional behavior is not crystal clear in Keynes’s works, his depiction of real world behavior is consistent with empirical evidence brought about by behavioral and experimental economics, and also with research in psychology not directly related to economics. And although at the time Keynes was writing his General Theory it was possible to claim that (op. cit., p. viii):
“It is astonishing what foolish things one can temporarily believe if one thinks too long alone, particularly in economics (along with the other moral sciences), where it is often impossible to bring one’s ideas to a conclusive test either formal or experimental.”
Today there are a large number of empirical studies about decision-making under different circumstances, highlighting the most important aspects of human behavior. If, according to Keynes, (op. cit., p. 147):
“(…) our conclusions must mainly depend upon the actual observation of markets and business psychology.”
It is possible to show that his conclusions are, by and large, broadly supported by the actual behavior of market participants and observation of business psychology, as the next section will demonstrate.
Behavioral Economics and the Economics of Keynes Main Findings Our third approach explores several features of the behavioral economics literature that are directly related to Keynes’s works, in particular The General Theory. All these works are directly related to decisions that are made under conditions of fundamental uncertainty, which is a very controversial concept in Keynesian Economics. Albeit this is controversial, we think that the definition of uncertainty used in behavioral economics and psychology viz., individuals do not know the probabilities of future events and their respective distributions for the judgments and decisions to be made, is adequate and close to the meaning of complete ignorance used by Keynes.
The purpose of this section is to identify the elements that corroborate Keynes’s psychological speculations, and show how behavioral economics can help to understand the economy from a Keynesian perspective. Also, we describe what is missing in Keynes’s view of human behavior, and what is lacking in the behavioral literature to deepen Keynes’s analysis.
Heuristics The first point to be analyzed in the economics of Keynes is the use of heuristics in decision making. During the last fifty years, the view that human beings act rationally has been constantly challenged. In 1955, Herbert Simon argued that people do not always seek to optimize. Instead, they rely on a series of rule of thumbs that act as a type of “satisficing” process. Then, a group of psychologists, in special Daniel Kahneman and Amos Tversky, launched in the seventies a research project on human judgment based on what they have called the “Heuristics and Biases Approach”. The evidence of their experimental studies is that people are in fact far from maximizing their actions. Instead, they rely on these heuristics, which are defined as fast decisions, using particular and simplifying rules of thumb. In general, they are very useful in providing guides for action and judgment of complex situations under uncertainty. Nevertheless, the dependence on these heuristics might cause systematic deviations from the definition of rationality. This article absorbs this approach, arguing that the analysis of heuristics provides a framework for understanding systematic economic behavior, at the same time allowing the investigation of other effects as consequences of peoples’ simplifying strategies.
From this literature, the three heuristics that caused most attention are:
Representativeness – Judgments of the likelihood of an event are based on how representative this event is within a class of events. To asses the likelihood of an object A belonging to class B, people compare how similar A is of B, that is, the degree to which A resembles B (Tversky and Kahneman, 1974). As an example, consider an individual who has been described by a former neighbor as follows: “Steve is very shy and withdrawn, invariably helpful, but with little interest in people, or in the world of reality. A meek and tidy soul, he has a need for order and structure, and a passion for detail.” How do people assess the probability that Steve is engaged in a particular occupation form a list of possibilities (for example, farmer, salesman, airline pilot, librarian, or physician)? It is clear that Steve’s description is very representative of a librarian, and this is actually the occupation chosen by the majority of the people as the most likely for him. Nevertheless, the number of librarians compared to salesmen, for example, is very small. This base rate analysis suggests that Steve is indeed more likely to be a salesman than a librarian, even if he is not representative of this occupation. Along with insensitivity to prior probability of outcomes (base-rate fallacy), other biases that are associated with this heuristic are: b) insensitivity to sample size ; c) misconception of randomness; d) insensitivity to predictability; e) illusion of validity; and f) misconceptions of regression to the mean (see Tversky and Kahneman, 1974 for a detailed description).
Availability - Situations in which people assess the frequency of a class or the probability of an event by the ease with which instances or occurrences can be brought to mind (ibid, 1974). For instance, people think that homicides, which are highly publicized, are more common than suicides, but actually the opposite is true. As homicides are much easier to retrieve than suicides, they appear to be more frequent in people’s minds.
Anchoring and Adjustment – Estimates that people make by starting from an initial value and then doing some adjustment to arrive at the final answer (ibid, 1974). As Slovic and Lichtenstein (1971) state, the initial value may be suggested by the formulation or some preliminary computation. In either the case, adjustments are insufficient, causing bias to the estimate.
There are several passages in The General Theory in which Keynes explicitly assumes that people are not able to maximize, and therefore some kind of “mental habit” is used to overcome the problem (1964, p. 51, italics added):
“For, although output and employment are determined by the producer’s short-term expectations and not by past results, the most recent result usually plays a predominant part in determining what these expectations are. It would be too complicated to work out the expectations de novo whenever a productive process was being started; and it would, moreover, be a waste of time since a large part of the circumstances usually continues substantially unchanged from one day to the next.”
“It would be foolish, in forming our expectations, to attach great weight to matters which are very uncertain. It is reasonable, therefore, to be guided to a considerable degree by the facts about which we feel somewhat confident, even though they may be less decisively relevant to the issue than other factors about which our knowledge is vague and scanty. (…) our usual practice being to take the existing situation and to project it into the future, modified only to the extent that we have more or less definite reasons for expecting a change.” (1964, p.148)
Notice how the three heuristics stated above can explicitly be applied to these cases: by using the most recent result, or how confident she is, the entrepreneur uses the information that is more representative of the situation and the information that is easiest to retrieve (availability). After absorbing this information, the final decision will be adjusted according to the entrepreneur’s expectations. These rules might work fairly well in general, but they also lead to several biases that are not expected by the decision maker. In any case, the Rational Expectations Approach is rejected as a meaningful explanation of individual behavior, and laissez-faire macroeconomic policies based on issues such as monetary and fiscal credibility should be challenged. According to the findings of behavioral macroeconomics, government intervention may have a stabilizing role.
In different circumstances, Keynes uses examples that suggest the use of different heuristics (1978, p. 57):
“As time goes on I get more and more convinced that the right method in investment is to put fairly large sums into enterprises which one thinks one knows something about and in the management of one thoroughly believes.”
Here, Keynes invokes the Recognition Heuristic as a fast and frugal rule of thumb to invest. This heuristic is defined as follows: “consider the task of inferring which of two objects has a higher value on some criterion (e.g., which is faster, higher, stronger). The recognition heuristic for such tasks is simply stated: If one of two objects is recognized and the other is not, then infer that the recognized object has the higher value."
(Goldstein and Gigerenzer, 1999).
Thus, the evidence that people use heuristics to make judgments and decisions can be extended to the insights brought by Keynes in his work. Using this approach, it is possible to understand human actions as systematic behaviors instead of a set of random actions that cannot be analyzed consistently. We believe that heuristics form the basis of judgments under uncertainty, and that they provide a useful tool to analytically explain economic outcomes.