1For a long time, focusing on institutions in the domain of the theory of growth and development has been the hallmark of heterodox approaches. It is no longer the case today. A new generation of economists trained in economic modeling and econometrics, has emerged, which shares the idea that institutions play a fundamental role in explaining the causes of economic growth. Some of the contributions of these economists provide a major source of inspiration for the international organizations (IMF, WB…) and for the new policies prescribed for developing countries. These contributions put forward new coordination mechanisms and evolutionary economic processes, which contrast with the usual market mechanisms and the steady-state equilibria of the traditional theory of growth. This paper offers an overview and an inquiry into what I will call the New New Institutional Economics (from now NNIE). The implications of this NNIE for the development policies recommended or requested by international organizations are also discussed. Those organizations have adopted an institutionalist point of view but this one does not include all the refinements the NNIE analysis implies. The first part of the paper looks into the genesis of the institutional conversion of those organizations dedicated to the problems of economic development. On the one hand, the merit of the emergence of this conception is due to North (1990) who gave birth not only to NIE, conjointly with Williamson and Coase, but also to a major part of the new trends in institutional economics. On the other hand, the market oriented transitional policies in East Europe countries and their spectacular failure has contributed to promote more institutional-based policies. Those events have boosted the credibility of the NNIE and reinforced this research program. The second part of the paper enters into some details of the NNIE’s analysis. Attention is paid to the main institutionalists’ generic themes rather than to the details of each contribution – which are often heterogeneous. The major topics selected from this literature to be exposed in the paper are the impact of informal institutions and social norms in economic process; the interaction between politic rules and economic rules; the conception of the institutional change from an evolutionary point of view. However, compared to the NNIE’s analysis, we will see that international organizations take for granted a more simple view to put capitalist institutions in place.
2The analysis and the recommendations of the economics of development that were previously focused on the role of markets – inspired by the standard economics – are now centered on institutions. Then, there is a growing literature devoted to the analysis of institutions and their functions in economies, both from a micro and a macro point of view, which I call the NNIE.
3In the following section, the move from market to institutions both in the theory and the political recommendations is taken, on the one hand, as a consequence of D. North’s contribution to institutional economics and, on the other hand, as a result of some failures in the transition experiments realized in countries where policies were in conformity with the main prescriptions of international organizations. These failures have been at the origin of an important change in policy orientations that have moved from “privatization, liberalization and stabilization” – encouraged in the 1980’s and also at the origin of the debates which have promoted the NNIE – to “governance reforms”.
4As already stressed in our introduction, the origin of the NNIE can be attributed to the seminal contributions of Douglas North (1990, 2005) to the revival and dissemination of institutional economics. After working in Cliometry, North has made a great contribution to New Institutional Economics (NIE) jointly with Coase and Williamson, applying contractual analysis (transaction costs, agent relationship...) to the theory of long-term growth. The magnitude to which the North’s approach to economic growth and development differs strongly from the mainstream conception is discussed. It is admitted that the North’s views have evolved from a conceptualization much in conformity with neoclassical economics to a less traditional analysis. Moreover, his analysis has been widened so much that it is now showing some similarities with older and heterodox approaches like those of the American institutionalism (Desquech, 2002 ; Rutherford, 1995). Whatever the changing orientation taken by North’s analysis, the emphasis on the main role of institutions in the long-term growth and in economic development remains an original point of view. Modern economic growth theories, like the so called “endogenous economic growth theories” try to show that economic growth mainly depends on the capacities of the various countries to produce technological innovations, in relation to public utilities, infrastructure endowments and the standard of manpower education. However, according to North these factors only provide the immediate, but not the fundamental causes of growth. North argues that beyond these factors, the main explanation for the differences in growth paths and rhythms between countries lies in the differences in their institutional architectures. If physical infrastructures correspond to the “hardware” side of the economy, institutions provide its “software” side (Johnson and Subramanian, 2005). North moreover explains that institutions are lasting and self-enforcing (path-dependent) and, consequently, institutions also contribute to explain long-term divergences in international growth trajectories. North also contributes to clarify how institutions affect economic performances. Institutions bring above all security and reduce uncertainty associated to all kinds of economic transactions. In addition, they generate some incentives for economic actions as, for instance, capital accumulation or education efforts since they condition payoffs for actions and investment. Economic evolution leads to an extension of markets and to long-distance and anonymous transactions that replace the past personal and repetitive relations. The growing number of impersonal transactions implies the introduction of security measures in order to replace the social sanctions, which were traditionally used in the case of broken commitments. Within this context, institutions play the role of enforcement for contracts and legal rules.
5Following North’s approach, more recent economic contributions look for empirical and econometrical confirmation of “the primacy of institutions” over other determinants in growth and development (Rodrik et al., 2004). On the one hand, these economists try to show that institutions play a more major role than other fundamental factors for long-term growth, notably geographical factors. In contrast, most of the contributions dedicated to the field of economic development conceptions attribute a major role to geography and consequently to natural endowments. Actually, there is a correlation between geographical location and development. However the econometric tests point out the predominance of the “quality of institutions” on the level of income. The influence of other factors like geography or integration in international trade appears to be only indirect via the influence on the quality of institutions. Another argument in favor of institutional factors over natural endowment is “the reversal of fortune” in economic prosperity. This “reversal of fortune” is illustrated by the case of the richer civilizations in the 1500s, which have become among the poorer countries of today (Acemoglu et al., 2005). Conversely, North America and other less developed territories in the 1500s are now among the richest in the world. A first consequence is that geographical factors cannot constitute the main cause of growth. The proposed explanation for the reversal in prosperity lies in the difference between both the colonial experiences of these two regions. The richer countries in the 1500s experienced a type of colonization devoted to the exploitation of resources and population. This form of colonization generated institutions that did not promote economic growth but only economic exploitation. On the contrary, the colonization of the less developed “new world” introduced institutions more favorable to promoting economic activity. Therefore, from this standpoint, economic institutions are the essential cause for economic growth and cross-country differences in economic performances over time. The emphasis on human institutions rather than on natural factors and other more traditional factors is the main characteristic of this perspective. Then a first feature of the NNIE is that it brings an empirical assessment and some measures for the Northian institutional economics. But transition experiments can be considered to have also contributed to the fact that in economic analysis, “institutions have moved front and center” (Stiglitz, 2001).
- 1 Quoted by Murell (2005).
6“The transition process has helped to change the very mode of analysis within economics” (Murell, 2005). Coase himself noticed in his Nobel address “the value of including […] institutional factors in the corpus of mainstream economics is made clear by recent events in Eastern Europe. These ex-communist countries are advised to move to a market economy, and their leaders wish to do so, but without the appropriate institutions no market economy of any significance is possible”1. The impact on the economic literature was noteworthy and Murell (2005) shows that the percentage of documents in the econlit database having transition as a keyword and simultaneously using institutions as a keyword climbs from around 6% in 1990 to more than 35% in 2002 (while those with transition and privatization, liberalization and stabilization fall to 10% after a maximum of 20% in 1995). Transition was a major event of the end of the twentieth century. It is the usual term to qualify political and economic changes that followed the fall of communism in Soviet Union and Eastern Europe. This process, which had begun in 1989-1991, implied 29 countries. Kornai compares the transformation that took place in those countries to those that happened from medieval to modern societies (Roland et al., 2008). The fundamental changes in China since 1978 are also associated to the transitional process in such a way that “fundamental institutional change affected about a quarter of the population of the world” (Berglöf and Roland, 2006). In the beginning of the process, international organizations advocated a set of reforms well known as “the Washington Consensus” consisting in the promotion of prices liberalization, of international free trade, of firms privatization, and of stabilization policies (fiscal stringency and “sound monetary” policy). This set of recommendations sometimes is ironically qualified as “the mantras: privatization, liberalization, stabilization”. They consisted in the promotion of market mechanisms in the regulation of most of the domains of economic life in emerging countries and in transition to market economy. But the measures have often been doomed to failure and have hastened the “Asian crises”. During the 1990s, the “good students” of the international organizations were also those who reached the weakest economic performance. During this decade, the growth in these countries was over half of what it was during the decade between 1950 and 1970 (Stiglitz, 2005). Indeed Washington Consensus policies had been associated with increasing inequality, unemployment and instability. Budgetary subsidies decline from 15% to 7% of GBP in two years in Russia (Murell, 2006) contributing to increase poverty. “Shock therapy” (drastic reductions in money supply in order to stop inflation) and standard price liberalization led to output fall (more than 10%) in the former Soviet Union (enterprises were relieved from the previously planned obligations) (Roland, 2008). As Stiglitz sternly states: “The economic theories on which the Washington Consensus rested had long been discredited. My own work on imperfect and asymmetric information and imperfect markets had contributed to undermining the theoretical foundations” (Stiglitz, 2008 p. 140) and finally “ignoring the social and political dimensions (i.e. institutional, I precise) – as the IMF and Washington Consensus have done – is not only bad social policy, it is also bad economic policy” (p. 18)
- 2 Based on Qian Y. (2002), “How Reform Worked in China”, W. Davidson Institute Working Paper, n° 473 (...)
7The most growing countries in the period were, in contrast, whose who had adopted heterodox policies like China, India, Vietnam and other Asian countries. The real GNP per capital of China has increased by nearly 700% between 1978 and 2003. The Chinese economy has opened its foreign trade representing over 60% of its GNP. But China implemented a market system grafted on a planned system, namely a mixed system of private property rights and of domestic firms owned by local governments, integrated into international trade but supported by strong protectionism and special economic zones. Murell (2005) identifies three specificities in Chinese transition2: 1) A dual-track approach to liberalization kept quotas and controlled prices on production up to those levels planned before the reforms while prices and quantities were free beyond planned production. This measure allowed developing production by preserving previous structures of production and avoiding a disruption in production. 2) A system of township-village enterprises protects decentralized property-rights when the state enables to guarantee more formal ones. 3) A fiscal arrangement between central and local governments with high fixed taxes to the center and high marginal rate for the localities. This system yields a steady stream of payments to the central government while the local governments’ interest to the performances of local producers is promoted. India, as well, has developed its economy sheltered from international competition. Most emerging economies grow with the help of active industrial policies and protectionism (Rodrik, 2004a). International comparisons apparently show the benefits of currency undervaluation and very few emergent countries in Asia have independent central banks (Rodrik, 2008). In 1993, the World Bank report on the East Asian miracle recognized that those countries didn’t follow the Washington Consensus principles. Even Chile, which figured as a good and successful student of economic international organizations did not entirely follow a process of “pure” liberalization: it did not fully liberalize its capital markets and kept some taxes over the short term capital inflows which prevent from volatility; it selectively privatized and around 20% of its exports come today from a public enterprise; the government puts emphasis on education and health expenditures in the direction of poor people (and inequality increase but less than in other comparable countries) (Stiglitz, 2005).
8While the most influential economic analysis of transition ignored institutions at the beginning, following debates later put them in the center. “Economists started taking very seriously the idea that the dismal performance of the Russian economy and of most of the former Soviet Union economies in the nineties could be attributed to institutional failures. The transition experience convinced a large part of the economic profession of the importance of institutions as the underpinning of the successful market economy” (Roland et al., 2008). International organizations like the World Bank or the IMF expressed a mea culpa after the deceptive results of the reform they promoted in the 1990s, and are now converted to the idea of the influence of institutions for supporting prosperity. Then “the institution fundamentalists” and the “augmented Washington Consensus” underlying institutional reforms took the place of “the market fundamentalists” (Rodrik, 2006). There was a general recognition that markets cannot be operational without the existence of a set of institutions, that is, rules to support economic activity and that reforms were promoted without considerations of their effects on the governance of contractual relations. The international organizations encountered the D. North’s view by stressing the fact that the performance of the economic organizations crucially depends on an institutional environment advantageous for economic development and growth. International organizations then attributed the transitions difficulties to “failures of governance”, that is, labour market rigidities, education system weaknesses, fiscal system inefficiencies, legal system and corporate governance defect, and substantial corruption. Then the priority for privatization of the State assets, like in Russia, created major production disruptions “while the lack of preexisting market infrastructure of market institutions did not allow the nascent private activity to compensate for the production disruption” (Roland, 2008) but rather see a “larger role for guns and mafia than for contract and court” (Murell, 2005). To avoid this failure of governance, financial loans granted for the developing countries became conditioned to “good governance”, reliable legal system, and protection against the risks of expropriation by the State, struggle against corruption, and improvement in political freedom. “The experience from transition provides strong empirical support for this shift in focus from market and price theory to contracting and the legal, social and political environment of contracting” (Berglöf and Roland, 2006).
9But this conversion to a form of an institutionalist credo is clearly questioned by the NNIE. The main argument invoked by some authors of the NNIE was that past errors can happen again if the existence of evolutionary processes and their effects are not seriously taken into account: economic reform never has systematic effects; a single measure can imply multiple issues, and conversely, various means can generate the same unique effect. In fact, the same institutional functions (coordination, reduction of uncertainty, security…) can be implemented under various institutional forms (Common Law or Civil Law). “The important point is that effective institutional outcomes do not map into unique institutional designs” (Rodrik, 2004b, p. 9). In the same way, if democracy and market economy – broadly speaking – favor economic growth, they are not however the unique solution. The comparison between the perceptions of the economic environment by foreign firms illustrates this point: whereas Russia frankly reformed its economy to pave the way towards a market system, foreign firms still however consider the communist China environment more reliable and more favorable than the Russian one for making business. China’s government was able to convince businessmen that their interests would be protected without the use of any formal law. Moreover, there are also numerous examples of countries that experience growth without “the good institutions” (Johnson et al., 2006). Finally, “the vast majority of growth take-offs are not produced by significant economic reforms, and the vast majority of significant economic reforms do not produce growth take-offs” (Rodrik, 2004a, p. 8).
10There is also some danger for international organizations to repeat old mistakes or to make new ones if they condition their loans to the adoption of standardized and uniform institutional reforms in poor countries. In fact, “it is impossible to have observed the patterns of development around the world during the last two decades and not realized that protection of property rights, contract enforcement, macroeconomic stability, integration into the world economy, and so one can be achieved in a number of different ways” (Rodrik, 2004a, p. 3). This quotation emphasizes that according to NNIE, the main mistake of international organizations when they impose the “bench mark” of Western legislation is the underestimation of the fact that the institutional arrangements are often partly indeterminate. Furthermore international organizations tend to mix up causes and effects regarding institutions and the economy. “The list of institutional reforms describe not what countries need to do in order to develop-the list certainly does not correspond to what today’s advanced countries did during their early development- but where they are likely to end up once they develop” (Rodrik, 2004a, p. 6).
11Then, the emergence of a NNIE can be considered as the result of both the North’s seminal contribution in the domain of economic analysis and of great changes in the economic world during the last decade. However, the NNIE’s message is not totally homogenous and can’t be totally assimilated to what international organizations are now recommending in terms of institutional reforms.
12If international organizations were unable to define efficient economic measures for developing countries it could possibly derive from the difficulty to reach an agreement about what an institution is exactly. We will indeed see that some authors in NNIE extends more and more the boundaries of the domain of definition of an economic institution whereas some other authors and traditional practitioners, in the usual institutional organizations prefer to stick to a more narrow conception. A growing literature is devoted to informal institutions. Those informal institutions are not circumscribed to developing economies but everywhere have a major function in economic activity. Economists also take the relationship between political and economic institutions and the nature of power seriously into account. The nature of institutional change is also very important to be explored.
13According to international organizations, as well as in North’s earlier works, economic institutions are often reduced to property rights and contract legislation. North defines institutions as all types of rules that organize individual behaviors: these rules are submitted to informal constraints as customs, traditions, or taboos, formal constraints as legal rules, property rights or constitutions; and enforcement devices dedicated to the application of all these preceding constraints. North favors formal rules in his early analysis and binds them to economic performances. Formal economic institutions are not only constraints but also incentives to contribute to wealth because “they determine the payoffs” of economic actions. His analysis of long term development in European countries and North America caused by a legal system appropriated to stimulate investment in education, capital and more generally to individual economic activity is well known. As a consequence for developing countries, absolute property rights may reduce expropriation, facilitate gains from trade and act as collateral for credit (De Soto, 2005) and transform property into productive capital. As Stiglitz explains, “the view that institutions (formal I add) arise to fill gaps in the market, and thereby increase economic efficiency, sometimes called the early North view, become very strongly held for a short while in North America – North has now reject it – by many of his early disciples, including some in the inter financial institutions” (Stiglitz, 2001). Many authors have constructed data sets that give a comparative view of institutional performance broadly focused on the legal system (Dajankov, et al., 2003), political system (Kaufmann et al., 2009) or more micro-economics indicators (number of days it takes to register a firm, settle a commercial dispute in courts...) like in the “Doing business” of the World Bank. The Worldwide Governance Indicators show a noticeable improvement in “institutional quality” during the second stage of transition. But this “highly normative conception” that inspires the spirit of reforms recommended by international organizations is today insufficient. “The issue of how institutional reform happens and how to do institutional reform in the context of transition (and development I add) is still poorly understood” (Roland, 2008). Both in Africa and Vietnam, relational contracting and renegotiating are a major way of doing business and they do no trust in law enforcement. However, growth is better in Vietnam and perhaps “it is not the inadequacies of the legal system that constrains Africa’s growth” (Rodrik, 2008). Therefore, not only property rights are neither the exclusive source of economic incentives nor the main explanation for economic performances but also there is not a single but many mechanisms that can implement property rights.
14As previous Stiglitz’s quotation suggests, North (2005) himself contributed recently to widen the usual conception of economic rationality. This is why this author now defines institutions as “mental models” related to the way agents see the working of the economic system positively or normatively. Therefore, these mental models are representations and support beliefs that guide actions. Among these models some are dominant and their domination allows the emergence of institutions. Thus, mental models and institutions are the two facets of the same phenomenon: on the one hand, beliefs are internal representations of the world and, on the other hand, institutions are external manifestations of those representations. Then while cognitive dimension of institution are now seriously considered, informal rules like norms and conventions are as important as formal rules.
15Undoubtedly, the authors of the NNIE pay major attention to the practical aspects implied by the necessity and the nature of institutional reforms. These empirical preoccupations may explain why some of them did not mainly focus on the notion of economic rationality. It can however be noticed that, following and extending North views on mind and knowledge, economists belonging to NNIE show a growing interest in such themes that traditionally pertained to economic sociology, in the tradition of Max Weber. They especially showed that culture defined as “those customary beliefs and values that ethnic, religious, and social groups transmit fairly unchanged from generation” directly affects economic outcomes (Guiso et al., 2006, p. 2). This influence is transmitted through expectations and preferences. These economists refer, for instance, to social capital, to trust, entrepreneurship spirit, attitude vis-à-vis redistribution, saving behavior, attitude towardsrisk, all factors which contribute to promote cooperation, to prepare future decisions and actions and therefore to determine, at least partially, economic dynamics. The utilization of these social attitudes is not new at all in social science. The real novelty lies in the application of usual economic methods as econometric tests – grounded on the World Value Survey –and experiments to show that belonging to a religion or an ethnic group may have a significant impact on some type of behavior more or less favorable to economic activity. Then, “these results support the hypothesis that cultural background plays a role in important economic choices” (ibid, p. 16). It is observable that those analyses do not lack of ambiguity. For some ones, institutions are seen as the social glue which produces cohesion and social dimension seems to come to the fore, bringing about cognitive aptitudes and predispositions. But for other ones, following a rule, conforming to a social norm is understood as a rational behavior, coping with moral hazard, imperfect or incomplete information, and compatible with self-interest hypothesis.
- 3 Now, Bardhan’s recognition of the survival and the importance of relation-based institutions did no (...)
16Furthermore, a good instance of the path followed by NNIE to develop and extend North’s research program can also be found in its analysis of social and economic interaction in modern economies. We indeed know that in mainstream microeconomics, institutions only appear in data as preferences, techniques and endowments. Moreover, we know that in this analytical context the core of a modern economy is reduced to a set of free competitive market interactions between anonymous agents. To a large extent the research program of neo institutionalism did not break with this conception of the economy. This remark is confirmed by the so-called “hold-up problem”. The existence of this problem indeed derives directly from a conception of market capitalism that remains stamped by the predominance of impersonal transactions only coordinated by the price system. Now we know that socio-economics has pointed out how social structures and personal networks affect economic outcomes and how they do so through influencing the nature of information (personal information does not have the same impact and quality as impersonal information), the efficiency of enforcement (reward and punishment) and the level of trust (Granovetter, 2005). It is striking to notice that far from sticking to the atomistic views of mainstream as well as new institutionalist approaches, NNIE is much closer to socio-economics as far as social interactions in a market economy are concerned. The concept of “embeddedness” of economic actions in social structures and networks (rather than impersonal exchange) is explicitly taken into account. To mention only some examples it is argued that ancient “trading diasporas” still survive in contemporary international trade (Bardhan, 2006a, 2006b); that family ownership dominates corporate governance in many countries (La Porta et al., 1999) like in France (Philippon, 2007); or that the gay community significantly contributed to the development of high-tech industry in Silicon Valley (Florida, 2002). Having a family and friends – which provide learning, finance, and networks- as entrepreneurs is one of the most important determinants for becoming oneself an entrepreneur(Roland, 2008).A more precise example might be taken in the field of development economics using Bardhan’s distinction between “relation-based” and “rule–based institutions”. “Rule-based institutions” refer to agent economic interactions supported by law and legal rules, while “relation-based institutions” refer to agent economic interactions related to network, group, community or family belonging. Bardhan shows that far from tending to disappear, “relation-based institutions” has survived the growth of the international trade. Thus, “trading Diasporas” or “clan-based organizations” still persist today and the principles that generate incentives to be fair in business in those types of organization are “multilateral reputation mechanism and informal codes of conduct and enforcement” (Bardhan, 2006a, 2006b3). On the other hand, “alternative coordination mechanisms – the State, the market, the community organizations – all play different roles, sometimes conflicting and sometimes complementary, in overcoming coordination failures, and these roles change in various stages of development in highly context-specific and path-dependent ways. To proclaim the universal superiority of one coordination mechanism over another is naïve, futile and a-historical” (Bardhan, 2006a, p. 4).
17In brief, in contrast with various previous traditional approaches, NNIE explicitly and a priori welcomes in its analytical framework a large (empirically observed) variety of alternative modes of coordination. Therefore, it drastically reduces the distance between economics and sociology. The market ceases to be the unique or even the optimal mode of coordination: economic and social history, economic-sociology and socio-economics can help to find other modes based on the State, on social groups, on networks, … One of the political implications is that, whereas formal institutions can be deliberately changed, informal institutions are less malleable. This could be the key reason for the difficulty of implementing reforms that are strictly oriented by formal institutional change. Then, “transition experience has helped to overcome an a-historic and naive view of institutions as something easily imported and exported” (Berglöf and Roland, 2006) and shows that transplanting political and legal institutions in alien cultural environment is not an easy task. “Resultant policy is a melange of the old and the new, a pattern typical of times of great institutional changes when revolutions in formal rules move for ahead of modifications in informal arrangements and behaviors” (Murell, 1996). In countries which applied “big bang” approach to transition, the establishment of a private sector by the transfer of state-owned firms in private hands suffered from a lack of policy to encourage the development of small private sector and firms in those countries that have mainly suffered from the insecurity of property rights and not from the lack of investment or from credit constraints.
18NNIE does not only incorporate in its analysis empirical facts or concepts related to sociology but also to political science.
19Of course, transition is also a political process and most appraisals of institutional change made by international organizations are concentrated on political indicators. Economic institutions are linked to political institutions. Political institutions are rules; laws and other practices determining ways people accede to political responsibilities. There is a general favorable bias in favor of the idea that democratic institutions promote prosperity. But the reality is not so simple and some non-democratic countries can sometimes grow faster than old democracies. But it is not clear however whether causality goes from democracy to development or from development to democracy. Murell (1991) tells that a Vietnamese official justify single party rule by noting the economic success of Taiwan and Singapore. Here again North has brought some new analytical elements on this subject. But once again, one can distinguish a first from a second North. For the early North, institutional change is driven by the demand of organizations seeking to reduce transaction costs. This conception performs a functionalist approach in which institutions appear to systematically increase economic efficiency. According to Stiglitz, institutions help above all to preserve some power status, vested interests and rents, not to promote efficiency (Stiglitz, 2001). In the later North, the institutional matrix reflects the beliefs of those that are in a situation to dictate the rules of the game. According to his views, political institutions are supported by dominant ideological or cultural beliefs and finally define economic institutions. This conception is stimulating but Aoki (2007) showed that it consisted in defining an institution as “the form of the game” and therefore in considering that institutions are exogenous to the general set of economic processes. In this framework, the economic institutions are finally the result of political institutions.
20NNIE now offers a more endogenous and processual view, which includes the political area and which takes power seriously. “Economics increasingly acknowledge the importance of institutions – the rules of the game in a society – and the nature of political and power struggles that lies behind them” (Rodrik and Rosenzweig, 2009) The framework of Acemoglu, Johnson and Robinson (2005) is convincing and is a good illustration of a more precise analysis. It can be summarized as follows:
21In this framework economic institutions not only determine economic outcomes but also the distribution of resources. This distribution gives power (“de facto power”) to some agents that have the capacity to influence political institutions favorable to the defense of their interests or rents. This is why they favor political institutions (form of government) that afford de jure power and the decision of economic rules (economic institutions) notably the rule of repartition and the rule concerning property rights. Finally those rules affect economic outcomes and distribution in a manner that reinforces the power of the dominant class. The preceding models explain why non-favorable institutions can persist if they are compatible with the conservation of a rent for the dominant class when “elite has a clear incentive to maintain existing power and distributional relationships trough institutions and institutional arrangements” (Stiglitz, 2001).
22Until recently, only heterodox economics (above all Marxism) regarded power as an essential aspect of an economy while “economists have treated power as the concern of other disciplines and extraneous to economic explanation” (Bowles and Gintis, 2008, p. 2). Even new microeconomics tends to reduce differences in bargaining power to “information asymmetries”. It is no longer the case today since the modern theory of the firm now includes the concept of power (Rajan and Zingalès 1998) and since the exercise of power on competitive markets (goods, labour and credit) is also considered (Bowles and Gintis, 2008). The same evolution can be noticed in the domain of development economics, where power and imperialism were traditionally and exclusively devoted to Marxian approaches. However, in this field, NNIE has now explicitly introduced the concept of power. For instance, Acemoglu and Johnson (2006) have built a model based on a Markov chain process and hysteresis phenomena, devoted to the explanation of both institutional change and persistence. In this framework, they also explain that in developing countries, elites, which possess the economic rents of the country, may endure and conserve the economic institutions favourable to their own interest while changing their de jure into a de facto political power. They show therefore that a country can switch from non-democratic to democratic political institutions but that the de facto political power of elites, based on wealth, strength and ability to solve “collective action problems”, can persist in order to preserve their influence over economic institutions.
23Then NNIE conceives economic dynamics as a process, which combines an economic form of causality stemming from the economic power of individuals and organizations, and an institutional form of causality, which comes from the existence of political power. For NNIE the direction of causality seems to go from the individual to the political power. Before that, the Public Choice School strongly influenced the economic vision of the public sphere in terms of rational choice: from its standpoint, public policies are interpreted as the result of the activity of groups lobbying. In terms of game theory, Aoki (2007) does not define an institution as “the rules of the game” but “as the equilibrium of a game” that is the result of agent interactions. But a less individualist vision of power must be envisaged. If economic power is a source of change, political power provides the source of both the selection of public purposes and rules and the regulation of the economic processes and practices. However public purposes affect individual economic interests and values. Using philosophical language, Searle refers to the “deontic power” of institutions: obligations, authorizations, and interdictions… But these deontic properties of institutions do not only constrain people; they also create new power relations which are in fact not only possibilities but also reasons for actions, yet independent from individual “desires” (Searle, 2005). Therefore, individuals do not always defend their own self-interests; they also follow collective rules and their community’s habits. In addition, they have a political capacity to invoke a general interest and not only their own interests. Therefore political outcomes certainly result from strategic interactions between interest groups, but this is not the whole story: a society is not a simple aggregation of individual “desires” (even if some of unexpected effects are now often taken into consideration).
24The studies devoted to transition learn much about the subtlety of the relations between political institutions and economic efficiency. China’s economic reforms are sustained without fundamental political reforms. Reforms occurred within “transitional and heterodox institution that manage to provide efficient incentive while maintaining the rents for whose politically powerful”. In the same time, “rents may well be a necessary condition for adequate levels of entrepreneurship to emerge” (Rodrik, 2008) whereas “big bang” transitions were developed without mechanisms to finance the restructuring investments.
25The first “Washington Consensus” was based on the idea that economic laws were absolute truths and that they were universal i.e. independent of the historical or cultural context. Its fallacy was probably due to the implicit idea that in a market economy, it is necessary to abandon inefficient customary rules and replace them by market mechanisms seen as the manifestation of natural laws supporting economic efficiency. Neoclassical economics that inspired those policies has always focused on the properties of equilibrium but analytical choice is contradictory with the analysis of structural change that is the main issue of economic development. The second “Washington Consensus” was more cautious concerning the problem of the efficiency of reforms and more sensitive to human distress and poverty. It however remained “universalist” – the universalism of institutions replacing the universalism of market (Gay, 2007). The New Washington Consensus is mainly based on a conception of the role of institutions, which stems from NNIE. Economic efficiency supposes that a country must tend to generate institutions analogous to those that prevail in Western capitalist developed countries. NNIE however develops a less mechanical vision of causality. It seriously considers the problems related to a general definition of “a good institution”. Designing the functions they have to perform is not the same thing that designing the shape they should take. The routes to these institutions are not unique and they are not a unique route between policies and results. Moreover, Rodrik (2008) suggests distinguishing “first best” from “second best” institutions. The first ones are those on which the early North and the NIE focused in order to minimize “transaction costs” (rule of law, democracy...) but, in developing countries, appropriated institutions may be different from those adopted by older industrial countries. Those “best practice” institutions are by nature non-contextual. From this perspective, “efforts at reforming formal contract enforcement institutions should focus on specific categories of firms that do not have access to relational contracting – foreign firms or new firms” (Rodrik, 2008), not on all firms.
26Some of the NNIE authors prefer adopting an evolutionary mode of reasoning rather than a more deterministic one (in terms of a simple cause and effect relations). They consider the diversity of contexts and the fact that institutions often evolve without designers and that what happens is not always what it was expected. These institutions are the product of history and of social and political evolution. This evolutionary process is neither entirely controlled, nor completely self-organised. The actions of individuals and collective actors, like firms or organizations are always and everywhere oriented and coordinated by rules (local or general, informal or formal…). But these actions governed by institutions do not always afford results, which are expected and endogenous innovations can emerge from the process. The new practices that can provide new opportunities for profit and new conflicts of appropriation are selected by authorities and are subject to new rules and regulations. This conception can explain why the construction of market institutions from the outside in developing countries cannot be easy, even if possible. This construction process is generally confronted with internal constraints related to national specificities, which can generate unbalanced situations. The evolutionary perspective also implies the consideration for the pace of institutional change related to “big bang approach to transition” and “gradualist approach to transition” (Roland, 2008). The Chinese reforms made a difference between two sectors: “ nascent private sector are governed by new institutions while some of the old mechanisms of governance prevented from convulsions in old state sectors” (Murell, 2006). From the point of view of the price liberalization, those gradualist reforms prevent poor consumers from price shocks, especially for basic goods they can buy at planned prices in state-owned shops (Roland, 2008). The transition experiments show that gradualism worked better than big bang reforms for adopting best-practice institutions, which strengthens an evolutionary approach.
27The evolutionary perspective helps as well to make a distinction between “fast moving institutions” like politic and legal rules and “slow moving institutions” such as culture and social norms (Roland, 2008) and the difference in pace of change might explain the difficulty of transplanting institutions from rich countries to developing countries.
28Referring to the difference between institutional functions and institutional forms, NNIE increasingly accept the idea that “high-quality institutions can take a multitude of forms and that economic convergence need not necessarily entail convergence in institutional forms” (Rodrik, 2004c, p. 28) and that “even within Europe… there has been only modest convergence among these arrangements in recent years, with the greatest amount of convergence taking place probably in financial market practices and the least in labour market institutions” (ibid.). Furthermore, a market economy is a system supported by a set of linked institutions. These institutions (of various types and various levels) must form a consistent system. More and more economists tend to accept the idea that crucial institutions (like the legal system or the political system) matter less than the self-consistency of a system of complementary institutions. Even in developed countries, market economies are compatible with different national institutional architectures. The idea is that even if “capitalist globalization” favours homogenization, nevertheless various “types of national capitalism” still persist. For instance, Amable (2005) distinguishes five types of capitalism: neo-liberal, continental European, socio-democrat, Mediterranean, the Asian. These “types of capitalism” are differentiated according to the institutional arrangements they generate in the following domains: competitive rules in product markets, labour market legislation and labour organisation, organization of the social security system, organisation of the educative system, organization of the research and innovation system, and organization of the financial system. In this framework, the convergence of various national growth trajectories or technological levels is not conditioned by a strict convergence of national institutional rules and of organizational forms. The idea can be extended to the importing of institutions by developing countries. A same single institution can generate very different outcomes according to the structure of the national institutional context in which it is inserted. Institutional complementarities explain why it is difficult to import structural reforms and behavioural rules that were not previously and endogenously formed. An institutional set up has its own consistency but also implies the existence of some inertia. To sum up, a particular institutional change is only advantageous if it is combined and synchronised with other changes in other domains. Moreover, if the design of institutional reforms seems to be a real challenge, the regulation is nevertheless a necessity.
29Within this context, markets institutions need not only to be created but also regulated: “markets need not be self-creating, self regulating, self-stabilizing, and self-legitimize… markets require extensive regulation, to minimize abuse of market power, internalize externalities, deal with information asymmetries, establish product and safety standards, and so on…” (Rodrik, 2004c p. 27). Rodrik and Subramanian (2003) distinguish several types of market institutions. First, institutions like property rights and laws on contracts are admittedly necessary to generate markets. A second type of institution is devoted to the regulation of market imperfections: public regulation authorities control activities that produce external effects (telecommunications, transportation, public utilities). A third type of institutions is dedicated to the task of stabilizing the macroeconomic evolution of the economy; it includes the Central Bank, as well as exchange regimes, or budget rules for instance. The utility of a last type of institutions is the legitimatization of market where institutions are supposed to organize income redistribution and social protection. We could mention pension systems, social security systems, and unemployment benefit systems. Finally, NNIE does not consider market economies as “spontaneous orders” but rather as networks of rules of various types and different levels that are “market-sustaining institutions”. If some of the NNIE authors are favourable to state regulation, however some other does not take for granted the importance of the State in upholding institutions. The State is often seen as a hindrance to market processes and to the freedom of entrepreneurship. The State is also often considered as the centre of corruption and the major risk of private property expropriation. However there is no market without a State to create and organize it, with the help of legal and working rules. Moreover, the historical perspective shows that the construction of the modern Western States was a precondition for the modernisation of the economy and of the emergence of market mechanisms. Today, experience shows that poorer countries are also those that have the weakest States. Conversely the most dynamic emerging countries are precisely those with have the most structured States.
30The research program we call NNIE sometimes uses quantitative methods like economic modelling and econometrics. This use explains why standard economists more and more accept NNIE. In the same time, it embraces a more extensive view of institutions. According to NNIE, economic rationality is no longer limited to usual rational choice and can include rule following as well as power relations. Its vision of economic process also appears to be less mechanical and deterministic since dynamic (or evolutionary) effects like “cumulative causation”, institutional persistence, unexpected effects are now taken into account. If North is recognized as a founder, ironically, his own research program retrospectively has some common features with older institutionalist traditions, German historical school (Stiglitz, 2000) but also American institutionalism (founded by Veblen and Commons and applied by Myrdall to analyse underdevelopment). To this extent, NNIE appears sometime ambiguously to stand halfway between mainstream economics and OIE. The impact of NNIE is especially strong in the domain of development economics. Today this impact is not entirely assumed by international organisations, which are however more and more driven to include institutional factors when they have to take decisions. Nevertheless, NNIE supposes a less deterministic conception of economic development able to better include the complexities of the economic process. It also assumes that the reality is submitted to a permanent process of change. It must be cautious to not “fetishize” institutions and institutional economics itself! It is admitted that “adequate” institutions are not easy to define and that often “the process of change is dominated by chance rather by choice” (Murell, 2005). It finally implies that economic policies can take specific contexts and experimentation into account. This is why NNIE developments might be useful for the future.