The irrationality of the ‘rationality’ assumption

Hugh Oliver


May 23rd, 2013

Rationality is often assumed, yet often imperfect. What fallacies of choice await, given our limitations?

Historically, the term ‘rationality’ has been ascribed various meanings within the sphere of economics.  Typically, rationality has been expressed in terms of the idea that consumers attempt to maximise utility by arriving at an optimal decision in light of a complete set of information relating to the market in which they operate.

That is, the rational person of neoclassical economics opts for the decision that is subjectively best for that person in terms of a given utility function.[1] Consequently, neoclassical reasoning relies heavily on artificial factual assumptions such as perfect information, rather than accepting the reality of limited information and cognitive capacity in making any given decision.[2]

Artificially imposing these neoclassical assumptions on consumer decision making is far from reflective of the impulsive nature of consumer behaviour (indeed, if consumers were rational they would not respond to advertising targeting their emotions – more on this later,) and the dynamic environment in which they reside. The reality is, the economy exists in a perpetual state of flux and motion. The environment is constantly changing and evolving. Tastes and preferences rapidly shift. Information is limited. Cognitive capacity varies for each individual. Human choice is fluid, protean, and unpredictable. It cannot be conceptualised in terms of simplistic patterns.

Consumer decisions are limited in terms of both the availability of information and the multiplicity of environmental variables. Collectively, these limitations operate so as to restrict the the ability of consumers to make the optimal or utility maximising decision that neoclassical economies would suggest is possible.

The line of distinction between an optimal decision and more realistic decision can be conceived in terms of a game of chess. In theory, each player is conferred a large number of strategies and potential moves that they can make at each given turn. In reality, players do not consider every potential move available but rather explore a limited number of moves at each stage in the game. Players act on the basis of incomplete information, as they have neither the time, available information, nor the mechanical capacity to consider all the possible alternatives.[3] The aim of each  choice is not to make an optimal decision in a vaccuum, but rather, obtain an advantage in the taking into account their previous position and their opponent’s last move. [4]


The chess analogy can easily be extrapolated so as to reflect consumer behaviour in an economic environment. Economics assumes consumers have complete preferences, that is, that they are able to rank and compare all possible market baskets that affect their decisions. This unbounded approach to rationality implies there are no limits to consumers and their choices in terms of the available information.

‘Rational’ consumers are able to mechanically process and decipher the abundant information available, and compartmentalise this knowledge into a objective relative ranking of their preferences of the market baskets from most preferred to least preferred. Yet, in reality, the consumer’s mind is highly impressionable, cognitively limited, easily manipulated by extraneous variables, and typically lacks all the information to make a utility maximising decision. The consumer is unlikely to make an optimal decision in light of the information available, but rather, makes a decision on the basis of incomplete information, therein obtaining a temporarily satisfactory as opposed to a permanently optimal result.

To illustrate these principles in practice, imagine a consumer in the market for smartphones. Say a consumer is debating which phone to purchase and is presented with a range of alternatives comprising the iPhone 5, and Samsung Galaxy S4 and the Motorola Droid Razr Maxx HD. In theory, if the consumer is conferred complete knowledge pertaining to the respective features of the three aforementioned phones, an impartial, objective analysis of these features may result in consumer A opting for either the iPhone, Samsung or the Motorola. Choosing a phone based on objective features such as battery life, screen size and camera quality (to name a few), and ordering the phones into an objective relative ranking on this basis, seems to be a logical extension of the utility maximising principle.

In reality however, consumers are regularly exposed to marketing campaigns attempting to differentiate the phones, as well as certain trends displayed by friends and family members who have opted for a certain brand of phone. Take the most common example, namely, the extensive marketing employed by the Apple brand. Through pervasive marketing, ‘Apple’ has been rendered synonymous with simplicity, elegance and modernity in the fabric of the consumer’s mind.

Carefully targeted advertising programs are utilised to convince consumers into artificially augmenting the perceived value of the Apple iPhone relative to its competitors. Even if consumers are aware of the objective differences between the three phones, the subjective value of the iPhone can be enhanced by advertising to the extent that many consumers would forgo the better camera of the Samsung or the longer battery life of the Motorola, so as to acquire the higher perceived value of the Apple brand instead. Consequently, irrespective of the fact that the iPhone may be relatively less effective in terms certain features compared to the Samsung or Motorola, if the consumer perceives the iPhone as superior, they will opt for higher priced Apple product.

Does a consumer really analyse all available and comparable phones on the market when they make a decision? Often, consumers purchase iPhones with minimal consideration of any viable alternatives – possibly because they had no inclination to consider these alternatives or possibly because they were convinced by marketing or popular trends into believing the iPhone was the best phone on the market. By failing to consider alternatives, consumers cannot rank the available phones on the market in order of most preferred to least preferred. By acting on imperfect and incomplete information, consumers do not possess the knowledge that enables them to make the optimal decision in light of all the potential alternatives. Accordingly, the consumer does not always make the optimal or utility maximising decision that the study of economics would lead one to believe is possible.

[1] Simon, H,  ‘The Behavioral Foundations of Economic Theory,’ The Journal of Business, (1986) 59(4), 209-224.

[2] Ibid, 209-224.

[3] C Tisdell, ‘Concepts of Rationality in Economics,’  Philosophy of the Social Sciences, (1975).

[4] Simon, H, ‘The Game of Chess’ (1992)

The views expressed within this article are those of the author and do not represent the views of the ESSA Committee or the Society's sponsors. Use of any content from this article should clearly attribute the work to the author and not to ESSA or its sponsors.

  • Owen Wakely

    Very interesting and well argued piece Hugh. One of the better recent articles on this site. I am an economics major (sophomore year) at Boston College in US and have been following ESSA for a few months courtesy of some of my australian buddies. Did you consider the concept of cognitive bias and also that the optimal decision is the one that satisfies most needs or the highest priority needs? Also i am sure some behavioural psychologists would debate you on some points. I look forward to seeing more from you Hugh and have emailed this on to some of my classmates.

    • Hugh Oliver

      Hi Owen,

      Thanks for taking the time to respond! Although I have not specifically explored the concept of cognitive bias, in retrospect, the idea that consumers are prone to cognitive bias is to some degree implicit in my discussion of consumers in the smartphone market. For example, to the extent that demand for Apple products can be attributed to consumers wanting to fit in within the everyone else (family, friends etc) who owns an apple product, this reflects the bandwagon effect in action. However, I definitely should have included a greater discussion of cognitive biases i.e. post purchase rationalisation, or even confirmation biases i.e. a consumer interpreting information regarding Apple products selectively in a manner which affirms their pre-existing beliefs that Apple products are superior. I completely agree with you that cognitive bias is a further example of consumers behaving in a manner contrary to rational choice theory.

      In terms of your question regarding the optimal decision (correct me if I’m wrong!) are you saying that if a consumer purchases an item which satisfies most/highest priority needs, even if it does not ‘perfectly’ fulfil their needs, it will amount to an optimal decision? The more extreme side of the unbounded rationality model seems more directed towards an ‘optimal’ decision in the sense that the decision made is the best outcome in the light of all the other available decisions. I was referring to ‘optimal’ decision in that sense. If an optimal decision was one capable of satisfying most needs or the highest priority needs, it would be possible for a consumer to make two or more decisions satisfying most needs/highest priority needs and therein have potentially two or more optimal decisions available. On the other hand, strictly speaking, the unbounded rational consumer is capable of making the ‘optimal’ decision. However, I agree with you that the optimal decision that satisfies most needs or the highest priority needs is a more accurate reflection of optimal decision making in reality. Hope this answers your question!

      • Owen Wakely

        Thanks for the considered response Hugh. I think, and would argue, that the degree to which we can predict consumer behaviour is product dependent. By that i mean whether the product is a durable, high cost purchase such as a car, or is a financial product such as a mortgage as opposed to consumer products with short and shorter product life cycles such as smartphones or other consumer electronic devices. I think its the latter category that is providing the greatest challenge and falls, i argue, well outside our ability to predict or “rationalise” decisions. Having said that of course you will be familiar with the theory of Diffusion of Innovations and the modelling of consumer adoption of new technologies and the notions behind that theory in terms of how decisions are made. I have a particular interest in consumer adoption of new technology. Roger and Everett’s original work from the early 1960s might be of interest to you. Cheers Owen

        • Hugh Oliver

          Thanks Owen, I’ll be sure to check it out! Again, I agree with you with respect to consumer decision making for products with short life cycles. In fact, I was thinking that one way of considering why consumers are less able to ‘rationalise’ decisions in the context of technological goods with shorter product life cycles can be expressed in terms of a general principle. By assuming that an individual’s potential ability to rationalise a decision is function of cognitive capacity, quality of available information and time, holding cognitive capacity and quality of available information constant, an increase in the time to make a decision should theoretically increase in an individual’s potential ability to make an ‘optimal’ decision. One possible explanation for this is that given more time, consumers are better able to weigh up the costs and benefits of any given decision. Conversely, given less time, consumers are less capable of analysing the costs and benefits of a particular good.

          In turn, it could be argued that the time taken to make any decision is product dependent. Thus, for a good such as a smartphone with a short life cycle, comprising a relatively smaller proportion of an individual’s income, less time will typically be spent in deciding which smartphone to purchase. If a consumer spends less time weighing up the available information when purchasing a smartphone, they are less able/likely to make a ‘rational’ decision. Another, more vivid example, is when consumers purchase supermarket items on impulse. Items such as chewing gum, confectionary, and soft drinks are always placed near checkouts, so as to induce consumers into buying without thinking, even if they don’t need the items on display. Again, it becomes clear that given less time, consumers would appear less able to ‘rationalise’ a decision. Conversely, in the case of a high cost purchase item such as a house or a car, given these items constitute a higher proportion of an individual’s income, it is likely that consumers will expend more time in arriving at a decision, thus ensuring consumers are more able and likely to weigh up the cost and benefits. In turn, a consumer who conducts a more thorough analysis of the costs and benefits of a decision is, in theory, more able to discern whether or not the decision is a viable investment.

          Secondly, if an individual has higher quality information available, this should theoretically improve the ability of the consumer to rationalise any given decision (holding other variables constant for the moment). Thus, returning to the example of smartphones, consumers often base their decisions on imperfect or heavily biased information such as advertising. In turn, it would appear that given smartphones are a ‘relatively’ smaller proportion of a person’s income in comparison to say a house, the consumer is more likely to rely on such imperfect information, and therein, be less able to ‘rationalise’ which phone on the market is most conducive to their actual needs. Conversely, in the case of purchase such as a car or house, given these items comprise a relatively higher proportion of a consumer’s income, it is more likely that consumers will take a greater number of steps in acquiring higher quality information so as to ensure they make the best possible decision. This may involve a consumer seeking out impartial evaluations from independent individuals in the property market etc. In short, the consumer would appear more inclined to obtain better quality information, which in turn, should lead to a more informed and therein ‘rational’ decision.

          Finally, a consumer less prone to cognitive biases, and therefore, having a better cognitive processing ability, should be able to ‘rationalise’ decisions better than an individual susceptible to cognitive biases. There are a range of cognitive biases which would appear to be more common in markets such as the smartphone market (bandwagon effect, confirmation biases etc) in contrast to markets such as housing, wherein, biases such as the bandwagon effect would probably be less pronounced. Thus, the greater number of cognitive biases likely to affect consumers in the smartphone market would further detract from a consumer’s ability to ‘rationalise’ decisions.

          Ultimately, an increase in any of these variables should increase a consumer’s potential ability to ‘rationalise’ any given decision. However, in reality, as I have argued in my article, there is a danger in drawing generalisations and relying too heavily on simplifying assumptions and models. Thus, while it would appear that given more time, or exposed to more information, or endowed with a higher cognitive processing ability, an individual would seem to have a greater potential to make a more ‘rational’ decision, it is unlikely such a model of consumer behaviour would translate seamlessly into practice. Consumers, even with lots of time, high quality information, and a high cognitive processing ability, may still make ‘irrational’ decisions. Accordingly, although it appears to be a reasonable assumption that consumers are less able to ‘rationalise’ decisions in the case of products with shorter life cycles (due to less time spent in making a decision, or possessing lower quality information, or being more prone to cognitive bias), this principle will never hold universally. There will always be consumers who are still able to make ‘rational’ decisions in the market for smartphones, technological goods etc, and on the hand, there will always be consumers who make relatively less ‘rational’ decisions in terms of high cost purchases, i.e. the housing market (GFC Housing Bubble etc).

          Sorry for the mini essay!

          • Owen Wakely

            Hi Hugh. Yours is a very thoughtful analysis of what is undoubtedly a complex issue. I encourage you to review some of the recent reviews of how innovation becomes distributed in the marketplace to expand on your comprehensive insights. I look forward to reading more articles by you Hugh. Cheers, Owen

          • Hugh Oliver

            Thanks Owen. Will Do!

  • Joey Moloney

    Great article Hugh.

    Couldn’t agree with you more. Behavourial economics is developing into a great add-on kit to mainstream economics. It always bothered me that a lot of economic models are built on rationalist axioms that don’t stack up empirically. The ‘satisficing’ consumer operating by ‘rule-of-thumb’ seems to be a much more realistic agent than the perfectly rational, super-computer-like optimizer that is often assumed.


    • Hugh Oliver

      Thanks Joey,

      I agree with you that the rationality assumption employed in economics lacks a firm empirical basis. On the other hand, I guess the perfectly rational consumer assumption will never be displaced as it ensures consistency in economic theory and simplifies complex behavioural patterns in a manner more conducive to economic modelling.

  • Vanessa Cheung

    I enjoyed reading your article Hugh. I agree with your views on the difficulty in predicting behaviour based upon the assumption of rationality. They are spot on. I guess the advertising industry as we know it taps into the emotional rather than the rational to effect our behaviour and if the rationality argument held there would be no market for advertising (simply boring infomercials!)


    • Hugh Oliver

      Thanks Vanessa,

      Good point – I guess it is good for some businesses that consumers are not perfectly rational, otherwise marketing would be ineffectual!

  • Dave Sharpe

    Cool article Hugh. Congrats on the “debut” contribution. Human behaviour and choice by their very nature are predictably irrational.

    • Hugh Oliver

      Thanks Dave!

      • Dave Sharpe

        keep them coming HUgh..i look forward to your next article. Glad to see you generated some interesting debate and discussion. Cheers Dave

      • Muhammad Rafiq

        But then how the consumer behavior can be modelled ?? It is indeed very interesting article

Founding sponsors




Gold sponsors


Silver sponsors