1. Introduction
Ha-Joon Chang (Reference Chang2011) argues that economists place too much faith in ‘liberalized’ institutions. Institutions matter for growth, he contends, but not the way institutional economists think they do. Challenging the conclusions made over the past two decades regarding the role of institutions, Chang is skeptical that institutions, particularly liberal institutions, are of prime importance for economic growth. His attack on the literature that points to the primacy of liberal institutions is three-fold. First, he insists that economists should promote economic growth rather than focusing on institutions since, in his view, economic development is a determinant of institutional quality and causation in the reverse direction is unsubstantiated. Second, he argues that the econometric studies supporting the ‘institutions matter’ hypothesis are fundamentally flawed. Last, according to Chang, the dynamic nature of institutions is not well understood and is poorly studied. As such, he suggests, economists do not know enough about how institutions work to purport the significance of certain institutions.
In this article, we offer a defense of the hypothesis that ‘institutions matter’ for economic growth and present several objections to Chang's arguments. Although Chang makes certain salient claims regarding the approaches to studying this relationship that are sometimes pursued, accepting his critiques still leaves intact the notion that institutions matter fundamentally for economic growth. Essentially, we find that Chang's assault on the ‘institutions matter’ hypothesis is unfounded and grows out of his mistaking the chaff for the wheat in that literature regarding the positive relationship between the institutions of economic freedom and economic growth.
2. Defending the ‘institutions matter’ hypothesis
Adam Smith (Reference Smith and Cannan1776) has offered, if not the first, arguably the clearest articulation of the ‘institutions matter’ hypothesis. ‘Little else is requisite to carry a state to the highest degree of opulence from the lowest barbarism,’ Smith (ibid., xliii) argued convincingly, ‘but peace, easy taxes, and a tolerable administration of justice; all the rest being brought about by the natural course of things.’ When the scope of the market and, so, the division of labor expands, nations become wealthier. When governments fail to clearly define and protect property rights or fail to enforce contracts, when they overtax their citizens, when they engage in internal and external predation, nations do not develop. Since Smith, there has been a raft of theoretical and empirical studies that have confirmed his formula (see, for instance, Hall and Lawson, Reference Hall and Lawson2011). Rather than engaging Smith's formula directly, however, Chang instead engages the weakest form of the ‘institutions matter’ hypothesis. Consequently, his arguments against the ‘dominant discourse on institutions and growth’ are not as convincing as he might have hoped.
The chief shortcoming of Chang's paper is that it does not provide a discussion about which institutions matter most for economic growth. Rather than focusing on private property, contract enforcement and the rule of law exclusively, Chang points to a variety of institutions that even proponents of the ‘institutions matter’ hypothesis would not describe as being essential for growth and concludes that these non-essential institutions are in fact non-essential. The institutions that Chang discusses, for instance, include a financial regulatory system with a central bank and securities regulation, a corporate governance system that is shareholder-oriented, a ‘flexible’ labor market, and even NAFTA (North American Free Trade Agreement). According to Chang (Reference Chang2011), these institutions are particularly ‘Anglo-American’ (ibid.: 2). He then concludes that:
Today's rich countries acquired most of the institutions that today's dominant view considers to be prerequisites of economic development after, not before, their economic development – democracy, modern bureaucracy, IPRs [intellectual property rights], limited liability, bankruptcy law, banking, the central bank, securities regulation, and so on (ibid.: 4).
In concluding that bankruptcy law or intellectual property rights or any of the other ‘Anglo-American’ institutions he points to represent products but not prerequisites of economic growth, however, he is not challenging the ‘institutions matter’ hypothesis, as he supposes. Instead, he is challenging the form of that hypothesis that is mistakenly embraced by some of the development and foreign aid organizations. His quarrel is, thus, fundamentally against the practice of economic development, not the scholarship regarding economic development.
A related concern with Chang's focusing on non-essential institutions and his argument that ‘higher-quality institutions’ are more likely to follow than precede economic development is that he privileges foreign aid without institutional strings attached over foreign aid that demands institutional reform. Chang writes:
If the causality runs more strongly in the direction of development to institutions, rather than the other way around, the financial and human resources that developing countries are expending in order to acquire GSIs [Global Standard Institutions] may be better used for other policies that more directly stimulate economic development – be they educational expenditure, infrastructural investments, or industrial subsidies – especially when they also indirectly promote institutional development, which can then further promote economic development (ibid.: 5).
Putting aside the feasibility of these programs for now, Chang seems to neglect that such programs require institutions (at some level) to work effectively. Indeed, it is unclear if we should really expect educational expenditure, infrastructural investments, or industrial subsidies to lead to wealth outside of a context where private property is respected and the rule of law preserved. Moreover, if wealth is paramount for the adoption of ‘superior’ institutions, then it follows that massive aid transfers to developing countries will put them on the path to good institutions. Without effective institutions, we cannot successfully interact to coordinate our activities, as we must in order to achieve our goals; this is certainly true if our goal is to create and improve an educational system, or to develop a nation's infrastructure or to subsidize industries. Thus, assuming economic development has an impact on institutions does not answer the truly important puzzle – how to achieve economic development in the first place. Western countries have poured over two trillion dollars into many African, Latin American, and Asian countries, however, with very few success stories in terms of growth-enhancing institutions, liberal or otherwise (Easterly, Reference Easterly2001). The progress made by these countries does not match the funds they received, and economists are becoming hyper-aware of the failures of foreign aid (Burnside and Dollar, Reference Burnside and Dollar2000; Svensson, Reference Svensson2000; Knack, Reference Knack2001; Alesina and Weder, Reference Alesina and Weder2002; Leeson, Reference Leeson2008; Carden, Reference Carden2009; John and Storr, Reference John and Storr2009).
While Chang is of course correct that economic development affects the quality of institutions, he does not demonstrate that certain fundamental institutions (namely, property, contract and the rule of law) are not essential for economic growth. Additionally, his critiques of the empirical efforts to tease out the connection between institutions and growth cut against the very evidence that he marshals in favor of his position. Chang (Reference Chang2011), for instance, claims that the evidence in support of the primacy of institutions is ‘a lot shakier than what the supporters of the dominant view want us to believe’ (ibid.: 11). Cross-sectional econometric studies, he suggests, are insufficient and, thus, time-series analysis will better determine how changes in institutions cause changes in economic outcomes. Furthermore, he claims that samples are treated homogeneously, even though the relationship between the dependent (economic growth) and independent variable (institution) in one country may not hold in another. Chang, however, uses poor empirics to make his case. He, for instance, points to Singapore as a prime example of how a strong state and supposedly illiberal institutions can result in an economic success story, while ignoring the pro-market institutions that allow Singapore to succeed, and never conceding that Singapore is but one observation and by no means a typical case. Similarly, his examples of African and Latin American countries (who he claims ‘faithfully reformed their institutions in the neo-liberal direction’; ibid.: 11), at best, demonstrate that official rules imposed from outside a society that do not match unofficial rules are not likely to have much success. These cases might indict the practice of top-down reform but do not overturn the notion that certain liberal institutions are necessary for economic growth.
Despite his efforts, Chang's assertion that good institutions follow growth rather than preceding growth is not convincing. And the assertion, to be plausible, requires that we conflate the academic literature on institutions and growth with the unfortunate structural adjustment programs that are often advocated by development agencies. These programs, however, have been convincingly criticized by supporters of the ‘institutions matter’ hypothesis and have been shown to matter less for appreciating growth than the institutional nexus in the country prior to ‘structural adjustment’ (see, for instance, Dollar and Svensson, Reference Dollar and Svensson2000). Chang's critique of Global Standard Institutions is, thus, most convincing when he laments (Chang, Reference Chang2011: 3) that the conditionalities placed on aid-recipient countries by the Washington Consensus organizations ‘are institutions that inherently favour the rich over the poor, capital over labour, and finance capital over industrial capital’ (ibid. 3). As he concludes, however, ‘the same institution in the same dose may be good for one country but bad for another’ (ibid.: 9) and, moreover, since institutions do not change easily, imposing institutions in certain countries may come at a very high cost (ibid.: 16–18).
3. Conclusion
Although Chang's effort to overturn the ‘institutions matter’ hypothesis is largely unsuccessful, his conclusion that the study of institutions and growth would be better served by analytical historical narratives and comparative historical studies because ‘“time-series” evidence . . . cannot capture complexities that characterize the domain of institutions’ (ibid.: 11) is quite compelling. Since culture legitimizes and delegitimizes institutions and, thus, lowers or raises the cost of enforcement of rules (Boettke, Reference Boettke and Boettke2001), we agree with Chang that an adequate understanding of a country's institutions requires history and ethnography (Boettke, Reference Boettke2000; Storr, Reference Storr2004).
Moreover, Chang's reminder that institutional change is costly so that it is possible that the cost of implementing liberal institutions may outweigh the benefits is intriguing and worthy of consideration. As he suggests:
Whatever the theoretical merits of the GSI approach to institutional reform . . . [they] would be a pie in the sky, if remolding non-GSIs into GSIs, or importing GSIs into countries with missing institutions . . . is very difficult . . . [or] if adopting different policies is even less costly (Chang, Reference Chang2011: 15).
While Chang employs examples such as patent law or the privatization of government industries, instead of property, contract and the rule of law, he is nonetheless correct that the cost of changing institutions (at least top-down approaches) might very well be higher than the benefits.
Additionally, Chang is correct that ‘institutions often have in-built mechanisms against change’ (ibid.: 18) but ‘economic development brings about cultural/institutional changes’ (ibid.: 20). He is quite right to point out that formal institutions are not so rigid that they can never be tinkered with or so soft that they can be easily transformed at will. As he writes, individuals are (in part) a product of existing institutions, institutions themselves often have rules that make change difficult, and some beneficial institutions can only function if complementary institutions are in place (ibid.: 17–18). Similarly, in the opinion of Chang, cultures should not be viewed as homogeneous and so easily characterized as being pro- or anti- development. He writes that ‘. . . a country's institutional complex contains various elements, and therefore can usually be described as pro-developmental, anti-developmental, or whatever we want, depending on which particular elements we choose to highlight’ (ibid.: 19; see also Storr, Reference Storr2006).
The dominant discourse on the relationship between institutions and growth and certainly the practices of the development agencies are rightly critiqued along some of the grounds that Chang suggests. But the conclusion that institutions (in particular property, contract and the rule of law) are unnecessary for growth has not been convincingly argued. In fact, when institutional analysis of the sort that Chang recommends – i.e. institutional economics which pays ‘more attention to the real world, both of the present and historical’ (Chang, Reference Chang2011: 22) – looks at the relationship between the institutions of economic freedom and economic growth, it tends to support Smith's hypothesis: ‘little else . . . but peace, easy taxes, and a tolerable administration of justice’ (1776: xliii).
