RaboResearch - Economic Research

Institutional quality and economic performance


  • Institutional quality is a broad concept that captures law, individual rights and high quality government regulation and services
  • Institutional quality and economic development reinforce each other over the longer term, but we argue that institutional quality leads this virtuous circle
  • Importantly, institutional development unlocks growth potential and does not intrinsically suffer from diminishing returns
  • Data show that countries with high institutional quality have been more successful in adopting frontier technology and productivity since the turn of the millennium

We build on the intuition that institutional development is the best indicator for structural development and long term welfare creation for a nation. Economic growth is key for determining the short term trajectory of a nation almost regardless of its origin, but institutional development determines whether short term gains are sustainable over the longer term. High quality institutions will not prevent the next economic crisis in a market economy, but they do raise the odds that a society can cope with and recover from such crisis and continue on its long term trajectory of progress.

There are at least two links business strategy. First, we focus on the long term economic potential of a country, from the premise that the contemporaneous growth profile may look appealing, but a firm really wants to understand a country’s long term potential to decide where it wants to develop lasting business. Second, much of the realization of potential in any foreign business strategy depends on a benign ‘enabling environment ’ over and above an economy’s short term growth success. We operationalize this enabling environment in a rigorous way and make it measurable at the macroeconomic level.

Institutions and economic development

The inquiry into the nature and causes of the wealth of nations dates back at least to 1776, when Adam Smith published his similarly titled landmark book (Smith, 1776). Smith eloquently describes how well-organized market places allow individuals, solely pursuing their personal interests, to collectively maximize economic welfare. This is the famous ‘invisible hand.’ While this principle has generated heated debates over the course of history as to the parameters within which a market is ‘properly organized’ as well as what ‘economic welfare’ is or should be, it has also raised many open questions that have spurred scholarly interest to this very day.

Fast forwarding to the second half of the twentieth century, Robert Solow (1956) made a landmark contribution, explaining that long-term economic growth is limited to the rate of technical progress, which at the time he assumed to be exogenous. His crucial revelation was that given constant returns to scale production technology all countries can converge to the income levels of the most productive nations and henceforward grow at the exogenous rate of technical progress. Gregory Mankiw, David Romer and David Weil (1992) in the early nineties made the point that this conclusion also extends to an augmented Solow model that includes human capital in addition to physical capital.

By that time, an equally heated debate had started about the endogeneity of technical progress. Paul Romer (1986) floated the idea that knowledge might be an input into production that has properties of increasing returns to scale. Robert Lucas (1993) extended this view with learning by doing, to operationalize the increasing returns properties of knowledge. This generated a vast enquiry into the production of knowledge, the distinction between codified and tacit knowledge, about publicly available ‘knowledge pools’ from which entire societies can benefit.

It is the leap of the economic growth literature to knowledge as a central input that has put the enquiry into the role of the institutional context to the forefront. Indeed, if knowledge is central to potential increasing returns to scale in production, yet at the same time can be freely copied and adopted by anyone at any time: i) how can we explain individuals’ investment in knowledge production, ii) why don’t we see frontier knowledge spread much faster over the whole world, to lift productivity levels of all countries to those at the frontier? Indeed the Solow prediction of structural convergence between economic welfare between countries has not been validated by the data since. The answer lies in how societies and markets are organized and institutions are front and center in that sense.

We assess the importance of institutional quality for economic development below in three chapters. We first address the definition and operationalization of institutional quality and argue why institutional quality leads the virtuous circle between institutional development and economic performance. We then elaborate on how exactly institutional quality transmits into economic growth and find that it shapes an enabling environment for individuals and firms to invest, innovate and grow. Third, we global data on institutional quality and economic performance and find evidence of institutional quality associating with faster rates of income convergence to the global frontier of productivity and economic welfare.

What are institutions and why are they important for economic development?

Although by now there is a common understanding that institutions and long term economic progress are intimately linked, there remains much debate about what those institutions are exactly. And this feeds through into the remaining discussion in the highest academic circles about what comes first: institutional development, or economic progress? Hence in the sections below we’ll first work towards a generic definition of institutional quality and operationalize it using objectively measurable, publicly available data. Then, we’ll address the crucial question as to whether institutions are leading or following economic development.

What are institutions?

Much of the pioneering work into institutional quality has been done by Douglass North (1981, 1990). North defines institutions as humanly devised constraints that shape interaction between people. Essentially, in North’s framework, institutional quality improves with the limitations imposed on executive power. Such limitations may be either formal rules or informal constraints and their strength is shaped by the characteristics of enforcing them. The idea being that limitations to executive power reduce the de jure position of a country’s executives to put themselves above the law. And that ensures individuals, entrepreneurs, challengers of the present economic system, that they are protected by the law in their ventures and their investments in human and physical capital as well as new technological endeavors. Such endeavors are crucial to speed up i) the widespread adoption of frontier technology available elsewhere and ii) to push out the technological frontier by investing in research and development, particularly in disruptive technologies. Such endeavors are highly uncertain by nature, and their disruptive character additionally makes them a challenge to those in positions of formal and informal power, be that political or economic. Hence the need for sufficiently high quality institutions to ensure that challengers and incumbents receive equal legal protection. The literature is far from limited to North, however, where the definition of institutions is concerned.

Acemoglu, Johnson and Robinson (2001, 2002, 2005) do not so much look at de jure executive power, but apply a wider perspective that takes into account both de jure and de facto power. Also, a distinction is made between not only into formal and informal power, but also into political and economic power. In such a more complex institutional setting –which we will elaborate upon later on– it suffices for there to be a proper balance of power, be it formal or informal, that ensures individuals, entrepreneurs, challengers of the present economic system, that they are protected de facto in their ventures and their investments in human and physical capital as well as new technological endeavors. Empirically, Acemoglu et al. (2001) use the perceived risk of expropriation as key indicator of institutional quality. The difference of this institutional view relative to that of Douglass North may seem small, but it is crucial in settings where de jure political power is concentrated in different groups than de facto economic power, limiting the exercise of the de jure political power. As we’ll see below, this nontrivial difference also plays a substantial role in the ongoing debate about the direction of causality between institutions and economic progress.

Yet another addition has been suggested by William Easterly (2001, 2013), who stresses the rights and the opportunities of the individual. With an explicit link to the line of thinking by Friedrich von Hayek (1948), Easterly (2013) builds the case that any type of economic progress that is to be lasting should be built on respect of the rights of the individual. His focus is on the very poor individuals in developing countries, but the line of enquiry applies to the poor and rich alike. The key message is as simple as it is crucial: lasting progress is always and everywhere the fruit of investment. Investments are almost inevitable sunk, be they investments in education, in physical capital, or in new technology. An investment is made based on an expectation of the conditions under which its fruits may be enjoyed. This may be the right to take a profession for which someone has been studying for some years, the right to operate a plant into which several physical investments have been made, or the right to develop and market new products and services that derive from the technology that someone has been financing research into. If these individual rights are (expected to be) easily violated, the investments will not be made and concomitant progress will not materialize.

There is an intimate link here with the strand of research focusing on ‘trust societies’ (Algan and Cahuc, 2010), wherein trust amongst individuals in economic transactions and trust of individuals in their legal rights is key in determining the institutional setting that generates economic progress. Legal rights in this sense can be interpreted as forms of formalized trust. In the line of enquiry pursued by Acemoglu, Johnson and Robinson (2005) such legal rights generate ‘inclusive institutions,’ institutions whose rights and protection loosely speaking include all, regardless of position in society or origin. This is opposed to ‘extractive institutions,’ which loosely speaking only serve to extract resources from the masses for the greater benefit of the ruling elite.

Last but not least, Easterly (2013) includes effective public services as an integral element of high quality institutions. From the point of view of developing countries, Easterly argues that legal and political rights are quite unproductive if they are faced with poor public services. Envisage machines not running because of chronic power failures as a result of a lack of public investment and maintenance of electricity supply infrastructure. The importance of such public services is easily extended to high(er) income nations, where for example infrastructure shortages, or cumbersome and ineffective bureaucracy hamper the full exploitation of business opportunities (e.g. Giordano et al. (2015) for an application to Italy).

Operationalizing institutional quality

The literature surveyed above lets us draw two important conclusions needed for bringing institutional quality to the data. First, institutions should be viewed as a basic requirement for economic success and long term progress. While we’ll spend more attention to the direction of causation later on, for now this implies that institutional quality should be constructed as a variable that is independent from economic growth and progress that it intends to foster. Second, institutional quality consists of a broad range of factors, some of which are hard to measure. For the sake of objectivity and replication, it is strongly preferred to base this indicator on publicly available data, that is as much as possible objectively assembled.

To that end, we construct institutional quality from seven variables, available from the World Bank website, being the six World Bank Governance Indicators and the Ease of Doing Business indicator. With our selection of indicators we follow the relevant literature. Easterly and Levine (2003), IMF (2003) and more recently Kuncic (2013) and Fabro and Aixalá (2013) all use the World Bank Global Governance Indicators (WGI). We construct institutional quality annually, so that we may also keep track of institutional improvements or deterioration over time. We add the Ease of Doing Business indicator, as it specifically captures the quality of the processes and administrative work, over and above the WGI’s ‘regulatory quality.’ Thus, for us, institutional quality consists of the following seven key dimensions:

  • Voice and accountability: capturing the extent to which a country's citizens can select and challenge its government, thus limiting executive power;
  • Political stability and absence of violence: the lower the probability of political instability and/or politically-motivated violence, the more a country’s citizens are incentivized to invest in their own prosperous future (e.g. Alesina et al., 1996);
  • Government effectiveness: capturing the quality of public services and the degree of its independence from political pressures, thus fostering a benign context for private investment;
  • Regulatory quality: the ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development, thus laying down uniform rules of economic engagement;
  • Rule of law: captures particularly the quality of contract enforcement, property rights, the police, and the courts, i.e. the enforcement of the rules of society;
  • Control of corruption: the stronger is control of corruption, the more economic success is a function of effort and competence, rather than connections and bribery;
  • Ease of Doing Business: captures a multitude of aspects that determine the extent to which the regulatory environment is conducive to business operation.

As a technical note, the Ease of Doing Business Indicator is transformed into a z-score using its absolute distance to frontier component. Subsequently we compute the simple average of this normalized Ease of Doing Business Indicator and the six WGIs (which are already expressed as normalized values). The institutional quality indicator thus operationalized must be interpreted as a relative score; it expresses institutional quality relative to the global average, normalized by the global variation around this average.

Institutional quality across Europe

Applied to the European[1] continent, our indicator for institutional quality illustrates the strength of European institutions relative to the global average (figure 1). European institutional quality outperforms the global average on all seven components of which it consists, although the lead is somewhat less pronounced in in terms of political stability and absence of violence. This observation applies to a lesser extent to control of corruption as well.

Figure 1: The European ‘institutional quality cobweb’
Figure 1: The European ‘institutional quality cobweb’Source: World Bank, Rabobank

The level of institutional quality nevertheless varies considerably across the continent, whereby EU and euro-area membership do not automatically correspond with higher levels of institutional quality (figure 2). Meanwhile, once poor former communist economies in Central Europe saw concurrent improvements in both institutional quality and income levels.

On average, the Nordic countries display the highest level of overall institutional quality both in Europe and globally. They perform strongly across all sub-indicators. Western Europe features the second-highest level of institutional quality, which, on average, is only slightly below Nordic levels. Within the region, Switzerland and the Netherlands perform strongest, whilst French institutional quality is weakest (comparable to Estonia’s, the best-performing Central European country). Ease of doing business constitutes a relative weakness in France and Luxembourg. Thanks to significant structural reforms since the end of communism in 1989, the Central European and Baltic countries take a middle position in terms of institutional quality in Europe today. Institutional quality in most countries is comparable to levels observed in Southern European countries like Portugal, Cyprus or Spain. Compared to its European peers, the region performs relatively favourable in terms of political stability, ease of doing business, and voice and accountability (with the notable exceptions of Hungary and Latvia, whilst the latest Polish changes to the constitutional court and media laws have not yet been reflected in the indicators).Government effectiveness and regulatory quality are mediocre, however, while relatively low levels of control of corruption remain a key weakness of the region. In terms of institutional quality Southern Europe is a mixed bag. Spain, Portugal, and Cyprus feature institutions that compare well with the tail-end of the Western European group, while Italy and particularly Greece more closely resemble lower South-Eastern European standards. The former three countries perform relatively strongly in terms of government effectiveness, but regulatory quality, as well as political stability and absence of violence/terrorism constitute challenges, particularly so in Spain (Bask separatism) and Cyprus (tensions with Turkey and Northern Cyprus). Italy and Greece, however, suffer from a combination of weak scores on corruption, rule of law, and government effectiveness, combined with low scores on political stability, particularly so in Greece.
Although institutional quality remained relatively stable in Italy over the past years, the country’s score on control of corruption slipped. Meanwhile, in Greece, recently implemented reforms in line with the country’s EU/IMF bail-out packages so far fail to bring about an improvement of overall institutional quality. While rules and procedures for doing business have been improved, the scores on rule of law and legal protection have deteriorated markedly, whereby the latter partly reflects reform-driven legal changes. Greece’s already limited score on control of corruption fell, as did the score on voice and accountability. Compared to its European peers, South-Eastern Europe features a very low level of Institutional quality. A detrimental combination of weak scores on legal protection and rule of law, control of corruption, government effectiveness, and lingering political instability constitute major institutional challenges, all of which exceed the global average. Though still trailing the rest of Europe by large margins, the region outperforms the global average in terms of regulatory quality and ease of doing business. Within the region, with the exception of Croatia and Montenegro, the relatively young Western Balkan countries are facing particularly big challenges institutionally, while Turkey, the region’s biggest economy, takes a middle position. While benefitting from relatively strong government effectiveness, regulatory quality and ease of doing business, Turkish institutional quality suffers from very low scores on political stability and voice and accountability, both of which have deteriorated since 2010. The two indicators bear witness to the increasing polarization of Turkey’s society and a concurrent deterioration of press freedom amid fears of increasing authoritarianism.

Europe has ranked among the global forefront in terms of having well-developed institutional frameworks for many decades. Consistent with our postulated connection between institutions and economic development, GDP per capita levels in vast parts of the continent also rank among the highest in the world. Across the European continent, the variation in institutional quality correlates with the level of economic prosperity. Whilst this corroborates with our assumption that institutional quality drives economic progress, it also naturally raises the question of the direction of causation between institutional quality and economic progress once more.

Figure 2: Institutional quality across European countries
Figure 2: Institutional quality across European countriesSource: World Bank, Rabobank

Do institutions lead or follow economic development?

This question continues to feed debate within the highest academic circles. The two sides of the debate are most effectively represented by Acemoglu, Johnson and Robinson (2005) –supporting the institutions-drive-progress direction of causality– and Glaeser, La Porta, Lopez-de-Silanes and Shleifer (2004) – vigorously supporting the reverse causality.[2]

Interestingly, both sides of the debate use the historical example of the Korea’s to build their cause. Acemoglu et al. (2005) refer to the Korea’s to pointing out that the two Koreas at the end of the second World War differed little in terms of economic endowments or structure. The main difference has been in terms of subsequent choices in terms of institutional organization. South Korea maintained a system of private property and an economic model based on private incentives and market forces. South Korea thus followed the path of inclusive institutions and prospered, turning into one of the ‘Asian economic miracles’ in the 1960s. North Korea in contrast followed the communist model in abolishing private property and installing an centrally planned economy. The North Korean regime chose extractive institutions and has seen its economy lag that of its southern neighbor since, even falling back in terms of absolute economic welfare since the 1990s. Acemoglu et al. (2005) see the Korean ‘natural experiment’ as a clear case supporting their view that institutional quality are an essential element in the enabling environment that drives long term economic progress.

Not so, say Glaeser et al. (2004) referring to the same experiment, pointing out that the institutional quality differential between the two Koreas was negligible when they separated into two distinct countries. According to Glaeser et al., it was the subsequent economic development that allowed South Korea to adopt a more inclusive institutional arrangement. Thus, essentially, they view the contemporaneous institutional quality differential between South and North Korea as a result of the economic progress seen in the South, rather than the other way around. Incidentally, but importantly, they implicitly view inclusiveness of institutions as a sort of luxury good, that societies can only permit themselves to invest in after they have achieved a minimum level of economic prosperity. They do not, however, explain what generated the South Korean lift off and why it never reached North Korea.

Acemoglu, Johnson and Robinson (2005) acknowledge the interplay between economic outcomes and institutional quality, yet maintain that institutional quality leads this dance. Figure 3 below outlines the mechanics as laid down in their analysis. Working from the left to the right, it starts by making the distinction between de jure and de facto political power. The former derives from the rules or institutions laid down in the past. The latter derive from economic clout of individuals or interest groups that are based on the command over economic resources. One may think of the political influence exerted contemporaneously by corporate lobbies or labour unions.

Figure 3: The interplay between institutions and economic results
Figure 3: The interplay between institutions and economic resultsSource: Acemoglu, Johnson and Robinson (2005)

The dynamic distinction between de jure and de facto political power is that the latter is transitory by nature; dependent as it is on the contemporaneous resource distribution. Only if economic power is sufficiently strong, lasts sufficiently long and / or is sufficiently concentrated within a politically homogeneous interest group, can it be wielded to re-forge, amend or invent the new institutional setting for the future. The contemporaneous de facto political power has then been solidified into future political institutions and will command de jure political power from that point in time onwards. That is, until the power balance once more changes the institutional setting. Changes in economic conditions, resulting in changing economic fortunes may influence the resource distribution and thus shift the political power balance, driving further change in the institutional setting.

Industrializing England in the eighteenth century serves as a telling example. With the machination of production facilities, infrastructural development in terms of (first) canals and (later) railways was crucial to get raw materials to the production facilities and afterwards transport the resulting product to the market. Many private investors leaped into this opportunity, locking in their wealth in infrastructural investment. With the investments thus sank, the incentives for the crown increased to expropriate these infrastructure investments and reap the benefits of their use for itself. An attempt was thus made to wield de jure political power to transfer economic resources from the private investors to the crown. De facto political power of the private investors in canals and railways was sufficiently strong to resist this push, however. Hence private property rights were effectively protected using transitory economic clout. This outcome stimulated continued private investment in infrastructural development and has greatly facilitated the speed of English industrialization.[3]

Compare this historical victory of the protection of private property rights with the Russian approach to foreign direct investment in natural resource extraction. Russian authorities repeatedly allowed foreign companies to invest in the application of state-of-the-art resource extraction equipment, which Russia itself sorely lacks, only to expropriate the investments made at some point down the road. As a result, Russian resource extraction still lags its potential, whilst the private incentives thus shaped are cast widely across Russian society and produce a less than mediocre business environment and context for generating lasting economic progress.

Also compare these examples to the ever-changing nature of institutional quality. The contemporaneous challenge to provide global intellectual property rights protection for instance is generated by changing economic fortunes. Fortunes that favored the Googles and Facebooks of today’s world, who now wield their de facto political influence to defend existing intellectual property rights and attempt to see them extended also into less accommodative environments like the Chinese society. Their success or failure to solidify their contemporaneous political power in to a changed institutional context will define the environment in which future entrepreneurs pursue their economic fortunes and drive economic progress.

Conclusion: institutions lead economic development over the relevant policy horizon

Institutional quality is both hard to measure and hotly debated as a driving force of economic progress. Regarding the operationalization of the institutional quality indicator, we have therefore constructed it from a broad range of publicly available, objectively assembled data. Applying this indicator to the European continent, it does a good job at first glance in discriminating countries in high and low development in terms of economic progress.

The crucial element in debate on the direction of causation between institutional quality and economic progress is that while economic fortunes may drive institutional changes in the long run, these institutions are slow to change. At the same time, these institutions importantly shape the context within which economic outcomes are derived. Hence institutional quality provides the context and greatly determines economic outcomes. Also, institutions are solidified political and economic power distributions and thus by their very nature are slower to change than economic fortunes. Put differently, institutional quality is more resilient, is the state variable, that drives economic change. Thus over the relevant policy horizon, the direction of causation runs from institutional quality to economic outcomes, rather than the other way around.

How do institutions transmit into economic development?

Although by now there is a common understanding that institutions and long term economic growth are intimately linked, there remains much debate about the precise nature of this interlinkage. Are institutions really the contextual framework within which economic growth thrives or falters? Or should we view institutional quality as yet one more factor of production, augmenting physical and human capital and the stage of technological development? And if so, can we really measure the institutional contribution to growth and development, with institutions being as broadly defined as they are?

In the sections below we’ll first discuss the contrasting views on the institutional role in economic growth and development. Then, we’ll draw upon historical narrative as well as news insights on the dispersion of breakthrough technologies to support our view that institutional quality should really be seen and treated as a contextual variable. This has important implications for the empirical strategy that we can adopt later on to deepen our understanding of the institutional impact on long term economic growth.

Contrasting views on the institutional role in development

The literature is not clear on the way in which institutional quality influences growth. On the one hand, Acemoglu and Robinson (2013) describe the institutional impact from a historical point of view as contextual; proving the environment that enables growth when institutional quality is sufficiently high, yet inhibiting growth otherwise. On the other hand, much of the econometric research includes institutional quality indicators in regressions explaining growth, thus implicitly treating it as an additional factor of production. We first have a closer look at these two fundamentally different views in turn. The next section will explore additional empirical evidence to pick our side in this debate.

I: Institutional quality is a factor of production

It seems logical to explore the explanatory role of institutional quality in cross country variations in economic growth. Intuitively, if institutional quality matters and makes people invest in their own economic fortunes, then long term economic growth should be higher in those countries where initial institutional quality was higher. Indeed, Glaeser et al. (2004) relate long term growth (growth in real GDP per capita between 1960 and 2000) to a set of control variables containing also indicators for institutional quality. While Glaeser et al. (2004) challenge the validity of most of the indicators used in the literature, they generally report a positive impact of institutional quality in growth performance over the four decades spanned by their analysis. Their empirical view is thus that institutional quality contributes to the factors that generate economic growth, hence should be considered as a factor of production.

In a more recent contribution, Góes (2015) makes this connection more explicit. Góes (2015) relates institutional quality to the per capita level of real income in a country. He incorporates the feedback loops between institutions and growth in line with the Acemoglu, Johnson and Robinson (2005) framework by means of empirically applying a structural vector autoregression (SVAR). Based on his results from 119 countries observed from 2002 to 2012, Góes shows that exogenous innovations to his proxy for institutional quality (Economic Freedom of the World Index[4]) have a positive and statistically significant effect on real per capita income levels. On average, a 1 percent shock in institutional quality leads to a peak 1.7 percent increase in real income per capita after six years. Other than Glaeser et al. (2004), Góes (2015) therefore explicitly translates an institutional quality ‘treatment’ into an ‘effect’ in terms of the level of income per capita. Moreover, he demonstrates that his results display different dynamics for advanced economies and developing countries; for advanced economies the peak statistically significant response is only 0.35, whilst for developing countries, the peak statistically significant response is 2.6 percent. This, in his own words, suggests “diminishing returns to institutional quality improvements” (p. 2).[5]

Hence, according to Góes, institutional quality is a factor of production and also one that displays diminishing returns. Institutional quality is thus more productive at lower levels and can be ‘exhausted’ when it reaches high levels. Note that papers such as for instance Gordon (2012) also follow this line of reasoning, specifically where it is argued that the growth generating effects of educational attainment (further facilitating technology creation and adoption) and globalization (pushing out the extent of the market that by Adam Smith limits the division of labour) are tapering off when these magnitudes reach plateaus.

II: Institutional quality is an enabling factor for economic growth

The view described above contrasts sharply with Acemoglu et al. (2005) and Acemoglu and Robinsons (2013) who position institutions as a state variable, from which structurally higher rates of economic growth and technical change can be realized. Higher institutional quality in their view unlocks potentially unlimited economic growth by providing individual entrepreneurs and inventors with the protection of their physical and intellectual property, so that they invest in disruptions of the contemporaneous economic (and political) system with their new products and / or innovations. This resonates with the views on economic growth that emanate from the seminal works of Adam Smith (1776) and Friedrich von Hayek (1948), both of whom build their economic views on the individual that is constantly looking for improvements to contemporaneous organization and technology, generating long term growth in economic welfare “as if guided by an invisible hand” (in Smiths’ words). The crucial element is that the invisible hand is now identified as consisting of the institutional context that allows these individuals to pursue their drive for individual improvement, protecting the investments they make to generate profits for themselves and technological innovations to the economic system.

Hence the argument is that an improvement in institutional quality – i.e. a level shift in institutional quality – will raise the growth potential of an economy, rather than merely lifting the level of economic activity by a limited amount. Or put yet differently, a change in the state variable ‘institutional quality’ will drive a change in the dynamics of economic progress. This also links up with Glaeser et al. (2004), who demonstrate the statistical property of stationarity of institutional quality, meaning that in and of itself institutional quality is not on an upward or downward trajectory over the relevant statistical horizon, whilst economic activity is non-stationary, i.e. is statistically characterized as having a constant rate of growth rather than a constant level. Such a view on institutional change unleashing forces of lasting economic growth are also fully in line with the literature on increasing returns to scale production technologies (Romer, 1986) by compound innovations and endogenous growth (Lucas, 1993).

Calling on data and history to break the impasse

Using data from the last two centuries for twenty-five major technologies, Comin and Mestieri (2014) document two facts that shed light on the debate discussed above on the way in which institutional quality impacts on growth: there has been convergence in the adoption lags for major technologies between rich and poor countries, while there has been divergence in the penetration rates of the same major technologies within countries. Why is it that technology diffusion between countries has picked up speed over the past two centuries, whilst technology diffusion within countries apparently lost speed? Put differently: what makes some countries better positioned to adopt the ever easier accessible frontier technology within their own economic systems than other countries?

The answer lies in high quality institutions, featured by inclusive institutions, effectively protecting property rights and entailing well-anchored individual rights. And this has important implications for the further understanding of the role of institutional quality in economic progress.

It is widely recognized in the economic literature that economic growth, technological change and economic prosperity are brought about by creative destruction. Disruptive innovations change the way we do things, but generate losers and winners across society. Acemoglu and Robinson (2013) characterize the resistance to technical change and creative destruction by the Luddites, the organized opposition of artisans against the newly invented machines that threatened to destroy their jobs. Such opposition may be the more visible type, whilst the political opposition of the elite that fears to lose political power is usually a more formidable force against the welfare enhancing forces of creative destruction. And the latter may in fact go a long way into explaining the lack of accelerating penetration rates of break through technologies over the past two centuries. Loss of political power may result for instance from a growing middle class that emerges from a process of accelerated economic growth driven by creative destruction (this resonates all too well with the contemporaneous Chinese economic transition for instance), but may also be the outcome of rising social unrest that follows the massive destruction of low-skilled jobs. As Acemoglu and Robinson (2013) phrase it, “growth thus moves forward only if not blocked by the economic losers who anticipate that their economic privileges will be lost and by the political losers who fear that their political power will be eroded” (p. 86).

Regardless of the precise shape and origin of the opposition to creative destruction, the above reasoning only serves to underscore the importance of institutional quality and its resilience in the face of such opposition to generate long term economic progress. And more importantly, in the process of overcoming resistance to change, social security systems that compensate the losers and provide them with income support during their transition to new jobs have the effect of reducing the strength or organization of such opposition. The latter aspect also underscores our emphasis before on the relevance of a wider definition of institutional quality.

Conclusion: institutions provide the environment for economic convergence and development

The review of the relevant literature above suggests that institutional quality drives economic progress predominantly by generating an environment conducive to technological change and adoption of innovations and new ways of organizing economic production.

This implies that institutional quality should not be seen as being restricted to ‘yet another’ factor of production that can be embodied in the macroeconomic production technology, but as a contextual variable allowing individual economic agents to generate, adopt and absorb global best practices in their production technology and organization.

In terms of the empirical approach best suited to demonstrate the fundamental role of institutional quality, we should thus be looking not at economic growth of a nation in isolation, but at the rate of convergence of economic prosperity to the global frontier country. Institutional quality makes or breaks a nation’s ability to approach the global frontier by removing the deadlock to change and innovation that may result within a low quality institutional framework. The deadlock emerges from opposition to change organized by those set to lose economic and / or political influence from it.

What do the data tell us about institutional quality and economic development?

Although by now there is a common understanding that institutions and long term economic growth are intimately linked, the relevant economic literature remains divided over the precise nature of this interlinkage. We have argued above that institutional quality generates economic progress more likely than the reverse direction of causation. We have also argued that institutional quality should be viewed and treated in empirical assessments as a contextual variable allowing economic growth, rather than as an additional factor of production operating within existing production technology. We now turn to the litmus test and analyze to what extent institutional quality drives economic progress by generating an environment conducive to technological change and adoption of innovations. In terms of the empirical approach, we look not at economic growth of a nation in isolation, but at the rate of convergence of economic prosperity to the global frontier country.

In the sections below, we’ll take this thesis to the data. We first look into economic growth and unconditional income convergence. Subsequently, we condition income convergence on our measure of institutional quality and find that income convergence kicks in only at reasonably high levels of institutional quality. We then elaborate on the robustness of these results and directions for further research.

Empirical evidence on income convergence and institutional quality

Economic growth is often seen at the success of economic policies pursued by the leadership of a country, despite the fact that empirical evidence demonstrates remarkably little persistence over time of ‘growth miracles’ (Easterly et al. 1992; Easterly et al., 1993; Hausmann et al., 2005; Rodrik, 2006). Indeed as Easterly (2013, p. 217) puts it, “national growth success is usually temporary and quickly reversed.” Figure 4 below demonstrates this feature of the data by plotting average annual per capita GDP growth in the nineties against that in the ‘extended zeros’ (2001-2014) for 196 countries. There’s hardly any relationship between the two, suggesting that whatever successful policies drove growth in the countries that proved successful in the nineties was either reversed (if one believes in the making of the success), or had disappeared a decade later. It must be said that the comparison presented in figure 4 represents a break with the pattern observed in the decades since the sixties, when the data displayed positive correlation between annual average growth between decades, though even then this correlation was weak at best.[6]

Figure 5 displays the extent to which the economies around the world exhibit income convergence, plotting average annual per capita GDP growth in the 2000-2014 period against the starting level of income in 2000 (the latter expressed in natural logarithm). The idea is that all countries should be able to access and adopt frontier technology, while those furthest away from the frontier should be able to generate the fastest growth rates from their catch-up potential pure and simple. Thus the convergence hypothesis states that countries that start out with low income should subsequently be able to generate faster growth rates, suggesting a negative correlation between initial income levels and subsequent annual per capita GDP growth. Indeed, Eichengreen et al. (2012) relate growth slowdowns of fast growing economies –including contemporary China– to per capita GDP levels achieved.

Figure 4: Little persistence in growth …
Figure 4: Little persistence in growth …Source: World Bank, Rabobank
Figure 5: … some pattern in convergence
Figure 5: … some pattern in convergenceSource: World Bank, Rabobank

The figure does exhibit a degree of income convergence in that sense, though the relation is weak. The linear trend line in the figure quantifies this relationship. At the estimate coefficient of -0.40, a country that starts out with per capita GDP 25% above the average, typically exhibits an annual rate of growth that lies 0.1% lower than the average. This is not a particularly strong rate of convergence, as we will also show below. Moreover, the R-squared statistic summarizes the share of the variation in average annual growth rates during the extended zeros that is explained by a country’s initial income position. At a mere 5% it signifies that 95% of the growth variation thus has other sources than a country’s starting position, indicating from another angle that the convergence pattern established in figure 5 is rather weak.

With the period of analysis being the extended zeros, including the Great Financial Crisis, the Great Recession and their prolonged aftermath, we may even have stacked the deck in favor of income convergence, as the crises curtailed economic growth predominantly in the industrialized, i.e. rich parts of the world. Indeed, earlier decades are best characterized by income divergence, rather than convergence. Contributions by for instance Barro (1991) and Mankiw et al. (1992) have actually been inspired by this empirical feature of income divergence. Their papers demonstrated the relevance of ‘conditional’ convergence, meaning that all countries around the world are indeed converging, but each one towards its own steady state, determined by its investment rate in physical and human capital and technology adoption. Conditional on such factors, countries do indeed converge toward their own steady states, and at a speed of approximately 2% per year.

Here is where institutional quality starts playing a crucial role. As we have argued before, supported by sizable parts of the relevant economic literature, institutions determine how conducive the environment is for individuals to drive change, for entrepreneurs to adopt frontier technologies to disrupt local economic conditions. But also, institutional quality determines the incentives people have to invest in physical and human capital to carve out their own economic success and drive macroeconomic growth in the process. Hence one might argue that the only condition for economic convergence is a country’s level of institutional quality. That idea is applied to the data and presented in figure 6 below. The data presented in figure 5 is now splined according to a country’s score on institutional quality (for which we apply the unweighted average of the six World Bank Development Indicators and its Ease of Doing Business Indicator) and color-coded accordingly.

The results presented in figure 6 below (based on data for the 176 countries for which institutional quality is also available) provide a compelling case for the importance of institutional quality in the analysis of income convergence between countries. Economies with below-average institutional quality (colored purple and light blue, capturing the data points with the lowest and second lowest quartiles of institutional quality respectively) display no income convergence at all. Put differently, their growth rates during the 2000-2014 time frame is in no way correlated with their initial income positions and thereby unrelated to their catch-up potential. Consistent with our reasoning of institutional quality providing an enabling environment for innovation and technology adoption, the convergence rates pick up when looking at the countries with above average institutional quality (colored dark blue and orange, capturing the second highest and highest quartiles of institutional quality respectively). Indeed, the rates of income convergence nearly double and more than triple relative to the unconditional convergence results in figure 5 above. For countries in the highest quartile of institutional strength (colored orange), a 25% higher level of initial income corresponds with an average annual growth rate that lies 0.33% lower, thus producing nearly 5% cumulative income convergence over the 14 years analyzed.

Figure 6 : Convergence conditional on institutional quality
Figure 6 : Convergence conditional on institutional qualitySource: World Bank, Rabobank

Note that our results are limited in the sense that whilst they demonstrate stronger convergence results among groups of countries with higher institutional quality, they do not allow us to predict which specific countries in these groups will demonstrate above-average or below-average rates of convergence. The key result is limited to the observation that cf. Acemoglu and Robinson (2013), we can discriminate groups of countries that have created institutional circumstances that position them better to reap the benefits of change and realize their catch-up potential (also see Rodrik (2006) on the limitations of attempting to predict growth successes in advance).

These results present compelling support for our hypothesis, that institutional quality provides an environment conducive to innovation and technology adoption, and more generally an environment that provides individuals with incentives to invest in innovative ideas as well as human and physical capital in order to carve out a better economic future for themselves. The results at this stage should nevertheless be interpreted as initial results, that also leave some questions unanswered while raising a few relating to their robustness to variations in the analysis. We turn to these issues now.

Open issues and robustness considerations

First of all, our analysis remains tentative in using ‘randomly’ splined samples. The nature of the research question allows for more formal econometric techniques that offer more detailed understanding of both what level of institutional quality associates with the start of income convergence and which elements of institutional quality drive these results in particular. Indeed, where we do not know what the different regimes of income convergence look like and we cannot precisely measure institutional quality that determines in what regime we are in, a switching regression framework seems appropriate (Goldfeld and Quandt, 1973). Institutional quality –whole or divided into its seven components– can be entered into the switching function. We see this as an important direction for further research, deepening our understanding of which elements of institutional quality drive the conditional convergence results.

Figure 7: Income convergence and institutional quality in a rolling sample
Figure 7: Income convergence and institutional quality in a rolling sampleSource: World Bank, Rabobank

A way to avoid advanced econometric techniques yet better assess the connection between institutional quality and income convergence is to apply a rolling sample analysis. The rolling sample always consists of 25% of the data set, starts out with the data points that have the lowest institutional quality score, and in each roll i) reports the speed of income convergence and ii) subsequently drops the data point with the lowest institutional quality score while including the data point with the next highest institutional quality score. Thus, based on the institutional quality measure used in figure 6 above, the rolling sample starts out with the lowest 25% of the sample (marking down in figure 7 the 0.09 rate of income convergence that is also reported in the purple trend line in figure 6 above) and moves through the sample to end up with the highest 25% of the sample (and marking down in figure 7 the -1.49 rate of income convergence that is reported in the orange trend line in figure 6). This draws the orange line in figure 7.

The rolling sample analysis provides a better picture of the institutional quality level at which income convergence markedly changes. Figure 7 shows that this seems to be shortly before hitting the third quartile (i.e. the 50th-75th percentile of the institutional quality distribution), where after income convergence become clearly visible.

Another advantage of the rolling sample analysis is that we can easily plot the income convergence line for an alternative indicator for institutional quality and compare the results. That relates to the second consideration, namely the robustness of our results over time. The extended zeros, incorporating the Great Financial Crisis and the Great Recession, may mark a rather specific episode in economic data, raising the question whether the impact of institutional quality on income convergence is similar across decades. Here we run into data limitations in measuring institutional quality for a broad set of countries over sufficiently long periods of time. Complete time series of country-wide institutional quality indicators have only become available in the last fifteen to twenty years, limiting our relevant history to the late nineties at best, a point of limitation also raised by Góes (2015). One way to circumvent this data limitation is to look into other indicators of institutional quality that have longer histories. Kuncic (2013) for instance constructs institutional quality indicators that are available from 1990 onwards.

Figure 8 reports income convergence results in rolling sample analyses using Kuncic’ measures of legal, political and economic institutional quality. Panels A–C use the institutional quality level in 2000 and plot income convergence during the extended zeros. While exhibiting their own profiles, there is a substantial similarity between the curves in panels A and C and the curve based on our own measure of institutional quality in figure 7. The similarity lies in the fact that both indicate that income convergence is concentrated among the high institutional quality countries. Moreover, both indicators mark the range of inflection –the institutional quality range where income convergence starts kicking in– at around 40%-50%. Income convergence looks somewhat different when conditioned on political institutions (panel B of figure 8), but there, too, income convergence is clearly concentrated among the highest institutional quality countries. However, the range of inflection lies substantially higher at around 60%-70%.

Interestingly, Kuncic’ measures for institutional quality run back to 1990, allowing a robustness check of our results with those obtained for the 1990s. Panels D–F therefore use the institutional quality levels in 1990 and plot income convergence during the 1990s. Now the convergence curves look much less convincing in the sense that for neither legal nor political institutional quality is there any clear relationship between initial institutional quality and the speed of subsequently observed income convergence. When applied to economic institutional quality (panel F of figure 8), though, there is weak support for receding income divergence when institutional quality improves, moving towards (weak) convergence in the very top levels of institutional quality.

All in all, the robustness check using Kuncic’ measures of institutional quality reveals that are results for the extended zeros are robust to alternative indicators of institutional quality. At the same time, the robustness over more extended periods of time requires further analysis.

                   Figure 8: Income convergence conditional on Kuncic’ measures of institutional quality        
                                 2000-2014                                                                                       1990s

Panel A: Legal institutions
Panel A: Legal institutionsSource: Kuncic (2013), World Bank, Rabobank
Panel D: Legal institutions
Panel D: Legal institutionsSource: Kuncic (2013), World Bank, Rabobank
Panel B: Political institutions
Panel B: Political institutionsSource: Kuncic (2013), World Bank, Rabobank
Panel E: Political institutions
Panel E: Political institutionsSource: Kuncic (2013), World Bank, Rabobank
Panel C: Economic institutions
Panel C: Economic institutionsSource: Kuncic (2013), World Bank, Rabobank
Panel F: Economic institutions
Panel F: Economic institutionsSource: Kuncic (2013), World Bank, Rabobank

It must be realized that institutional quality still correlates with per capita income levels, as figure 9 readily displays, augmenting the rolling window analysis using institutional quality and reporting the average per capita income level observed in the rolling sub samples. The figure clearly shows that lift off in terms of income convergence coincides with a regime shift towards a trend of steadily rising per capita incomes.

Figure 9: Rolling windows and level of income
Figure 9: Rolling windows and level of incomeSource: World Bank, Rabobank

Note though, that this relates to initial income, whist the convergence results pertain to subsequent average per capita growth over more than a decade. Our analysis and results thus sit squarely in the debate on the direction of causation, wherein we have argued that over the relevant policy horizon, this direction runs from institutional quality to income convergence.

Conclusion: institutional quality allows countries to achieve long term income convergence

We have assessed to what extent institutional quality drives economic progress by generating an environment conducive to technological change and adoption of innovations and new ways of organizing economic production. In terms of the empirical approach, we have looked at the rate of convergence of economic prosperity to the global frontier. While our analysis so far is rough, leaving open many issues that warrant further analysis, we present clear preliminary results that suggest that institutional quality matters for countries to achieve long term income convergence, to adopt available technologies that are superior to those currently in use, and reap the benefits of their catch-up potential.


[1] This section draws heavily from the institutional description in Briegel and Bruinshoofd (2015).

[2] Both sides acknowledge the significance of colonial origins as well as geographical and ecological advantages or disadvantages (e.g. Diamond, 1997; Easterly and Levine, 2003).

[3] This does not disregard the path-dependencies in institutional development and economic progress. See for instance Allen (2009) for an analysis that of how economic forces drove the institutional changes that curbed the power of the Crown and thus facilitated the industrial revolution in England.

[4] Fraser Institute's Economic Freedom of the World Index (EFW) takes into account five institutions-related subcomponents, namely: legal system reliability, monetary stability, burden of regulation, size of government, and freedom to trade internationally.

[5] Hence at the same time supporting the view that institutional development leads economic progress.

[6] The Great Financial Crisis of 2008/09 and its prolonged aftermath may have played a role here, or the collapse of communism in the nineties generated a rather fat left tale in the growth distribution in the nineties.


Figure 9: Convergence conditional in institutional quality in four separate figures

Panel A: Low institutional quality
Panel A: Low institutional qualitySource: World Bank, Rabobank
Panel B: Medium-low institutional quality
Panel B: Medium-low institutional qualitySource: World Bank, Rabobank
Panel C: Medium-high institutional quality
Panel C: Medium-high institutional qualitySource: World Bank, Rabobank
Panel D: High institutional quality
Panel D: High institutional qualitySource: World Bank, Rabobank


Acemoglu, D., Johnson, S. and J. A. Robinson (2001), ‘The Colonial Origins of Comparative Development : An Empirical Investigation,’ American Economic Review 91(5), 1369–1401.

Acemoglu, D., Johnson, S. and J. A. Robinson (2002), ‘Reversal of Fortune: Geography and Institutions in the Making of the Modern World Income Distribution,’ Quarterly Journal of Economics 117(4), 1231–1294.

Acemoglu, D., Johnson, S. and J. A. Robinson (2005), ‘Institutions as a Fundamental Cause of Long-Run Growth,’ in: Aghion, P. and S. N. Durlauf (eds.), Handbook of Economic Growth, Volume 1A, Elsevier B.V., 385–472.

Acemoglu, D. and J. A. Robinson (2013), Why Nations Fail: The Origins of Power, Prosperity and Poverty, Profile Books: London.

Alesina, A., Ozler, S., Roubini, N. and P. Swagel (1996), ‘Political Instability and Economic Growth,’ Journal of Economic Growth 1(2), 189–212.

Algan, Y. and P. Cahuc (2010), ‘Inherited trust and growth,’ American Economic Review 100(5), 2060–2092.

Allen, R. C. (2009), ‘The Industrial Revolution in Miniature: The Spinning Jenny in Britain, France, and India,’ The Journal of Economic History 69(4), 901–927.

Barro, R. J. (1991), ‘Economic Growth in a Cross-Section of Nations,’ Quarterly Journal of Economics 106(2), 407–443.

Briegel, F. and W.A. Bruinshoofd (2015), Institutional quality in Europe: diverging trends in challenging economic times, Rabobank Economic Report, 7 May.

Comin, D. and M. Mestieri (2014), If technology has arrived everywhere, why has income diverged? mimeo April 2014.

Diamond, J. M. (1997), Guns, Germs and Steel: The Fate of Human Societies, W. W. Norton & Co, New York.

Easterly, W. (2001), The Elusive Quest for Growth: Economists’ Adventures and Misadventures in the Tropics, The MIT Press, Cambridge (MA).

Easterly, W. (2013), The Tyranny of Experts: Economists, Dictators, and the Forgotten Rights of the Poor, Basic Books, New York.

Easterly, W., King, R., Levine, R., and S. Rebelo (1992), How Do National Policies Affect Long-Run Growth? A Research Agenda, World Bank Discussion Paper 164, Washington, DC: The World Bank.

Easterly, W., Kremer, M., Pritchett, L. and L. H. Summers (1993), ‘Good planning or good luck?,’ Journal of Monetary Economics 32(3), 459–483.

Easterly, W. and R. Levine (2003), ‘Tropics, Germs, and Crops: How Endowments Influence Economic Development,’ Journal of Monetary Economics 50(1), 3–39.

Eichengreen, B. Park, D., and K. Shin (2012), ‘When Fast-Growing Economies Slow Down: International Evidence and Implications for China,’ Asean Economic Papers 11(1), 42–87.

Fabro, G. and J. Aixalá (2013), ‘Do the Models of Institutional Quality Differ According to the Income Level of the Countries? The Case of the Low-Income Countries,’ Review of Public Economics 206(3), 11–26.

Giordano, R., Lanau, S., Tommasino, P. and P. Topalova (2015), Does Public Sector Inefficiency Constrain Firm Productivity: Evidence from Italian Provinces, International Monetary Fund Working Paper WP/15/168.

Glaeser, E., La Porta, R., Lopez-de-Silanes, F. and A. Schleifer (2004), ‘Do Institutions Cause Growth?,’ Journal of Economic Growth 9(3), 271–303.

Góes, C. (2015), Institutions and Growth: a GMM/IV Panel VAR Approach, IMF Working Paper WP/15/174.

Goldfeld, S. M. and R. E. Quandt (1973), ‘The estimation of structural shifts by switching regressions,’ Economic Research Program Research Memorandum 147, Princeton University.

Gordon, R. (2012), Is US Economic Growth Over? Faltering Innovations Confronts Six Headwinds, National Bureau for Economic Research Working Paper 18315.

Hausmann, R., Pritchett, L., and D. Rodrik (2005), ‘Growth Accelerations,’ Journal of Economic Growth 10(4), 303–329.

Hayek, F. A. (1948), Individualism and Economic Order, The University of Chicago Press, Chicago (IL).

IMF (2003), World Economic Outlook: Growth and Institutions, Washington: International Monetary Fund.

Kuncic, A. (2014), ‘Institutional Quality Dataset,’ Journal of Institutional Economics 10(1), 135–161.

Lucas, R. E. Jr. (1993), ‘Making a Miracle,’ Econometrica 61(2), 251–272.

Mankiw, N. G., Romer, D., and D. N. Weil (1992), ‘A Contribution to the Empirics of Economic Growth,’ Quarterly Journal of Economics 107(2), 407–437.

North, D. C. (1981), Structure and Change in Economic History, W. W. Norton & Co, New York.

North, D. C. (1990), Institutions, Institutional Change, and Economic Performance, Cambridge University Press, New York.

Rodrik, D. (2006), ‘Goodbye Washington Consensus, Hello Washington Consensus? A Review of the World Bank’s “Economic Growth in the 1990s: Learning from a Decade of Reform”,’ Journal of Economic Literature 44(4), 973–987.

Romer, P. M. (1986), ‘Increasing Returns and Long-Run Growth,’ Journal of Political Economy 94(5), 1002–1037.]

Smith, A. (1776), The Wealth of Nations: Books I – III, Penguin Classics, London, 1999.

Solow, R. M. (1956), ‘A Contribution to the Theory of Economic Growth,’ Quarterly Journal of Economics 70(1), 65–94.

Allard Bruinshoofd
Rabobank KEO

naar boven