Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Key Points

This book is about clusters of institutions and economic systems. It proposes an original typology of capitalist varieties for those countries whose capitalist nature has seldom benefited from much analysis so far: poor and emerging developing countries. Typologies are sometimes criticized as being based on a limited number of features that then tend to be generalized across national cases. The typologies of mature capitalisms, carefully elaborated in an extensive literature, can, however, scarcely be criticized as over-generalizations (Jackson and Deeg 2006). As for developing economies, they exhibit capitalist features that are likely to be fairly different from the well identified ones that have been shown for the mature country systems. Researchers, therefore, have to face both complex and highly “exotic” institutional systems, with very few tools adapted to analyse them.

A typological approach certainly constitutes a useful approach for identifying and comparing developing country capitalisms, for the simple reason that, so far, there have been very few proposed elaborations. The absence of firmly-based typologies of developing countries’ capitalisms does not mean that those typologies cannot be found. Neither does it mean that such typologies would not be useful if they could be elaborated. We believe, on the contrary, that a typological approach provides researchers with a holistic framework that is capable of tackling expected high cross-country heterogeneity. Addressing the likely “exotic” nature of developing countries’ institutional systems, that is, their possible singularity with respect to the models describing mature capitalist systems, means adopting an agnostic approach. In other words, since ideal-types of emerging capitalisms cannot be a priori derived from existing models or from more formal theoretical elaborations, they should be inferred from a systematic analysis of the sectoral institutional data to be observed for a sufficiently broad sample of developing countries. In order to reconcile the typological approach, commonly used by comparative capitalism (CC), with the rigorous statistical analysis of observed institutional forms and clusters, advocated by the new institutional economics (NIE), the present book has, therefore, adopted an original methodology. We have first addressed the extreme diversity of developing countries’ institutional systems, and then reduced this diversity by clustering our countries into a limited set of models, understood as ideal-types.

Our statistical approach has, accordingly, enabled the diversity of institutional configurations across countries to be assessed and subsequently crystallized. It has also generated various innovative results concerning the varieties of institutional system to be observed in emerging and developing countries. Since the effects of institutions on economic development tend to cluster (Besley and Persson 2011), our core assumption has been that this needs analysing at system level, which involved the mobilizing of several dimensions of institutional governance: agriculture, education, environment, finance, product market, labour and social protection. Our approach, by thus departing from the NIE empirical logic, with its use of one-dimensional scalars to represent the whole socioeconomic system, offers an alternative contribution to the widespread literature on the institutional dimension of economic development.

At this point, we are now able to address the main questions raised in the Introduction. How can we empirically analyse developing countries’ systems of socioeconomic governance? What types of institutional complementarities are revealed by this analysis? Are there sufficiently marked, homogeneous and consistent clusters of countries to enable various models of capitalism across developing and emerging countries to be differentiated? These questions are addressed in the next two sections (Sects. 14.2 and 14.3). We then proceed to highlight two key aspects of the cross-country institutional discrepancies that emerge in Sect. 14.4. The first of these concerns the type of state intervention in socioeconomic governance; the second aspect refers to the role played by experimentation in shaping this type of socioeconomic governance. The policy implications of our results are then addressed in Sect. 14.5, with special focus being put on institutional reforms in poor countries. Finally, new horizons for research are suggested in Sect. 14.6.

2 Institutional Clusters and Complementarities

In line with the CC literature overviewed in Chap. 2, the approach adopted in this book has been geared towards explicitly assessing how sectoral institutions actually coalesce at system level, in complementary or non-complementary fashion. According to the standard definition, two institutions are complementary if they demonstrate higher efficiency, when combined, than when separated. Although institutional complementarities have been extensively studied in CC literature, those studies have almost always used pre-defined ideal-typical institutional models. The CC ideal-typical approach underlying institutional complementarities results in institutional isomorphism being applied indiscriminately to all sectors of the institutional system.

The approach adopted here, however, has chosen to identify and compare developing countries’ capitalisms by clustering them in terms of their sector-specific regulation. This meant considering a flexible notion of complementarities, one that is necessarily more akin to the idea of institutional coalescence or congruence than to the standard definition. Developing countries’ institutional systems tend to associate sectoral regulations that have been articulated over time on a somewhat makeshift basis. Most developing countries, in fact, show essentially transitional and hybrid institutional systems, whose functionality is generally difficult to assess, because the researcher has very few theoretical criteria concerning what is complementary or not. It was this lack of clear criteria that led us to define two types of institutional complementarities: de jure and de facto.

De jure complementarities concern those to be expected on purely theoretical grounds. These complementarities generally correspond to a theoretical construct that is strongly inspired by a first-best functionalist logic: (1) each specific and isolated institution is designed ex ante to minimize transaction costs for the sake of collective efficiency, and (2) sectoral institutions are considered as most effective when isomorphic to the others. De facto complementarities are, conversely, defined as institutional arrangements lacking a priori theoretical justification on first-best or isomorphic grounds. They result, instead, from an institutional experimentation process, with the hybrid system’s functional efficiency appearing ex post, sometimes rather unexpectedly.

Bearing all this in mind, we went on to analyse each sector (labour, competition, finance, social protection, education and training, agriculture and environment) separately. The sample countries could thus be clustered with respect to their similarities in terms of the institutional variables governing each specific sector. As this procedure was replicated for all seven sectors, a corresponding set of seven institutional governance types, one per sector, was ascribed to each country, in order to depict its institutional system (see general Appendix). At a second stage, the sample countries could then be clustered with respect to their specific sets of sectoral institutional governance types. The resulting clusters could finally be characterized and labelled as distinctive models of capitalism. Each cluster, therefore, groups the countries that are most similar in terms of their socioeconomic system.

Although emerging market economies share similar economic performances and tend to be perceived as somewhat undifferentiated institutional systems, our analysis shows that they have very different institutional characteristics and cannot, accordingly, be treated as one group of homogeneous countries. Several crucial lines of differentiation between these various types of emerging capitalist economies are identified in the present work: the degree of institutional formalization, the type of state intervention in the socioeconomic system and, more surprisingly, the degree of institutional experimentation. Moreover, various forms of non-isomorphic institutional congruence have emerged from the analysis, with these de facto complementarities being functionally efficient, even though they significantly depart from de jure complementarities. This finding alone amply justifies our chosen approach, since these de facto complementarities would have been inaccessible to traditional ideal-typical approaches based on a priori defined models. The high degree of institutional interaction within and between each institutional domain, as well as the strong likelihood of unexpected de facto complementarities in developing countries, cast doubts on the relevance of the ideal-type-based approach in the context of emerging capitalisms. As a result, a priori typologies, like those proposed by the CC literature, generally built around broadly drawn types of regional capitalisms, are unable to seize unexpected or unorthodox configurations.

3 The 2+4 Models of Capitalism

Our analysis led to six (2+4) models of capitalism being identified: the Liberal Market, Coordinated Market, Globalization-Friendly, Statist (Resource Dependent), Hybrid-Idiosyncratic and Informal models. The two OECD country models, Liberal Market and Coordinated Market Economies, are mainly differentiated by their labour market regulation, centrally Coordinated for the latter and Deregulated for the former and, to a lesser extent, by their financial systems. Liberal Market Economies are characterized by contractual and flexible capital-labour relationship, a high degree of market competition, liberal social protection and a deep and broad financial market. Coordinated Market Economies articulate coordinated labour market, decommodified social protection, democratic universal education, a liberalized competitive product market and intermediate bank-oriented finance sectoral models of governance, together with effective environmental governance and highly formalized and productive agriculture.

As for emerging economies, some of them share common features with the OECD group. The Copenhagen criteria have, unsurprisingly, led to a form of institutional convergence for the Central and Eastern European Countries (Bulgaria, Czech Republic, Estonia, Hungary, Lithuania, Romania, Slovenia and Slovakia) whose product market regulation needed to comply with Western European standards. These countries massively reformed their institutional systems in order to be integrated into the European Union, thereby converging towards Coordinated Market Economies. Latin American economies, such as Argentina or Chile, also share similarities with Coordinated Market Economies, because the huge market reforms towards liberalization that they undertook during the 1970s and 1980s, led them to converge towards OECD standards of regulation.

Apart from those very few emerging countries that have converged towards the OECD mature capitalisms, the bulk of developing countries can be located in four clusters that are highly specific to emerging capitalisms.

The first cluster, labelled Informal (Weak State), covers most of the poor economies that exhibit similar sets of sector-specific institutional governance types: informal labour market, outward-oriented agriculture, statist-protectionist product market regulation, embryonic finance (mostly informal), social insecurity, weak environmental governance and narrow education. All these sectoral types of governance, characterized by their high degree of informality, tend to coalesce; this is especially true of the poorest countries in our sample. The informal institutional arrangements used, in each sector, as substitutes for the missing state-enforced regulations and rules, are strongly isomorphic. Although informal institutional arrangements do allow the economic system of poor countries to hobble along, they also deliver bad socioeconomic outcomes, such as strong economic insecurity, high social vulnerability or low productivity, which certainly hamper long-term economic development. The strong internal consistency of all these isomorphic dimensional regulations lock the country into an institutional poverty trap, that is, a situation in which private and public investment are not coordinated because the economy lacks the necessary protection and incentives that make investment possible. The Informal (Weak State) model can, therefore, be considered as resulting from regressive de jure complementarities.

The second cluster, called Statist (Resource Dependent), brings together large, densely populated economies such as China, India, Mexico or Russia which, although they have partly liberalized their external trade, remain characterized by massive state regulation of their domestic labour capital and product markets. The Statist (Resource Dependent) systems are characterized by sets of dimensional regulation that are highly isomorphic, with the operation of markets being restrained by heavy regulation and direct state intervention. In most countries of this cluster, high levels of market regulation and social transfers to the poor, often financed by export rents, are commonly used as political and economic risk reducers. The economic rationale is that the state must organize the process of economic modernization by controlling prices and entrepreneurship, with this high level of economic interventionism and loss of freedom being counterbalanced, or legitimized, by high levels of redistribution, for workers and the poor, and by opportunities of regulator capture for capitalists and firms. This group includes most of the large middle-income economies, like China, Egypt or India, which share certain features with poor countries but are, in many important respects, characterized by very singular institutional patterns. First, their product markets have partly opened up to world competition and been liberalized, at least with respect to international trade and investment. Second, they have developed financial systems that are mainly bank-regulated (higher access to financing, higher lender and borrower protection). Third, they show highly discriminating and generally dualistic labour markets, in which widespread informal sectors survive alongside more modern, often extraverted, market segments.Footnote 1 Hence, those economies remain largely under-reformed, their institutional structure corresponding more to that of low-income than of high-income economies.

It is worth noting that such isomorphic statist regulations of labour, capital and product markets can deliver positive outcomes for poor countries, since they are well suited to state-led big push policies. The Statist (Resource Dependent) model seems to be the only one that can be considered as delivering progressive de jure complementarities, at least during the first stage of accumulation-led economic development. Those statist regulations may, nevertheless, become a major stumbling block for middle-income countries facing insufficient output diversification and sophistication (Rougier 2016). The Statist (Resource Dependent) model is strongly determined by path dependency and by the particular mode of state capacity building. The typical mix of high protection and patron-client relationships, directly inherited from the Natural State model discussed by North et al. (2009) is prevalent throughout the system. Unsurprisingly, ex-socialist and natural-resource rich economies are over-represented in this cluster, given the strong historical presence of a centralized state and administration in most of them. When looking at the historical trajectory of those countries, it appears that many have adopted, at some juncture in their development path, a socialist economic (and often political) model, or some alternative based on high degrees of centralization and state control over the economy.

The third cluster, called Globalization-Friendly, is mostly composed of smaller emerging market economies, with highly liberalized economies, as well as high levels of state intervention. The Globalization-Friendly model is characterized by the domination of such sectoral governance types as deregulated labour, education biased towards high school and export-oriented liberalized competition. It is worth noting that the countries classified in this model have generally adopted regulations that optimize their integration into the world economy. In countries like South Africa and Malaysia, the social contract tends to be based on significantly improved access to economic opportunities for broad shares of the population and on a low degree of government-based welfare. The Globalization-Friendly cluster includes small countries that have not experienced similar historical trajectories and that were able to set up their institutional system from scratch, or to carry out reforms independently of their former institutional characteristics. The Globalization-Friendly model also exhibits strong internal consistency, although it has been shaped by a process of experimentation of sectoral institutional governance driven by the common objective of facilitating integration towards international trade and production networks. Accordingly, this cluster shows de facto complementarities that can be considered progressive, since the group’s levels of economic and social outcomes are significantly higher than the sample average. The Globalization-Friendly model shows, for example, better performance than the other three developing country clusters for all governance indicators, and also exhibits higher average levels of real GDP per capita, HDI, openness and integration to globalization.

The fourth cluster, called Hybrid-Idiosyncratic, includes both poor and emerging countries. This model brings together countries whose highly original institutional systems cannot be subsumed under a clear-cut set of distinctive and common features. Our analytical framework has enabled a set of idiosyncratic forms of sectoral institutional governance to be identified at sector level, with these idiosyncratic forms finally tending to cluster at the second stage of the analysis. This cluster also includes countries that exhibit hybrid systems of sectoral models of governance, namely, models that are not usually associated because they do not show de jure complementarities. Hybrid-Idiosyncratic countries have, therefore, experimented original institutional sets that have been able, in some cases, to drive positive outcomes in terms of economic development.

In what concerns the hypothesis of institutional advantage, although our results are inconclusive as to which model shows the best performance in terms of economic growth, Globalization-Friendly countries have, nevertheless, reached higher average levels of GDP per capita, human development and trade integration than the other three clusters including developing countries. Globalization-Friendly countries also recorded higher Gini coefficient levels over the 1990s, before inequality went on to become significantly higher in the Hybrid-Idiosyncratic models during the following decade.Footnote 2 As for the Statist (Resource Dependent) model, despite higher than average population size, it showed significantly lower than average levels of poverty, from 1990 onwards, and of inequality in the mid-2000s. This cluster, as well as the Informal (Weak State), also shows greater reliance than average on agricultural production, with exports being less diversified than the sample average. As for the broader assumption that each institutional system is better at reaching its prioritized goal than the others, it seems that the Globalization-Friendly does reach its goal of higher levels of trade and capital integration,Footnote 3 sometimes at the expense of rising inequality, while the Statist (Resource Dependent) has demonstrated efficiency in durably keeping poverty at low levels, albeit at the expense of microeconomic efficiency.

4 “Stateness” Variations and Institutional Experimentation

It is commonly considered that the degree of state intervention in markets is a critical factor of differentiation between mature capitalisms (Hall and Soskice 2001). As for developing countries, the role of the state is twofold: it should regulate markets, but also trigger long-term economic transformation. Besley and Persson (2011) explain successful economic development by the extent to which the state is able to use its legal and fiscal capacities to invest in public goods while simultaneously regulating sociopolitical violence. In their analysis of the historical formation of committed-to-development states, North et al. (2009) have also argued that equal access to different types of public goods is usually provided in a certain sequence, starting with the rule of law, followed by mass education and infrastructure and, finally, by equal participation in labour markets, including the provision of social insurance systems (Gollwitzer and Quintyn 2012). These three stages possibly correspond respectively to what is referred to in the present book as institutional formalization, investments in public goods and the establishment of more inclusive institutions via the channels of labour regulation and social protection. But this possible matching would require further empirical investigation, which is beyond the scope of the present book.

Almost two decades ago, Evans (1997) had claimed that forms of stateness, defined as “the institutional centrality of the state”, were likely to vary significantly among nations. Accordingly, such stateness requires finer analysis by social scientists, if political behaviour and institutions are to be better understood, especially in developing nations. Our work has demonstrated that stateness should not be measured merely by composite scores of market state regulation, as in the International Country Risk Guide “Investment Profile” or the World Bank Governance Indicator “Quality of regulation” indicators, or the proposed “state capacities” composite indicator of Besley and Persson (2011). The present book shows that the degree of liberalization is not the prime factor of differentiation between the models of emerging capitalism. Institutional formalization, that is, the shift from local informal rules to centrally established and enforced formal rules, as well as the degree of institutional experimentation, have come over as being more decisive in explaining differences between our models.

As for the most basic dimension of stateness, namely, institutional formalization, heterogeneity is high across developing countries. Most countries of the Informal (Weak State) cluster have not, so far, reached the first stage of institutional formalization, which requires centrally-enforced rule of law. Their weak state is unable to invest in the type of public goods that support the emergence of self-sustaining individual and collective preferences for economic development. Statist (Resource Dependent) countries have succeeded in formalizing and enforcing the rule of law, especially for private actors, but their pattern of investment in such public goods as education is counterbalanced by market regulations that are not as inclusive as those to be found in Globalization-Friendly, CME and LME clusters.

Thus, there is not one unique model of “institutional centrality of the state” at work across emerging economies. On the contrary, those economies tend to cluster into three very different models with respect to the role of the state: the Globalization-Friendly model, in which the state is interventionist in order to increase competitiveness; the Statist (Resource Dependent) model, with the state exerting strong control over the economy and actively supporting households’ means of existence; and the Hybrid-Idiosyncratic one, which is more heterogeneous.

It may well be the style of state intervention, therefore, and not its mere intensity, that finally matters most in understanding why some countries have managed to develop, whilst others have not. The present work has notably highlighted the fact that institutional experimentation offers a crucial explanation of the style of state intervention variations. Institutional experimentation was, in fact, demonstrated at two levels. At sector level, original de facto institutional arrangements have emerged. For instance, the export-oriented goods market, upgrading skills-exporting education and private transfer-based social protection, found predominantly in Globalization-Friendly systems, all signal emerging countries’ efforts to adapt their sectoral governance to the requirements of the globalized economy. In addition, it proved possible to identify models of environment regulation such as that of the focus-on-biodiversity. At system level, the Hybrid-Idiosyncratic and Globalization-Friendly models offer two good illustrations of economic systems crafted experimentally, on the basis of de facto institutional complementarities, with these experimental complementarities having delivered, in some cases, strong economic benefits. The Hybrid-Idiosyncratic cluster signals two forms of de facto complementarities: experimentation of innovative forms of sectoral governance, and hybridization of polymorphic institutions across the different sectors of the system.

China probably provides the best illustration of this experimental approach to reforming. Even though this country has finally been classified as Statist (Resource Dependent) by our analysis, its post-1978 institutional reforms have tended to be fundamentally experimental (Qian and Xu 1993). Innovation and experimentation have reinforced the adaptive efficiency of Chinese capitalism’s institutional system (Qian 1999; Ahrens and Jünemann 2011). While the household responsibility system introduced strong microeconomic incentives to increase productivity and trade newly produced goods on nascent markets, the dual-track system allowed a process of smooth transition from state-owned enterprises to private firms, by enhancing the efficiency of the former. Simultaneously, decentralization led to growing fiscal and legal autonomy for local governments, which were held responsible for the financial success of the former State-Owned Enterprises that became transformed into Town Village Enterprises (TVEs), endowed with quasi-private company objectives. Local government could thus implement institutional solutions fully adapted to the specific needs of local actors, those local actors being consequently driven by ever-stronger economic incentives to invest in new activities. Those simultaneous reforms have, in turn, gradually increased the individual and collective opportunity costs of reforms reversals, and provided growing support for openness and privatization, with the initial reluctance regarding such reforms progressively fading away (Lau et al. 1997; Qian 2003; Bardhan 2010). Meanwhile, the domestic market was preserved by strong centrally-governed incentives to limit local protectionist policies, promote trade across provinces, and foster competition between TVEs (Qian and Weingast 1996). The process has been fully adaptive, in the sense that political reformers have always used informative feedback from the productive economy to re-shape institutions, and entrepreneurs have also used their economic power to divert the process of institution building towards their objectives (Ahrens and Jünemann 2011).

Other countries have enacted similar hybrid reforms, although by using fewer experimental ingredients than the Chinese-style dual track and TVEs. The comparative case studies conducted in Chap. 13 have described the various elements driving long-term institutional change via incremental adjustments of sectors of the whole system, thereby leading to the formation of hybrid systems. Malaysia and Brazil have experimented original institutional articulations all along their post-independence trajectory of reforms. Various elements have simultaneously conditioned and oriented the institutional experimentation process: historical critical junctures, development strategy and political economy. Critical junctures concern either political or socioeconomic realignments, that is, periods of bifurcation in socioeconomic or political governance. Such critical realignments inform about the influence that underlying sociopolitical coalitions had on policy choice. Whereas the development policies that were chosen after independence exercised a durable influence on socioeconomic institutions, as in the case of the Mexican Statist-redistributive model, subsequent strategic inflections also had significant consequences on institutional configuration, with parts of the system being liberalized, whilst others remained highly state-regulated. Finally, changes in the dominant sociopolitical equilibrium also dramatically influenced the hybrid and experimental nature of these systems.

What finally differentiates the countries that succeeded and those that failed may be located in the willingness to reform and the autonomy of reforming. In most cases, such locally adapted institutional fitness was progressively set up by an evolutionary process of trial and error, with the institutional system adapting to changes in both the local and external environment. In countries in which trade and capital account liberalization reforms were externally imposed, notably in the context of structural adjustment, such reforms generally resulted in high internal inconsistency and ineffectiveness. This was the case of numerous countries classified as Informal (Weak State) or Hybrid-Idiosyncratic. At system level, Hybrid-Idiosyncratic and Globalization-Friendly systems provide two illustrations of economic systems having been crafted experimentally, on the basis of de facto institutional complementarities, with these experimental complementarities having delivered, in certain cases, large economic benefits.

5 Lessons for Institutional Reform in Developing Countries

Despite the poor effectiveness of top-down benchmarked institutional reforms in developing countries, they have tended to proliferate during the last 30 years (Rodrik 2008; Andrews 2013).Footnote 4 This proliferation is based on the widespread belief that the idealized institutional design drawn from the theory of competitive markets, that is, open markets and free prices, with limited state intervention in production and distribution, is optimal for economic development (Chang 2011). As Evans (2004) puts it, the “monocropping view” of institutional reforms imposes a standardized institutional technology on undifferentiated countries. It assumes that institutional effectiveness is independent of local conditions regarding development level, sociocultural preference or social contract history. Institutions that are transferred to developing countries are, therefore, benchmarked against the allegedly successful institutions of developed countries, with a strong preponderance in favour of those inspired by the LME model (Chang 2011). Certain regions, like Eastern Europe, have adopted the CME benchmark, which can be explained by the influence of the European institutional model on the content of transition countries’ reforms.

At sector level, too, numerous examples of transplantation of Western practices via market or public administration reforms should be mentioned (Andrews 2013). The functionalist approach to institutional reform, which has always been prevalent among international financial institutions and aid agencies, considers that one given function should be assumed by only one type of best-fitted institutional form, whatever the national context. This approach considers that bringing developing countries’ systems in conformity with the institutional frontier mix of institutions is the first best policy (Rodrik 2008). According to this de jure approach to institutional fitness of shape, the minimum level of enforcement of this best-fitted institution should automatically engender a highest expected economic outcome than a higher level of enforcement of any alternative institution. This does not leave much room for institutional experimentation of possible de facto complementarities.

Surprisingly, however, very few developing countries have effectively introduced fully-fledged market-based institutional systems. We were able to identify various alternative forms of functional effectiveness generated by de facto complementary and hybrid sets of institutions, with some of them being highly singular. Many developing countries have, for instance, retained high levels of state intervention, along with progressive and asymmetric market liberalization. Fine-grained case studies would even suggest that most of the systems observed at country-level could, in fact, be typified as being hybrid, and not mere transplantations of Western-style institutional benchmarks. Since a great deal of innovation and experimentation has been observed in post-independence institutional reforms, the dominant benchmark logic thus seems to be contradicted by the historical facts. This apparent contradiction needs to be further elucidated.Footnote 5

One first functional explanation, clearly highlighted by the comparative description of national path of reforms in Chap. 13, is that reforms can be deliberately driven by the search for adaptive efficiency of the institutional system. Top-down technocratic-like reforms, which consist of transplanting allegedly optimal models of institutions, on the grounds that they are supported by economic theory or appear to have been successful in other countries, often fail to improve the institutional system’s overall institutional functioning. Institutional hybridization is a natural response to the high uncertainty about the capacity of such imported institutions as formal contract law or capital account liberalization to fit in with the local context. Hybrid institutions, therefore, manage to increase both the social acceptability of and political support for the new rules. This is what our comparative case studies perfectly illustrate: the shape of an institutional system cannot be understood independently of its underlying historical and political foundations.

The second explanation may be related to the weakness of the commitment to reform by developing countries’ governments. According to Andrews (2013), many developing countries’ governments feel obliged to adopt international standards of institutional reforms, simply because they would not otherwise be given financial help. Incumbent governments struggle for external political and financial support, using their institutional reforms to signal their political ability and willingness to adopt that objective. In such cases, best-practice institutional reforms, like privatization, fiscal rules or meritocratic and performance-based administration may be introduced essentially for signaling reasons, thereby tending to be only superficially or partially implemented, with possible deterrent medium-term functional consequences for the whole institutional system.Footnote 6

In the case of Ghana and Brazil, domestic political leadership has also impacted both the willingness to reform in depth, and the autonomy to do so in conformity with the national conditions. The influence of political leadership on economic performance and institutional design has become an emerging issue for development economists (Jones and Olken 2005). Under strong leadership, not only are reforms more deeply and time-consistently implemented, but their shape is also more fitted to the specificities of the national context, thereby driving durably positive functional effects (Andrews 2013). Our comparative case studies have contrasted Ghana and Côte d’Ivoire in this respect, with the former having demonstrated leadership autonomy resulting in the gradual making of an effective, albeit hybrid, institutional system.

The third explanation is that the experience drawn from the last three decades of institutional reforms in developing countries demonstrates that institutional change faces sizeable information problems. There is a great deal of uncertainty about the economic outcomes to be expected from a given institution, mainly because the action of this specific institution is complementary with that of the rest of the institutional system. Second, the interaction of this new institution with the local political and sociocultural context introduces an additional source of uncertainty. The whole institutional system can become dysfunctional and ineffective when the disjunction between formal rules and the underlying structures of power and practice becomes too great. Even though formal rules are changed, informal institutions, like collective norms of political and economic activities or resource management and cultural dominant traits, continue to influence the day-to-day working of the system (Roland 2004). The effectiveness of formal rules reforms may be thwarted by the survival of such informal institutions, as shown by Andrews (2013). In the case of Argentina, discretionary patron–client relationships between central and province-based governments, and soft budget constraints, have systematically undermined the effectiveness of public finance reforms. Likewise, high degrees of personalization of politics and institutions, and the persistence of a highly partial political culture in Malawi, by limiting the emergence of a modern and formalized public sector have accentuated informality and corruption.

The process by which institutional reforms are carried out is, therefore, as crucial, in terms of economic effects, as the specific shape of the institution to be introduced (Evans 2004; Andrews 2013). Experimentation and learning enable institutions to be adjusted to the prevalent system as well as to the sociocultural context. Since they ease the emergence of political consensus and support for the new institutions, such mechanisms as political deliberation have equally been advanced as essential ingredients of a successful institution building process (Roland 2004; Evans 2004). The country-case studies in Rodrik (2003) have also pointed to the critical roles of both the fit of institutions to local conditions and the organization of political support by political leaders, in ensuring the successful economic trajectories of China, Botswana and Mauritius. Institutional reforms, therefore, imply adjustments to local conditions that can significantly transform the shape or enforcement mode of the institution. This can explain why so many developing countries’ systems exhibit idiosyncratic modes of sector governance.

Institutional change is increasingly described as a bricolage, with solutions emerging from available resources and extant rules, that are simply incrementally reorganized, and not by application of fully-fledged transplanted best-practice solutions (Campbell 2004; Andrews 2013). Institution building, therefore, corresponds to an evolutionary process of trial and error, based on incremental innovation and adjustment of the existing system, with such a system finally imposing high path-dependent inertia on the institutional configuration (Acemoglu and Robinson 2012), and with certain institutional dimensions moving faster than others (Roland 2004). Institutional reform in developing economies needs, therefore, to be informed by a clearer understanding of how the various institutional dimensions actually combine together in a more or less complementary fashion. This book represents a contribution to this goal.

Our framework and results have, in addition, two key implications for policy issues and institutional reforms. First, reasoning in terms of systems of institutions and of institutional complementarities helps our understanding of why some reforms (like importing a common law system, or deregulating labour or capital markets) do not necessarily raise system effectiveness, and can even, if the other institutions do not fit, cause serious inefficiency.

Second, there is no one single best way to escape from the low Informal (Weak State) model but, instead, a variety of institutional trajectories. Although all poor countries tend to have adopted institutional configurations that look similarly and consistently informal, the sheer variety of middle-income and emerging countries’ institutional systems suggests that there are different ways of escaping from the informal institutional trap and acceding to prosperity. What chiefly opposes our institutional clusters is that, whereas certain countries have adopted more or less statist or globalization-friendly systems showing internally consistent isomorphic sectoral governance mechanisms, others have experimented original, sometimes unintended, institutional configurations.

It is therefore suggested that, for countries trapped into the Informal (Weak State) institutional configuration, the path to institutional formalization goes through a transitional phase that may follow one of three different trajectories. One trajectory is based on the empowerment of the state as a central actor of economic and social life. As shown in Chaps. 12 and 13, that trajectory is generally historically or physically conditioned, that is, state antiquity and/or the availability of natural resources have contributed to the maintenance and strengthening of path-dependent, often authoritative, national varieties of the developmental state. Another trajectory may involve adapting the institutional system to globalized economy requisites. A third trajectory would consist in proceeding to national institutional innovation and experimentation. This could take one of two different forms: (1) hybridization, merging well-identified area-related institutional models in a highly original way; or (2) innovation, essentially mixing idiosyncratic area-related models.

Throughout the book, strong emphasis has been placed on emerging countries, because they exhibit differentiated, often innovative, ways of organizing their economic systems, with positive development outcomes. By clustering institutions of socioeconomic governance, our empirical analysis has been able to reveal the variety of developing countries’ institutional systems, thereby demonstrating that, for the less developed and formalized economies, different strategies of institutional reforms can escape informal dysfunctional systems. Thus, there is not one unique way of “being happy”, that is, building institutional systems that enable sustained economic development. According to Besley and Persson (2011), all successful countries look the same, namely, they cluster all the good institutions and outcomes by simultaneously building up state legal and fiscal capacities whilst avoiding internal violence, whereas the countries that fail to develop also fail, each in its own specific way, to simultaneously address those three keys of success. We complement this by showing that for those conditions to be met, there is a variety of institutional options, some of which being extremely innovative.

6 Paths for Future Research

The present book has raised both analytical and policy issues. The main analytical issue concerns the empirical approach most adapted to addressing the sui generis constructive complexity of institutional systems. We have argued in this book that, although this issue has been diversely addressed by various strands of literature, developing countries’ institutional systems have seldom been systematically scrutinized and compared. When institutional systems are analyzed as clusters of sectoral institutions, the standard opposition between state and market does not fully describe the real content of institutional reforms in developing and emerging capitalisms. One crucial explanation of institutional diversity within developing countries may well, in fact, be the degree of institutional experimentation. Experimental institutional systems, based on hybrid or unusual sets of institutional governance models, can certainly be opposed to the more standard institutional systems that are more conventionally accepted as internally consistent configurations.

The main policy issue, therefore, concerns the process of institution building and that of institutional reform in developing countries. Our results shed light on the puzzling issue of ineffective institutional reforms in developing countries. We argue that hybridization is a common practice for the institutional systems that are embarked on a trajectory of highly frequent incremental reform: by simply combining pre-existing elements with new ones, the job can be done, given the resources at hand. The U-shaped distribution of countries over the two institutional formalization and experimentation axes certainly needs further research to be understood more fully, but it clearly shows that the way towards institutional sophistication and economic development may go through a phase of institutional innovation, one in which each country experiments its own institutional arrangements. The intermediate position of emerging economies suggests that accounting for institutional diversity and experimentation is crucial in order to identify what does and does not enable a poor country to escape underdevelopment traps and accede to sustained economic progress.

What could be the next steps or additions to the present work? First, different strands of the new-institutionalism, like CC and NIE, should be increasingly integrated. Our work represents a first and modest attempt aimed at doing that. But there are undoubtedly other possible fruitful paths to be followed with that aim in mind. Second, our understanding of developing countries’ economic systems should be improved, notably by examining more closely the connection between the institutional structure, and its different components, with economic performance or vulnerability. Performance and vulnerability can be gauged at both micro- and macro-levels. Hybrid systems provide the different parts or sectors of the economy with sometimes contradictory microeconomic incentives. This may induce inefficiencies that are worth being identified and corrected by well-informed policy reforms. Equally, economic growth may be affected by the contradictory effects of the finance and labour market regulations on private investment.

Third, our final analysis has led to interesting and new results with respect to the main concern of assessing the institutional architecture, as well as the forms of internal consistency or inconsistency, to be observed among developing countries. Our focus has been more specifically put on these so-called emerging capitalist systems, which, although based on significant but highly original forms of state interventionism, seem to successfully deliver sustained growth and international competitiveness. This variety of original blends of statism and globalization-compatibility has not, so far, been sufficiently analysed in the existing literature. Yet, such a variety of institutional models challenges two very conventional and prevalent views: (1) the “monocropping” view, according to which the best-fitted institutional forms (LMEs and CMEs in our analysis) necessarily come from advanced mature capitalism, and (2) the view that globalization necessarily leads to convergence towards the liberal model of capitalism, with highly flexible markets and low collective social protection.

Technical change and globalization seem, in fact, to have increased economic complementarity between the two models of mature capitalisms (Acemoglu et al. 2012), with the high income levels of CME countries offering markets for the radical innovations of LMEs. This book contributes to broadening this framework by revealing a larger set of differences and potential economic complementarities between capitalist models, notably in what concerns emerging capitalist economies, whose static and dynamic properties require to more systematic study.