Ha-Joon Chang - Economics - The User's Guide

Institutions are not just constraints: contributions and limitations of the New Institutional Economics

Deploying the concept of transaction cost, the NIE has developed a wide range of interesting theories and case studies. One prominent example is the question as to why, in a supposedly ‘market’ economy, so many economic activities are conducted within firms. The (simplified) answer is that market transactions are often very costly due to the high cost of information and contract enforcement. In such cases, it would be much more efficient if things were done through hierarchical commands within the firm. Another example is the analysis of the impacts of the exact nature of property rights(the rules on what owners can do with which kinds of property) on patterns of investments, choice of production technologies, and other economic decisions.

Despite these very important contributions, the NIE has a critical limit as an ‘institutionalist’ theory. It sees institutions basically as constraints – on unfettered self-seeking behaviour. But institutions are not just ‘constraining’ but can also be ‘enabling’. Often institutions limit our individual freedom exactly in order to enable us to do more collectively – traffic rules, for example. Most members of the NIE would not deny the enabling role of institutions, but by not talking about it explicitly and continually referring to institutions as constraints, they convey a negative impression of institutions. More importantly, the NIE fails to see the ‘constitutive’ role of institutions. Institutions shape the motives of individuals and do not merely constrain their behaviour. Missing out on this critical dimension of what institutions do, the NIE falls short of being a full-blown institutional economics.

One-sentence summary: We are not smart enough, so we need to deliberately constrain our own freedom of choice through rules .

The Behaviouralist school is so called because it tries to model human behaviours as they actually are, rejecting the dominant Neoclassical assumption that human beings always behave in a rational and selfish way. The school extends this approach to the study of economic institutions and organizations – for example, how best to organize a firm or how to design financial regulation. The school thus has a fundamental affinity, and some overlap in membership, with the Institutionalist school.

The Behaviouralist school is the youngest of the schools of economics that we have so far examined, but it is older than most people think. The school has recently come to prominence through the fields of behavioural finance and experimental economics. But it has its origins in the 1940s and the 1950s, especially in the works of Herbert Simon (1916–2001), the 1978 Nobel economics laureate. [103] Simon was the last Renaissance Man, as I call him in Thing 16 of my book, 23 Things They Don’t Tell You about Capitalism . He made path-breaking contributions not just in economics but across many fields. He was one of the founding fathers of artificial intelligence (AI) and of Operations Research (OR, a branch of business administration). He also wrote one of the classics in the field of public administration ( Administrative Behaviour , published in 1947) and was a leading scholar in cognitive psychology. So he knew a thing or two about how people think and act.

Limits to human rationality and the need for individual and social rules

Simon’s central concept is bounded rationality. He criticizes the Neoclassical school for assuming that people possess unlimited capabilities to process information, or God-like rationality (he calls it ‘Olympian rationality’).

Simon did not argue that human beings are irrational. His view was that we try to be rational but that our ability to be so is very limited, especially given the complexity of the world – or given the prevalence of uncertainty, if you want to formulate it in the Keynesian way. This means that often the main constraint on our decision-making is not the lack of information but our limited capability to process the information we have.

Given our bounded rationality, Simon argued, we develop mental ‘shortcuts’ that allow us to economize on our mental capabilities. These are known as heuristics(or intuitive thinking) and can take different forms: rule of thumb, common sense or expert judgement. Underlying all these mental devices is the ability to recognize patterns, which allows us to abandon a large range of alternatives and focus on a small, manageable but most promising range of possibilities. Simon often used the chess masters as an example of someone using such a mental approach – their secret lay in their abilities to rapidly eliminate less promising search paths and converge on a sequence of moves that are likely to yield the best outcomes.

Focusing on a subset of possibilities means that the resulting choice may not be optimal, but this approach enables us to handle the complexity and the uncertainty of the world with our bounded rationality. Therefore, Simon argues, when they make their choices, human beings satisfice, that is, we look for ‘good enough’ solutions rather than the best ones, as in the Neoclassical theory. [104] Some Neoclassical economists have tried to reinterpret bounded rationality so that it fits into optimization models. Some argue that bounded rationality simply means that we need to see economic decision as the ‘joint-optimization’ of resource costs (a traditional Neoclassical concern) and the costs of decision-making. In another common reinterpretation, people are seen to optimize by choosing the best decision rules, rather than trying to make the right choice in every decision instance. Both these reinterpretations in the end do not work because they assume even more unrealistic levels of rationality than the standard Neoclassical model does. How can agents that are not even rational enough to optimize on one front (resource costs) optimize two (resource costs and decision costs)? How can agents that are not smart enough to make rational decisions on individual occasions design decision rules that will allow them to make optimal decisions on average?

Market economy vs. organization economy

Even though it starts with the study of individual decision-making, the interest of the Behaviouralist school stretches much further. According to the school, it isn’t just at the individual level that we build simplifying decision rules that help us operate in a complex world with our bounded rationality.

We build organization routinesas well as social institutions so that we can compensate for our bounded rationality. Like heuristics at the individual level, these organizational and social rules restrict our freedom of choice but help us make better choices because they also reduce the complexity of the problem. Particularly emphasized is the fact that these rules make it easier for us to predict the behaviour of other related actors, who would follow those rules and behave in particular ways. This is a point that the Austrian school also emphasizes using slightly different language, when they talk about the importance of ‘tradition’ as the basis for reason.

Adopting the Behaviouralist perspective, we begin to see our economy in a way that is very different from the dominant Neoclassical one. The Neoclassical economists usually describe the modern capitalist economy as the ‘market economy’. The Behaviouralists emphasize that the market actually accounts for only a rather small part of it. Herbert Simon, writing in the mid- 1990s, reckoned that something like 80 per cent of economic activities in the US happen inside organizations, such as the firm and the government, rather than through the market. [105] H. Simon, The Sciences of the Artificial , 3rd edition (Cambridge, MA, The MIT Press 1996), p. 31. He argued that it would be more appropriate to call it the organization economy.