North, D.C, “Epilogue: Economic Performances Through Time” in Empirical Studies in Institutional Change edited by L. Alston, T. Eggerston and D. North, pp 342-356 (USA: University of Cambridge Press, 1996)

This is a crosspost from Global History @ LSE. All my posts leading up to the start of my course will be crossposted to Left Outside, but once my course starts you’ll have to visit Global History @ LSE if you want to keep up. These are my notes on North’s “Economic Performances Through Time” and is one of seven items of preliminary reading I have to cover before my course starts.

Taken from here. North is attempting to delineate what can and what cannot be learned about the way societies change over time. Section I is on the process of economic change, section II is on what questions we should ask and section III speculates on what questions can be answered and which cannot.


North explains that “a theory of economic performance through time would entail an integration of institutional change, demographic change and the change in the stock of knowledge.” In this essay he chooses to focus on the change of institutions because economics is the study of a process and processes are embodied in the institutions of a society.

Institutions are created by humans to structure human institutions in order to reduce uncertainty in pursuit of their goals (or those making the rules) in social, political and economic exchange.

Institutions are defined as the formal rules (constitutions, statute and common law, regulations etc.) , informal constraints (norms of behaviour, conventions, internally imposed codes of conduct etc.) and the enforcement characteristics of each.

Institutions are not Organisations. Think of Institutions as the rules of the game and Organisations as the players. “Organizations and their entrepreneurs are the actors; they will introduce new institutions or technology when they perceive that they can improve their competitive position by such innovation.”

North now goes into more detail on the processof institutional change.

The institutions of a period define what organisations will exist. Organisations exist under a condition of scarcity and therefore competition. The rate of innovation and change in the institutions which govern our organisations is governed by the incentive structure constructed by this competition.

This is not the competitive conditions of the economic theory of perfect competition but rather the institutional environment of the organizations–the framework of rules and norms–which determines the incentives for innovative activities. An improvement in an organisation is not necessarily an improvement in productivity but could be the creation of a monopoly or the redistribution of wealth in some advantageous way.

Sometimes the innovation involves a change in formal rules and structures [like the repeal of the corn laws] so a study of institutional change must take embody a theory of political change. Sometimes changes involve a gradual change in informal norms of exchange [like the end of religious prohibitions on usury].

Whether organisations change institutions in ways which boost productivity or by redistributing wealth depends on the incentive structure of currently existing institutions. If institutions and the way they evolve are the key to economic performance through time what determines the way they evolve?

The immediate answer is that individual entrepreneurs who are in the position of modifying the rules of the game in political or economic markets and have implicit or explicit theories about the consequences of policies act upon those theories to modify the rules to improve their competitive position; the perceptions of entrepreneurs shape their policies and over time it is the way these perceptions evolve that determines the way institutions evolve. There is no implication that results of the choices that are made will coincide with intentions; indeed more often than not the belief systems that underlie perceptions produce unintended and unanticipated results.


This approach to understanding economic performance through time is at odds with the mainstream literature.

Arnold Toynbee popularised the term “Industrial Revolution” in a series of lectures in 1880-81 and historians have ever since usually onsidered technological change to be the most important determinant for economic performance. For example “take off” theories of development. This is wrong, technologies give us our upper bound, but do not tell us about why this upper bound is so often missed. In contrast, Malthusian pressures have determined the gloomy lower bound.

The economic growth literature has seldom posed the questions which North wants answered. Growth accounting can explain the proportions with which human capital, physical capital, increasing returns, or technology contribute to economic growth but it cannot explain why poor countries don’t invest more in human or physical capital.

Studies of the Third World have often implicitly assumed incentives are aligned correctly, when they are almost always provide incentives for redistributive institutions are larger than those for productive activity. The failure of humans to organise themselves the provide the right incentives must be the centre of inquiry.

Too much study in economics looks at static equilibria, the study of economic performance through time needs a dynamic approach. Given the above, North now reveals his questions.

  1. Are institutions really the carriers of the process of economic change and if so what are institutions?
  2. If institutions play such a role how do they interact with the other key actors in the process of economic change, demographic change, and changes in the stock of knowledge?
  3. What are the sources of institutional change?
  4. What is the process of economic change? How useful are models drawn from evolutionary biology?
  5. How can we explain the diversity of economic performance through time?
  6. How far can we go in constructing a framework that can provide guidelines to improving the performance of third world and transition economies? Can we construct a dynamic theory of change?
  7. Where are we going?


In this section North  approaches each question in term.

Are institutions really the carriers of the process of economic change and if so what are institutions?

Mainstream economics neglects institutions. The evidence of the myriad levels of economic performance even amongst countries facing similar technological and demographic limits should illustrate how important institutions actually are.

However, while mainstream economics do not incorporate institutions into their theory, when they concern themselves with policy they do so in discussing changes in laws and regulations – formal institutions. But Informal institutions are also very important.

If institutions play such a role how do they interact with the other key actors in the process of economic change, demographic change, and changes in the stock of knowledge?

They interact in many ways, most of which we are yet to fully understand. Modern economic growth has its source in the stock of knowledge which is rooted in the scientific revolution of the sixteenth and seventeenth century, yet the roots of this revolution are far from clear.

The formal and informal institutions which led to this scientific revolution need to be understood to explain how it led to the massive changes in demographics and technology which followed.

What are the sources of institutional change?

The source of change reflects the perceptions of economic and political entrepreneurs who perceive ways in which they can improve their position.

In this context, actors do not act as neoclassical theory would suggest. This approach assumes pervasive scarcity and individuals making choices reflecting their preferences. The aggregate of this preference is then constructed  in the context of fixed resources, private goods and given technology. This model is good for illustration the benefits of a decentralised market system but is less useful in understanding institutional change.

When solving economic problems frictions arise [transaction costs] is the context of imperfect information and imperfect enforcement of agreements, and markets are the creatures of political forces. In the real world beliefs define the actions of actors.

We must model beliefs if we are to understand any social science.

Risk versus uncertainty comes into play when discussing someone’s beliefs. Risks can be described as a probability distribution and can be insured against. This is impossible for uncertainty and economics has not created a body of work on this subject. However, all human endeavour is conducted on the basis of some sort of uncertainty – religions, myths, taboos and half baked ideas all serve as the basis of (sometimes life and death) decision making.

The study of economic performance through time must entail the study of how humans learn and meld beliefs and preferences; why they develop these choices in the face of such uncertainty; and why some ideas die out and others prosper.

What is the process of economic change? How useful are models drawn from evolutionary biology?

There are parallels between Darwinian Evolution and the process of economic change. However, Institutional change is largely Lamarkian, change is intentional and is intended to improve the position of those making the changes.

Part of our the scaffolding on which society is built is genetic, part of being human, and part of it is cultural, also I guess part of being human, but in a different way.

Experimental Economists have found evidence which lends support to the position of evolutionary psychologist, human beings cooperate in small groups when transaction costs are small but noncooperative outcomes are favoured in large groups or under conditions of private information. Stephen J Gould and others have suggested that cultural evolution still has a large role to play in the architecture of how humans act. A look to the variety of human societies which share the same genetic material, suggests that cultural evolution play a large role.

Initially the architecture of the  structure is genetic, but interaction with the physical environment and those from the socio-cultural linguistic environment affect this. Categories of experiences are formed which become models on which we decide actions and predict outcomes, as we use these models we update them through confirmation and falsification.

No one individual can understand the whole world but each develops their own models. These models would diverge were it not for ongoing communications from those of a similar cultural background. The interaction of these models creates a culture, provides a method for intergenerational transfer, a means for internal communications, and a method to explain experiences outside the experience of our specific individual. This last function explains the helpes existence of irrational beliefs, some explanation is better than no explanation.

All this implies that individuals are constantly somewhat irrational, unknowledgeable and subject to experiences which may reinforce rather than dismiss incorrect beliefs. North proposes some more questions on what this means for the study of economic performance through time.

  1. What difference does it make that the agents fall far short of substantively rational behavior (full knowledge of all possible contingencies, exhaustive exploration of the decision tree, and a correct mapping between actions, events,and outcomes)?
  2. Is it possible that the agents simply get it wrong in terms of modeling the process of change or representations of the environment?
  3. It may be that the past experiences of the actors lock them in to a belief system and institutional framework that however well they have solved previous problems are strikingly bad at solving new problems.

How can we explain the diversity of economic performance through time?

History is a record of unanticipated consequences and outcomes of decisions made in the face of uncertainty. It could hardly be otherwise.

Economic History is the study of the huge increase in wealth and well-being of humanity, but is is also the study of actions which have produced death, war and famine on a colossal scale. Even what look today like the best laid plans (the US Constitutions) have in fact been aided hugely by chance.

One of the biggest challenges to economic growth for a society is the move from personal to impersonal exchange – a necessary component for greater specialisation and enlarged trade. For example, Genoese traders outperformed Islamic traders in the 11th and 12th century because the Islamic traders relied on in-group social communication networks to enforce collective action while the Genoese formed formal and legal enforcement mechanisms. Thus, due to path dependency, the Genoese could gain more from specialisation and enlarged trading networks.

Institutions change through time and impact one another in unpredictable ways. Because Islamic traders relied on personal exchange, they couldn’t specialise as much as the Genoese, which impaired change in the stock of knowledge which altered the path of demographic change which impacted the change of institutions etc.

How far can we go in constructing a framework that can provide guidelines to improving the performance of third world and transition economies? Can we construct a dynamic theory of change?

A theory of dynamic change is not possible in the same way that a theory of general equilibrium is. We make changes based on new information all the time. To predict those changes would require us to predict that future knowledge, at which point it ceases to be future knowledge and simply becomes knowledge and then becomes now. The past can inform the present and the future, but it cannot predict it.

We know something about the evolution of formal institutions but we know very little about how informal rules change, and these can be just as important.

We need to understand path dependency because it plays a major role in constraining change, if institutions exist which retard economic growth then we need to know hoe easy it is to change them.

Where are we going?

The foregoing discussion suggests that our potential foresight is relatively limited. Forecasting what humans will learn and how the human environmental landscape will change in consequence of that learning and non-human alterations is beyond our capacity and in consequence imposes severe temporal limits on our understanding of economic performance in the future.