Category: Economics

On the Homo Economicus Assumption

As I read more economics, I realized that much of the vitriol between mainstream economists and their challengers, say from the Marxist or heterodox front, is based on the misunderstanding of their own theory and the theory of their opponent.

For instance, consider the oft-mentioned assumption in neoclassical economics, believed by both its proponents and opponents – the homo economicus, the rational, utility-maximizing agent. Supposedly, because agents (consumers and producers) are rational, you will have demand curves sloping downwards with price and supply curves sloping upwards with price.

Now the assumption of a rational agent may seem uncontroversial to neoclassicals. Critiques argue that it portrays people as perfectly rational beings who always make optimal decisions based on self-interest. They argue that the assumption oversimplifies human behavior by disregarding the complexities of emotions, social influences, supposed natural desire for cooperation, and cognitive limitations.

There are mainstream replies, of course, some are cheeky – like altruism can be part of the utility function, and some are useful – like bounded rationality, or that a model’s fitness is not on the realism of its assumptions, but its capacity to predict. Critiques might reply: of course, it can predict things, because capitalism contorts people to become homo economicus. People are advised to be selfish, so competition can exist! And so the debate rages on between economics and sociology departments.

But as I learned later, it is actually a myth that the assumption of homo economicus is needed! Gary Becker, writing “Irrational Behavior and Economic Theory” in 1962, argued that the so the so-called demand curve – the curve of the demand satisfied by supply – will be sloping downwards with price, even if agents have randomized behavior. Instead of rationality, the cornerstone of Becker’s argument is scarcity – the budget constraint. He argued, mathematically, that as long as the budget constraint exists, then budgets will generate an aggregate demand curve that slopes downward.1

I won’t be discussing Becker’s result, but you may want to check out his paper. May be a bit mathy though, but it checks out.

The next question is, where does the budget constraint come from? Mainstream economists treat the budget as an “initial endowment”, which is fair if your focus is on the subsequent exchange and bargaining that leads to a price. But it is also possible to think of it as “income” from a previous transaction, say work or the receipt of dividends from capital.

Here, you can take the neoclassical view and say that the marginal productivity of the factor used to generate that income (say a mix of labor or capital) is your reward, which is your income2, or, you can take the political-economy view that if you are a worker, you are the sole source of the surplus value appropriated by the capitalist (since capital is just bought and consumed at its value), so you deserve both the wages (the cost of your labor power) and the capitalist profits.
In any case, incomes and budgets exist, and that is enough for a downward supply curve. No need for homo economics.

So maybe let’s move on from this and explore other ways to annoy each other.

  1. The story is different for total excess demand curve – the curve of the demand unsatisfied by supply, i.e. shortage. Naturally, you will think that as price increases, the total shortage shrinks. But the Sonnenschein–Mantel–Debreu (1973) theorem shows us that even if consumers are well-behaved and rational, then it is not necessarily the case that the total shortage shrinks as price increases. In any case, this only buttresses the point that homo economicus is unnecessary in neoclassical economics. In fact, its inclusion does not necessarily improve its case. ↩︎
  2. There are results saying that mere random market exchanges guarantee a power law distribution of wealth, or more precisely, a Gibbs distribution – the distribution of energy in an ideal gas in equilibrium. It is actually amazing how literal the reality of market “atomization” is. ↩︎

Flaws in Microfoundations

My last article delved into the mathematics of the modern approach to macroeconomics – reducing macroeconomic behavior as an ensemble of optimizing, intelligent, microeconomic agents – the so-called “microfoundations” approach. New classical economists sees the microfoundations approach as the final bridging of macroeconomics and microeconomics, spurring hopes of a single economic theory that would explain both the individual and aggregate economic phenomena. (note how this parallels physicists’ dream of uniting large-scale relativistic physics with quantum mechanics). This spurred an orientation in economic research and pedagogy characterized by complex mathematical models capturing “deep” parameters in taste, technology, and expectations.

Recently, the microfoundations approach came under attack after models with “deep” microeconomic parameters supposedly failed to predict and recommend effective policy recommendations to mitigate the current global economic crisis. Even recent Nobel Laureate Thomas Sargent – one of the pioneers of modern macro – is under fire. Why this is so – as well as earlier, almost forgotten challenges to the microfoundations approach – is the subject of the survey paper below. Read the abstract and full text:

Abstract

The history of economics, for the most part, has been bifurcated between the study of individual economic decisions (microeconomics) and the aggregate economic phenomena (macroeconomics). The attempt to marry the two, via incorporating “microeconomic foundations” or “microfoundations” to explanations for macroeconomic observations and predictions, has so far taken sway a majority of mainstream economists with the failure of Keynesian models to accurately predict aggregate behavior in the presence of government policy. Robert Lucas Jr. posited that people form “rational expectations” of government policy and act so as to render forecasts unstable.

However, there are some persisting theoretical and empirical challenges on this research direction – the empirical instability of macro-models which incorporated microfoundations, the Sonnenschein–Mantel–Debreu result which may spell the theoretical dead end to economic aggregation, the still unresolved Cambridge capital controversies started by the reswitching argument by Italian economist Pierro Sraffa and American economist Joan Robinson in the 1960s, and the missing “representative consumer or firm” that can take into account the behavior of the aggregate. These challenges give the idea that aggregate economic behavior is almost impossible to deduce from microeconomic behavior of agents. Post-Keynesianism – which asserts that long-term expectations are largely determined by non-economic, psychological processes exogenous to the model – is posited as a possible way forward.

Continue reading “Flaws in Microfoundations” →

Optimal Control in Agent-based Economics

Nobel laureate in Economics Robert Lucas once described economics in terms that would have been familiar to computer scientists:

“This is what I mean by the ‘mechanics’ of economic development – the construction of a mechanical, artificial world, populated by the interacting robots that economics typically studies, that is capable of exhibiting behavior the gross features of which resemble those of the actual world.” – Robert Lucas, Jr., “On the Mechanics of Economic Development”. Journal of Monetary Economics. 1988.

The slide presentation below attempts to survey the use of optimal control theory – a staple in dynamic optimization problems – in economics, one in which markets are populated by hyper-rational, machine-like beings with fixed preferences and ability to calculate in infinite time.

The presentation is also available here: http://www.slideshare.net/jmmiraflor/optimal-control-in-agentbased-economics-a-survey