Much Ado About Rational Expectations

In the last decade after the 2008 financial crash, the field of neoclassical economics has been impugned by heterodox and casual economists (Chakrabortty, 2013), with the orthodox presumption of rationality generating much agitation. This entire discourse bases itself on the definition of rationality equating to economic agents, including consumers, workers, firms, or the government, aiming to maximise their utility by making decisions based on what they derive the most benefit from in their narrow self-interest.

The primary argument against assumptions of rationality is that humans are oft irrational beings, who are easily coerced by emotional, social and psychological factors, and it is impossible to mathematically quantify their behaviour in order to stereotype their behaviour and appraise the economic consequences. Behavioural economists claim that “agents reason poorly” and “act intuitively” (Tevy Chawwa, 2012) because of the presence of imperfect/asymmetric information, bounded rationality (coalescence of limited ability for the human mind to process and evaluate all information, agents receiving incomplete/unreliable information, and limited time to make decisions, overall indicating that human ability to make fully rational decisions is restricted) and cognitive biases (systematic patterns that lead to agents deviating from rational decision-making). This may lead to the potentially tautological argument that government should intervene with ex-ante mechanisms to prevent irrationality and promote wellbeing, despite the state being vulnerable to the aforementioned flaws too.

The assumption of rational expectations (RE) is a vital microfoundation to DSGE models, which allow us to conceptualise the macroeconomy and forecast impacts of economic policy. One alternative is to presume adaptive expectations (AE) instead, which rules economic agents as naïve and making decisions based on the past, e.g. basing their inflation expectations for the current year on last year’s rate. However, in contrast to the forward-looking nature of RE, the acceptance of AE boils human behaviour down to depending solely on past observations, irrespective of current monetary policy or the state of the economy, which leads down a dangerous path as humans do react to economic events and utilise new information to their benefit. The Lucas critique and Goodhart’s law inform us that any alternative to RE would eventually undo itself when economic policy changes and agents react in a differing manner to before, as they begin to anticipate such events and act upon it in a manner that benefits them, e.g. with adaptive expectations, if inflation is normally at 3% then workers will demand 3% wage increases the following year, but if government expands money supply to a level where inflation is at 6% and workers still demand 3% wage increases, then their expectations have been exploited as their real income decreases — but, if this policy was tried more than a few times, workers would change their behaviour to demand higher % wage increases.

However, presupposing RE would mean that agents know possible future outcomes with a strong economic insight formulated on the grounds of a correct model of the economy, which is an exaggeration considering how information gaps, bounded rationality and rules of thumb lead to humans satisficing with decision-making rather than maximising utility — it should be understood that this is not irrational behaviour, instead an attempt to be as rational as possible when considering the limitations through bounded rationality. Therefore, RE should not be considered as a perfectly viable description of human behaviour, rather as an assumption that provides a benchmark against which real-life human behaviour can be evaluated — similarly to the notion of perfect competition.

To conclude, I believe that any discourse of overthrowing neoclassical economics because of its misunderstanding of human behaviour, and consequentially its failure to properly understand the economy, is misplaced because it evokes a sense that contemporary economics is useless, fanatically dogmatic and ideological — this could not be further from reality for our dynamic and diverse discipline, with simple Google searches for economics papers attesting to this as one can explore research on topics varying from central banking to gender, culture or happiness! Instead, economists should be incorporating some of the valuable insights from behavioural science, e.g. cognitive biases such as heuristics (Ormerod, 2006), into their methodology for modelling as it will ensure that despite being unable to perfectly represent an economic agent’s conduct, we can conceptualise our economy in a more meaningful manner — as long as there is rigorous empirical testing to explore these hypotheses. An example of such research is Xavier Gabaix’s acclaimed paper “A Behavioural New Keynesian Model” (Gabaix, 2018), where he analyses the impact of bounded/incomplete rationale on monetary and fiscal policy in an advanced theoretical model.

Related posts:

Anxiety attack? Ask yourself why to determine what triggered you. Recent failure? Ask why to find the lessons from that experience. Big decision? Ask why to know which path to take. Bad habit? Ask…

Capita partnered with AIR Serenbe, the non-profit artist-in-residence program at Serenbe, to produce a short film featuring eight of America’s leading spoken word artists reflecting on the experience…

Projeto criado na aula de fotografia da professora Márcia Molina, com o propósito criar uma segunda temporada para a série 13 Reasons Why, na qual pessoas que já sofreram bullying são os modelos. As…