Institutions and growth

Posted in development, economics, institutions by jrahman on November 10, 2012

I’ve been reading about growth lately. There’s the new Acemoglu and Johnson book and the accompanying debate.  Plus, for professional reasons have had to think about growth and development.  So, I figure, I should jot down my ever-evolving thoughts.  The idea is to do a series.  In practice, probably it will grind to a halt.  We’ll see.

Over the fold is the first instalment — a review essay I did in 2004.  Yes, the literature has moved since then.  But the contour of the debate hasn’t.


There are strong reasons to think that institutions are an important underlying determinant of economic prosperity.  Recent empirical evidence supports this.  However, measures of institutional quality used by the literature are not satisfactory, and instruments used to account for their endogeneity may not be valid.  These cloud the evidence gathered thus far.  Future research should construct better measures and institutions.  A rigorous theoretical framework will help.

I. Introduction

Is there some action a government of India could take that would lead the Indian economy to grow…?  If so, what, exactly?  If not, what is it about the ‘nature of India’ that makes it so? … Once one starts to think about them, it is hard to think about anything else.                        Lucas (1988)

These questions motivate theorists, empiricists and other practitioners in the discipline of economic growth.  In recent decades, the discipline has made strides in understanding what Lucas (1988) calls the mechanics of economic growth.  Theorists have developed models explaining the roles factor accumulation and technological progress play in driving economic growth.  Empiricists have a better understanding of the relative importance of accumulation and technology.  And policy makers have developed a list of remedies, the so-called Washington consensus, designed to help poor countries lift their growth rates.

However, many governments find it difficult to stick to the policy prescriptions designed to promote growth.  Meanwhile, theorists have asked the question, why did some societies manage to accumulate and innovate more rapidly than others?  One view that has gathered currency in recent years, and is the focus of this essay, is that economic growth and prosperity depends not on specific policies, but on institutions, such as a respect for property rights, which constrain the powers of the state.

Institutions are not the only underlying source of economic growth and prosperity – the role of geography is often stressed for instance.  On the other hand, there is an argument that good institutions are costly, and can be afforded only by countries that have attained a minimum level of prosperity.  According to this view, economic growth leads to good institutions, rather than the other way around.

In this essay, I review the recent literature on the role institutions play as an underlying source of economic growth and prosperity.[1]  I find that:

  • institutions are important for economic prosperity – there are strong reasons to suggest that institutions, together with other factors such as geography, affect economic prosperity, perhaps by making it easier for good policies to succeed; but,
  • the relationship between institutions and prosperity and growth is not well understood empirically – measures of institutional quality typically used in the empirical studies are not satisfactory, while the widely cited instruments are perhaps not valid, clouding the evidence gathered thus far; and,
  • the task ahead includes a more rigorous theory and innovative empirical approach – the task of building theoretical models linking institutions and growth have merely started, while we need to make improvements in the empirical front.

In Section II, I list various underlying sources of economic prosperity.  Section III summarises what we know about the link between institutions, other factors, and economic growth and prosperity.  In the process, this section also discusses various shortcomings of what we know, and highlights what we need to find out.  In the last section, by way of concluding, I outline what should be in the future research agenda.

II. Underlying sources of economic prosperity

Per capita income in the richest region in the world was about 19 times the per capita income in the poorest region in 1998, compared with an interregional spread of about 1.1 around 1000 (Maddison 2001).  Why did some countries grow faster than others did over last 1000 years?

Growth theory and empirics suggest that countries that accumulate human and physical capital or promote innovation will grow faster.  Factor accumulation fails to account for the bulk of the cross-country difference in per capita income (Easterly & Levine (EL) 2001) or output per worker (Hall and Jones (HJ) 1999).  To say that some countries are rich because they innovated begs the question why they innovated while others did not.  Thus, Rodrik, Subramanian & Trebbi (RST) (2004) suggest that factor accumulation and innovation are at best proximate sources of growth.[2]  To understand why some countries are more prosperous, we need to look for deeper underlying sources.

What are the underlying sources of prosperity?  The usual suspects are institutions, geography/endowments, integration, policies and culture (EL 2003, Acemoglu, Johnson & Robinson (AJR) 2004, and RST 2004).[3]  This section discusses how these factors affect economic prosperity.

Let us begin with institutions.  What exactly are institutions?  AJR (2004) use the North (1990) definition: Institutions are the rules of the game in a society or, more formally, are the humanly devised constraints that shape human interactions.[4]

In rich countries, institutions are such that property rights are generally respected, the rule of law holds, private incentives are aligned with social objectives, and citizens have civil liberties and political freedom.  Thus, rich countries have good institutions.  The institutionist view holds that the rich countries are rich because they have these good institutions.  Of course, the opposite is also likely – rich countries have good institutions because they are rich (Barro 1999).

Geography is frequently considered as another underlying source of economic prosperity.  Tropical or land-locked countries are more likely to have lower growth rate (Sachs & Warner (SW) 1997).  Temperate climate and east-west continental alignment explains why the agricultural revolution took place in the temperate and not tropical regions (Diamond 1997).  Tropical environments tend to have poor crop yield and worse disease environment (Sachs 2001).

The institutionist view argues that geographical endowments affected initial institutions, which affected economic outcome.  Western European countries developed good institutions, became more prosperous, and colonised the rest of the world.  However, the colonisation strategies they adopted, and thus the institutions they introduced in the colonies, depended on geographical endowments.  Where the Europeans settled, they set up good institutions that respect property rights or civil liberties.  Elsewhere they set up institutions designed to extract resources.  Upon independence these institutions persisted, leading to settler colonies being more prosperous (AJR 2001).  Further, even among the countries where the Europeans settled, differences in land endowments resulted in different institutions, and thus different economic outcomes (Engerman & Sokoloff 2000).

Integration, the degree to which a country is open to the rest of the world, is also cited as an underlying source of growth (SW 1997).  That international economic integration can be a driver of productivity improvement is accepted by most economists.  However, the institutionists argue that to realise the gains from trade, a country needs a minimum quality of institutions.  Further, as discussed above, geography affects how much a country can trade with the world.  Finally, some argue that rich countries have more to gain from trade, and thus the direction of causation runs from growth and prosperity to trade.

The degree to which a country integrates with the world is at least in part a policy choice.  Other policies promoting macro stability, micro competition or sound public finance can also affect growth.  International financial organisations often stress that government policies can overcome geographical barriers to growth and prosperity, but many governments find it difficult to stick to good policies (Easterly 2002).  This is because of poor institutions, the institutionists argue.

Finally, culture is considered as a source of economic prosperity.  Culture is viewed as a key determinant of the values, preferences and beliefs dominant in a society (AJR 2004 summary of Weber 1930).  Different societies have different cultures because of religion or history.  Religious beliefs about heaven or hell appear to be positively correlated with economic growth, but religious acts such as church attendance seems to depress growth (Barro and McCleary 2003).  For a given quality of institutions, different cultures can lead to different economic outcome.  Nonetheless, the link between institutions and culture in the current literature is rather nebulous.  Perhaps more importantly, culture may be endogenous – rising prosperity may change the values, preferences and beliefs of a society.

From the above, several themes appear:

  • the factors discussed above usually affect economic growth and prosperity by interacting with each other; and,
  • institutions interact with all the other factors, and is likely to have a major role in shaping economic outcome; but,
  • except for geography, all the other factors, including institutions, are affected by rising prosperity, and it is likely that there is reverse causality.

The task is to untangle the web of interactions and reverse causalities, and determine what are the underlying sources of economic prosperity.

III. Institutions and their discontents: the evidence

How do we go about finding what are the underlying sources of economic prosperity?  The first task is to measure these factors.  Then we need appropriate instruments to show that these factors are independent determinant of economic prosperity, and not consequences of higher incomes.  Finally, by running appropriate regressions using proper instruments, we should be able to separate the significant factors from the rest.

The recent literature (HJ 1999, AJR 2001, EL 2003, RST 2004) suggests that good institutions are a very important, if not the, underlying source of economic growth and prosperity.  However, more recent literature challenges this view, alleging or acknowledging that one or more of the above three tasks are not performed well (Glaeser et al 2004, Rodrik 2004).

In this section, I consider how the recent literature addresses these tasks.  I find that measures of institutional quality typically used in the empirical studies are not satisfactory, while the widely cited instruments are perhaps not valid, clouding the evidence gathered thus far.

a. Measuring the factors

Geography is perhaps as exogenous a variable as economists can hope to get – short of invading the neighbours to its or north or south, a country is not going to change its latitude as it develops.[5]  Distance from the equator is a frequently used measure of a country’s geography (RST 2004).  Other measures include landlockedness, mean temperature, prevalence of diseases such as yellow fever or malaria, and share of natural resource export in GDP (SW 1997, AJR 2001, Sachs 2003).

Openness ratio (the ratio of total trade to GDP) or the extent of tariff and non‑tariff trade barriers are common measures of a country’s economic integration (RST 2004).  There are several measures of economic policy.  Central government budget position (SW 1997), inflation (SW 1997, EL 2003), and real exchange rate overvaluation (EL 2003) have been used to measure macroeconomic policy.  SW (1995) define a country as open based on its tariff and non-tariff barriers, other trade policies and the role of market in the country’s economy.  This is often used as a measure a country’s trade policy (EL 2003, RST 2004).  The proportion of a country’s population affiliated with different religions is an oft-used measure of culture (HJ 1999, EL 2003).

Institutions, or more accurately institutional quality, are not easily measured.  A range of perception-based indicators of the rule of law or property rights is usually used to measure institutional quality.  Kaufmann et al (1999, 2002) provide indicators on: voice and accountability, political instability and violence, government effectiveness, regulatory burden, rule of law and graft.  Averages or composites of these indicators are commonly used (EL 2003 and RST 2004).  Political Risk Services, a firm providing assessments of risk to international appropriation to international investors, also compile indices that have been widely used (HJ 1999, AJR 2001).  Another measure of institutional quality is an indicator of the constraints on the executive branch of the government (AJR 2001).

Measuring institutional quality with such perception-based indicators is problematic.  Recalling the definition of institutions, Glaeser et al (2004) note that good institutions should provide durable constraints on economic agents.[6]  For instance, a good institution would have durable constraints on the government, prohibiting the government from expropriating private property.  Suppose a government is not constrained from expropriating private property, but chooses to abstain from expropriation voluntarily.  The perceptions‑based indicators will misleadingly tell us that this government is working under a good institution.  A government’s policy choice tells us little about the underlying quality of the institutions.  Nor can we gain much insight about institutional quality by merely looking at the legislations and constitutional rules constraining the government.  Rodrik (2004) illustrates how ‘the rules of the game as written down in the books’ mean little with an example involving Russia and China.

The post-Soviet Russia enshrined private property rights in its constitution in a way that is yet to be matched in China.  However, the investors perceive property rights to be more secure in China than in Russia.  Most of the measures of institutional quality used in the literature suggest that China has better institutions than Russia.  However, China’s superior performance may largely be because of policy choices taken by China’s leadership since Deng Xiaoping.  To the extent that a future leadership can reverse these policy choices, China does not have good institutions.  Indeed, Glaeser et al (2004) show that the usual measures of institutions are highly volatile.

The above discussion suggests that:

  • the measures of institutional quality are not satisfactory – until we improve our measure of the quality of institutions, any empirical work aiming to establish the importance of institutions, or lack thereof, should be interpreted with caution; and,
  • policy matters, sometimes – to the extent that what we use as measures of good institutions actually tell us more about the effectiveness of good policies, we need to better understand why and when good policies are adopted.

b. Appropriate instruments

As discussed above, rising prosperity affects several factors mentioned in Section II.  To discern their impacts on economic prosperity, we need proper instruments.  A good instrument, for say integration, is a source of exogenous variation in economic integration that is uncorrelated with other plausible determinants of economic prosperity.  One instrument used for integration, measured by trade/GDP ratio, is a trade/GDP ratio constructed from a gravity equation for bilateral trade flows (Frankel & Romer 1999).

RST (2004) notes that this instrument passes the ‘AER test’ because it was published in the American Economic Review.  The shortcomings of the measures of institutional quality notwithstanding, I outline the instruments used for institutional quality in the recent literature in this subsection.

HJ (1999) instruments are various correlates of Western European influences.  Western Europe discovered the ideas of Adam Smith, the importance of property rights and representative government.  Thus, more influenced a country is by Western Europe, better its institutions are likely to be.  The instruments they use are latitude and the prevalence of English or other Western European languages.[7]

AJR (2001) use mortality rates of European settlers as an instrument for institutional quality.  They argue that Europeans settled in the colonies with low mortality rates, and in these colonies, they erected good institutions that respected property rights or accorded civil liberties.  Elsewhere, they erected institutions designed to extract resources.  As institutions are durable, AJR (2001) argue that settler mortality rates are therefore a good instrument for institutional quality.

AJR (2001) instrument passes RST (2004)’s ‘AER test’, and has received a lot of attention.  However, Glaeser et al (2004) provide a strong rebuttal of the validity of the AJR (2001) instrumentation approach.  They show that AJR (2001) instruments for institutional quality are even more highly correlated with human capital.  This suggests that mortality rates of European settlers, or other predictors of European settlement, may not be valid instruments for institutional quality.[8]

In addition to the Glaeser et al (2004) critique of AJR (2001) instrumentation approach, using the correlates of Western European influence to instrument for institutional quality has another shortcoming.  This approach implies that the variation in economic prosperity around the world today is fundamentally due to different forms of European colonisation.  How then can we account for the variation in economic prosperity in countries that have never been colonised by Western Europe (RST 2004)?

The above discussion suggests:

  • we need better instruments – in addition to finding better measures for institutional quality, we need to develop better instruments.

Perhaps constructing a better measure of institutional quality will help us find better instruments.

c. Institutions versus other factors

Economic development is a complex phenomenon.  Historians and other social scientists would suggest that all the factors mentioned in Section II have influenced economic prosperity or lack thereof.  However, economists like to quantify things, and the recent literature (HJ 1999, AJR 2001, EL 2002, RST 2004) show that good institutions are a primary source of economic prosperity.  Of course, to the extent that there are shortcomings in the measures of institutional quality and the instruments used, these results should be taken with caution.

An intuitively appealing way to determine whether institutional quality is more important than other factors is to regress the current level of economic prosperity (usually per capita income) on these factors, after having them appropriately instrumented.  Levels, rather than growth, are preferred because they capture the difference in long-run economic performance (HJ 1999).

EL (2002) take this approach.  They regress the levels of income on various measures of endowments, institutions and policies.  They find that institutional quality influences economic prosperity.  They also find that endowments and policies do not have any influence on economic prosperity other than through institutions.

RST (2004) also follow in the same path, regressing levels of income on geography, institutional quality and economic integration.  They find that institutional quality overshadows the influence of other factors.  Once institutions are controlled for, integration has no direct effect on prosperity, while geography has a weak direct effect.

RST (2004) consider regressing income level on policies problematic.  They argue that policies pursued over a shorter time period is like a flow variable, while economic prosperity is like a stock variable that is a result of much longer historical process.  This means that level regressions with policy as an independent variable conflates stocks and flows.  To the extent that measures of institutional quality actually reflects good policy choices (the way for instance low inflation reflects good monetary policy), any regression of income on institutional quality would be equally problematic.

Glaeser et al (2004) provide a different interpretation of the AJR (2001).  Recall the AJR (2001) story is that Europeans brought good institutions with them to the countries where they settled, and this explains the observed differences in the economic prosperity among former European colonies.  Glaeser et al (2004) argues that European settlers brought themselves, ie their human capital, to the settled colonies.  By way of evidence, they present instrumental variable specifications predicting economic growth where human capital performs better than institutions.[9]

The above discussion suggests that:

  • institutions appear to rule, but we are not sure if they really do – until we construct better measures of institutional quality, and find more valid instruments, the empirical evidence gathered thus far suggesting the primacy of institutions will remain suspect.

IV. Concluding remarks

Discussions in Section II suggest that there are strong reasons to suggest that institutions, together with other factors such as geography, affect economic prosperity, perhaps by making it easier for good policies to succeed.  But discussions in Section III highlights how far away we are from establishing exactly how important institutions are.  We have neither a reliable measure of institutional quality, nor a good econometric instrument for it.  Addressing these concerns should top the future research agenda.  Indeed, if a better measure is developed, that itself could help find an appropriate instrument.[10]

We also need rigorously developed models explaining the mechanics of how institutions affect growth.  AJR (2004) sketches the outlines of such a model.  In their framework, political institutions and distribution of resources are the state variables.  These variables evolve because current economic institutions affect future distribution of resources, and groups with current de facto political power changes the political institutions to increase their future de jure political power.  Economic institutions promoting growth emerge when political institutions promote property rights and constraints on the power holders.  A full exposition of this, or other similar, model would allow us to form testable hypotheses.

Meanwhile, what of policy?  If we think that governments do not stick to good policies because of bad institutions, but cannot actually measure how bad an institution actually is, what role then do we have for policy?  On the other hand, to the extent that what we use as measures of good institutions actually tell us more about the effectiveness of good policies, we need to better understand why and when good policies are adopted.

A recent study finds that a surprisingly small change in policy can lead to a sustained increase in economic growth (Hausmann, Pritchett & Rodrik 2004).[11]  To quote Rodrik (2004):

Instigating growth is a lot easier in practice than the standard Washington recipe, with its long list of institutional and governance reforms, would lead us to believe.

This is surely welcome news for policy makers.






V. Reference

  1. Acemoglu D, Johnson S & Robinson JA, The colonial origins of comparative development: an empirical investigation, American Economic Review, vol 91(5), pg 1369-1401, 2001.
  2. Acemoglu D, Johnson S & Robinson JA, Institutions as the fundamental cause of long-run growth, NBER Working Paper no 10481, 2004.
  3. Barro RJ, Determinants of democracy, Journal of Political Economy, vol 107, no 6, pt 2, pg s158-183, 1999.
  4. Barro RJ & McCleary R, Religion and economic growth, NBER Working Paper no 9682, 2003.
  5. Diamond J, Guns, germs and steel: the fates of human societies, WW Norton, 1997.
  6. Easterly W, The elusive quest for growth: economists’ adventures and misadventures in the tropics, MIT, 2002.
  7. Easterly W & Levine R, It’s not factor accumulation: stylized facts and growth models, World Bank Economic Review, vol 15(2), pg 177-219, 2001.
  8. Easterly W & Levine R, Tropics, germs and crops: how endowments influence economic development, Journal of Monetary Economics, vol 50, pg 3-40, 2003.
  9. Engerman SL & Sokoloff KL, Institutions, factor endowments, and paths of development in the new world, Journal of Economic Perspectives, vol 14, no 3, pg 217-32, 2000.
  10. Frankel J & Romer D, Does trade cause growth?, American Economic Review, vol 89(3), pg 379-399, 1999.
  11. Glaeser EL, La Porta R Lopez-de-Silane F & Shleifer A, Do institutions cause growth?, Journal of Economic Growth, vol 9 (3, Sep), pg 271-303, 2004.
  12. Glaeser EL & Goldin C, Corruption in America: an introduction, NBER working paper no 10775, 2004.
  13. Hall RE & Jones CI, Why do some countries produce so much more output per worker than others?, Quarterly journal of Economics, vol 114(2), pg 83-116, 1999.
  14. Hausmann R, Pritchett L & Rodrik D, Growth accelerations, NBER Working Paper no 10566, 2004.
  15. Kaufmann D, Kray A & Zoido-Lobaton P, Governance matters, World Bank Policy Research Department Working Paper no 2196, 1999.
  16. Kaufmann D, Kray A & Zoido-Lobaton P, Governance matters II – updated indicators for 2000-01, World Bank Policy Research Department Working Paper no 2772, 2002.
  17. Lucas RE, On the mechanics of economic development, Journal of Monetary Economics, vol 22, pg 2-42, 1988.
  18. Maddison A, The world economy: a millennial perspective, OECD, 2001.
  19. North DC, Structure and change in economic history, Norton, 1981.
  20. North DC, Institutions, institutional change, and economic performance, Cambridge University Press, 1990.
  21. North DC & Thomas RP, The rise of the western world: a new economic history, Cambridge University Press, 1973.
  22. Rodrik D, Getting institutions right, CESifo DICE Report, Summer 2004.
  23. Rodrik D, Subramanian A & Trebbi F, Institutions rule: the primacy of institutions over geography and integration in economic development, Journal of Economic Growth, vol 9 (2 June), pg 131-165, 2004.
  24. Sachs JD, Tropical underdevelopment, NBER Working Paper no 8119, 2001.
  25. Sachs JD, Institutions don’t rule: direct effects of geography on per capita income, NBER Working Paper no 9490, 2003.
  26. Sachs JD & Warner AM, Economic reform and the process of global integration, Brookings Papers on Economic Activity, vol 1, pg 1-118, 1995.
  27. Sachs JD & Warner AM, Fundamental sources of long-run growth, American Economic Review, vol 87(2), pg 184-88, 1997.
  28. Weber M, The protestant work ethic and the spirit of capitalism, Allen & Unwin, 1930.


[1] I do not replicate any of the studies I cite.  Nor do I devote too much time on specific econometric modelling issues.  Data limitation is an issue that plagues empirical studies in economics, and I do not dwell on it.

[2] AJR (2004) go further by quoting North & Thomas (1973): the factors we have listed (innovation, economies of scale, education, capital accumulation etc) are not causes of growth; they are growth (emphasis in original).

[3] EL (2003) list geography/endowments, institutions and policy.  AJR (2004) list economic institutions (different from political institutions), geography and culture.  RST (2004) list geography, institutions and integration.

[4] Glaeser et al (2004) use a similar definition: a set of rules, compliance procedures, and moral and ethical behavioural norms designed to constrain the behaviour of individuals in the interests of maximising the wealth or utility of principals (North 1981).


[5] Economic prosperity can allow military prowess, which in turn can lead to a country expanding its border.  This rather laborious possibility is discounted in the literature.

[6] Recall the definitions mentioned above: institutions are the rules of the game in a society or, more formally, are the humanly devised constraints that shape human interactions (AJR 2004 quoting North 1990); and, a set of rules, compliance procedures, and moral and ethical behavioural norms designed to constrain the behaviour of individuals in the interests of maximising the wealth or utility of principals (Glaeser et al 2004 quoting North 1981).

[7] In the literature, these instruments are also used as a measure of geography or culture in their own rights.

[8] Glaeser et al (2004) reinterpretation of AJR (2001) story is discussed below.

[9] Glaeser et al (2004) regress economic growth rather than income level, and do not stress the role of human capital too much, noting the tentative nature of the finding.

[10] Researchers in all likelihood have started work in this direction.  Glaeser & Goldin (2004) provide an example regarding measures of corruption.

[11] It appears that like the non-accelerating inflation rate of unemployment, there is a missing derivative in growth acceleration.  Perhaps growth increase or income acceleration would have been more appropriate.  Nonetheless, Hausmann, Pritchett & Rodrik (2004) passes the ‘NBER test’!


2 Responses

Subscribe to comments with RSS.

  1. Diganta said, on November 11, 2012 at 10:45 am

    Want to read in details and study further. Just leaving a thanks for this writing for now 🙂

  2. BDAF said, on November 24, 2012 at 4:39 pm

    An interesting review on the book by Jared Diamond [author of Guns, Germs & Steel], here:

    I like this review, primarily because it covers some ground that the authors miss. It covers the issue in a more holistic manner, I think. Readers please check out.

Comments are closed.

%d bloggers like this: