Review – How Institutions Evolve

How Institutions Evolve: The Political Economy of Skills in Germany, Britain, the United States, and Japan
By Kathleen Thelen
Cambridge University Press, 2004

How Institutions Evolve, Kathleen ThelenKathleen Thelen’s How Institutions Evolve: The Political Economy of Skills in Germany, Britain, the United States, and Japan is an elegantly argued book on the comparative political economy of skill formation. Using institutional theory within a varieties of capitalism framework to explain the causes for variation in vocational training and skill formation amongst Germany, the United Kingdom, Japan, and the United States, Thelen traces the interaction between and amongst states, firms, and labour movements from 1870 to 1920. In my view, the work is a very well argued piece; however, in looking to contribute to both varieties of capitalism and institutionalism, the author opens up questionable points in her comparison. The review is structured as follows: (1) an analysis of her case and theoretical choices; and (2) a section which focuses on her comparative method and its weaknesses.

Cases Selection

Thelen’s setting begins with a focus on large machine and metal working firms pursuing industrial expansion and following similar cost-effective strategies regarding skill formation in the 19th century Germany, United Kingdom, Japan, and the United States. What justifies her case selection and time period? First, these were consolidated outcomes no longer subject to change, but moreover, since Thelen is using historical institutionalism in the varieties of capitalism literature, it was necessary to look at the institutional genesis of similar institutions in their national political settings and trace the evolution overtime. With the conflation of politics and economics due to the pressures of increasing industrial development, social and class restructuring, and competition amongst great powers, nation states were the most appropriate unit of primary of analysis. Drawing from the categories in the varieties of capitalism literature on the differentiated ‘high-skill’ economies of Germany and Japan from the ‘low-skill’ economies of United States and Great Britain (Allan 2006, Hall and Soskice 2001, Streeck 2009, Steinmo 2010), she groups these four cases into these categories following the similarities of the state’s role in societal and economic development. German and Japan industrialization occurred under the auspices of an uncontestable, authoritarian state, which follows a coordinated system of bargaining. Conversely, the ‘liberal systems,’ Great Britain and United States, were grouped because of the historical strength of the market and firm.

Second, a comparison within the institutional setting of these states allows the systematic assessment of skill formation’s development without attributing explanatory power to deterministic, non-falsifiable factors. In analyzing the emergence and development of different kinds of capitalism, the types of skills that matured over time have implications to the competitiveness of the economy and society; liberal regimes have been associated with wider gaps of opportunities to different types of workers, thus implicating income inequality. While other factors can also influence capitalism – such as geopolitical climate, the existence of important resources, modernization, ‘social systems production’ or ‘cultural’ ideational explanations – focus on these ‘harder structural’ or bias for ‘softer agential’ explanations alone have fallen out of favor in varieties of capitalism (Hall and Soskice 2002, Streeck 2009). As such, a structure-agent, nondeterministic approach can be found in her institutional approach, which theorizes the interaction of and outcomes from, strategically located actors in institutional settings (Hay 2002, Streeck and Thelen 2009). This point solidly contributes to the varieties of capitalism literature which argue that the political economy is a mode of varying, integrated systems in which institutional arenas – capital infusion system, vocational education and training systems, transnational governance, and industrial relations – coalesce in crucial moments to form a type of capitalism.

Could Thelen have analyzed other units besides the ones already mentioned? Since the categories of liberal and coordinated systems go beyond her four cases, she could have tested more or possibly chosen better cases: Ireland, New Zealand, and Australia for liberal systems (Becker 2006, Croates 2008); France and Scandinavian countries for the coordinated systems (Becker 2006. Croates, 2009). However, her strategy lies not in picking cases with broadly similar conditions, but in cases where it is possible to locate and trace the institutional genesis of capitalism. It is important to start at the origin of an institution because crucial choices made set a path that would influence further policy choices. Pierson remarks that “cumulative commitments on the existing path will often make change difficult and will condition the form in which new branching will occur.” (Pierson 2004; 52) Russia was only able to industrialize after World War I; France, the Netherlands, and Belgium were still relying on largely protectionist, agriculturally based economies; while New Zealand and Australia largely developed after World War II (Allan, 2006, Becker 2006, Croates 2008). For Thelen’s case, as these other cases come later historically, they were much less relevant than those cases where institutional genesis could be found (Moore, 1973). Also following Barrington Moore and Theda Skocpol’s works, there is a bias that ‘larger’ units are more important than smaller ones (Moore, 1973). Since Germany, the United Kingdom, Japan, and the United States were trendsetters in industrial expansion and liberal trade policies, they were the larger, earlier, and more important units.

Furthermore, her selection of approach and design contributes to institutional theory by demonstrating institutional change and continuity. The idea of a ‘locked-in’ or ‘sticky’ institution does not account for gradual institutional change caused by the choices of actors in historical junctures (Hay 2002, Pierson 2004, Mahoney and Thelen 2009, Streeck and Thelen 2009). Past research focuses on ‘path-dependence’ or institutional reproduction by observing exogenous shocks or critical junctures to perpetuate institutional change (Mahoney and Thelen 2009, Streeck and Thelen 2009). Following this, Thelen disagrees with the idea that institutional change is a process that happens because of ‘exogenous’ shocks such as invasions and wars. She believes that change and continuity are interlinked processes and it is important to see which components are reproduced over time and which are not. So instead of a static, ‘locked-in’ phase or a discontinuity in the form of a ‘critical juncture,’ she argues that institutional reproduction and change should be studied together and are very much closely linked (Mahoney and Thelen 2009, Streeck and Thelen 2009). She shows in her analysis that positive feedback and agential choice could incrementally change institutions; the idea of a ‘locked-in’ path should be qualified if it is indeed so or is it a path with a ‘busted lock.’

Despite not mentioning it in her book, I feel that she is contributing to the debate of various forms of institutionalism in preferences formation and causation. As Thelen is located in the historical institutionalist camp (Mahoney and Thelen 2009), there are three other major camps, which are the rational, sociological, and the discursive. Broadly speaking, rational institutionalism believes that there is a universal claim to rationality; preferences are fixed on maximizing gains from institutional decisions; and that outcomes are products of actor preferences with institutional rules (North 1990). Sociological and Discursive institutionalism believes that there are no universal claims on rationality; preferences depend on the appropriateness of norms in the period; and outcomes depend on the transfer of those norms to reality (Schmidt 2008). By showing that preferences and rationality are historically made, as in the interaction of firms and unions, she affirms the stance of historical institutionalism that past policy paths matter in shaping outcomes and trajectory of the present.

Comparative Method, Path Dependency, and Causation

In endeavoring to identify general patterns, Thelen uses two methods of comparison. First, she uses the ‘indirect method’ of difference, which creates and compares two comparative categories with one another, and the logic of concomitant variation, which explains the divergence of the cases in one category from each other. She locates the critical juncture near the 19th century, when large machine and metal working firms were pursuing the same interest and following common strategies concerning skill formation. An important outcome was that of the association of traditionally skilled artisans– gatekeepers of skills which the firms needed. State policies differ on this aspect: in Germany and Japan, state policy sharpened and institutionalized the divide between artisans and skilled workers in the industry; while in the United Kingdom and the United States, state policy obscured the difference between these two, eventually merging them into skill-based unions. Whether the artisans separated or merged with organized labor, this had a crucial impact on the legitimacy and effectiveness of the skill training system.

In Germany and Japan, firms and organized (mostly unskilled) labor formed coalitions to battle against these privileges. German firms wanted the power to certify their own workers and the Japanese firms lobbied for in-house firm training. The variation in coordinated regimes lies in that certification and training were institutionalized in the national level in Germany and in the firm level in Japan. In both scenarios, the interest of the firms and organized labor converged; it set the path towards investment on skill and the legitimacy of the system.

In the liberal regimes, where artisans merged with organized labor, the interest of the firms and organized labor were at odds; it set the path towards the lack of investment and diminishing legitimacy of the system. The battle in Great Britain was to reassert managerial control in the shop floor and, in the US, firms had the goal of reducing their dependency on skilled labor through research, innovation, technological change, and product standardization. Overall, judging from historical works done in the subject, her empirical data and her evidence appears to be accurate  (Allan 2006, Becker 2006, Buigues and Sekkat 2008).

Thelen argues that ‘the behavior and strategy of leading skill-intensive industries, particularly the machine and working industry’ in the 19th century, was a ‘path dependent’ choice and the most crucial variable that led to divergent trajectories. She argues that Japan, similar to Great Britain and the United States, actively dismantled its traditional artisanal sector. The state policy should have led Japan to the path taken by the two aforementioned countries, but firm strategy to employ artisans as oyakata to train new skilled workers directed Japan back to a similar path taken by Germany. The Japanese state was facing pressure for modernizing and developing state policies, so it intervened by recruiting skilled workers from abroad; sending students to study industrial relations; and redressing the labor shortage. She further argues that state policies towards the artisanal sector could not explain variation in skill formation taken by these countries and were only able to account for the outcome of the aforementioned sector; thus, opening up the critical juncture. As Thelen is contributing to historical institutionalism, she best highlights multiple paths that firms could have taken during the critical juncture. In doing so, she implies that if the Japanese did not employ the artisans as oyakatas, the conflict could have followed the craft control battles of Great Britain or the United States.

There are two major weaknesses in her application of the comparative method. First, her analysis is very neat, but her generalization through the method is inapplicable to Germany. In Japan, Great Britain, and the United States, the state actively dismantled the traditional artisanal class and after doing so, the state no longer had a hand except in indirect policies regarding skill formation. In Germany, state policy reorganized and institutionalized the artisanal sector into a ‘parapublic’ institution, granting the sector monopoly to certify skills, thus making it a key sector in the political economy. Therefore, Germany did not start from the same ‘critical juncture’ as the other cases. Germany appears to be the ‘odd one’ out.  The big difference lies in the absence of other paths from which the German firms could choose from; state policy was the most crucial variable. They did not have any other choice but to deal with and contest skill certification from the public institution. In that case, is it even possible to draw a level of causal generalization if there appears to be a level of determinism in the German case? There are therefore questionable premises in applying the indirect method of difference, the concomitant variation, and ‘path dependence’ argument since Germany’s path appears to be already determined and its critical juncture could be found elsewhere. Thelen could have looked for the juncture somewhere else or perhaps argued that German firms could have acted differently.

Second, an unmentioned dimension in her work is the interrelation of these units in the 19th century which led to ‘emulation’ and ‘borrowing.’ A further exploration of this dimension could weaken her work. I will focus on two possibilities here. First, these nation states were engaged in various modes of emulation, and in constant, relentless economic competition. Auggelo and Guidi argue that there were international institutions designed to learn from its members; particularly, international economic societies had members which served as key advisers to the government and companies (Augello and Guidi, 2001). So rather than locating causal generalization in the strategy of firms during the critical juncture, could causality be a product of an emerging, common economic paradigm among these economic societies?  How much of an influence were these members to state decision making? These questions bring up questionable issues of causality and outcomes in her work.

Second, decisions of these firms were influenced by important international shareholders. Ferguson argues that the declining imperial dominance of the Great Britain at that time resulted in capital transfer in favor of Germany and the United States. With higher of capital, firms gained the capacity to invest in new technologies, as in the case of the US, or to bargain with an institutionalized artisan class, as in Germany (Allan 2006, Buigues and Sekkat 2008, Steeck 2009, Steinmo 2010). Therefore, if firm decision is not an independent variable, not a critical juncture which would set path dependence, but is also a product of interacting processes among these actors, is it still possible to infer causal generalization and path dependence? Thelen could have explored these interconnections and preempt possible objections to her argument.


Augello, Massimo and Guidi, Marco. 2001. The Spread of Political Economy of the Professionalisation of Economists. Economic societies in Europe, America, and Japan in the nineteenth century. London: Routledge.

Allen, Mathew. 2006. The Varieties of Capitalism Paradigm: Explaining Germany’s Comparative Advantage? (New Perspectives in German Studies). UK: Palgrave Macmillan.

Becker, Uwe. 2006. Open Varieties of Capitalism: Continuity, Change and Performances. UK: Palgrave: Macmillan.

Buigues, Pierre-Andre and Sekkat, Khalid. 2009. Industrial Policy in Europe, Japan, and USA Amounts, Mechanisms, and Effectiveness. UK: Palgrave Macmillan.

Croates, David. 2008 Varieties of Capitalism, Varieties of Approaches. UK: Palgrave Macmillan

Hall, Peter and Soskice, David. 2001. Varieties of Capitalism: The Institutional Foundations of Comparative Advantage. USA: Oxford University Press.

Hay, Colin. 2002. Political Analysis. London: Routledge.

Mahoney, James and Thelen, Kathleen. 2009. Explaining Institutional Change: Ambiguity, Agency, and Power. UK: Cambridge University Press

Moore, Barrington. 1993. Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World. USA: Beacon Press

North, Douglass. 1990. Institutions, Institutional Change and Economic Performance (Political Economy of Institutions and Decisions). UK, Cambridge University Press.

Pierson, Paul. 2004. Politics in time: History, Institutions, and Social Analysis. USA: Princeton University Press.

Schmidt, Vivien A. 2008.”Discursive institutionalism: The explanatory power of discourse.” Annual Review of Political Science 11: 303–326.

Steinmo, Sven. 2010. The Evolution of Modern States: Sweden, Japan, and the United States (Cambridge Studies in Comparative Politics). UK: Cambridge University Press

Streeck, Wolfgang. 2009. Re-Forming Capitalism: Institutional Change in the German Political Economy. USA: Oxford University Press

Streeck, Wolfgang and Thelen, Kathleen. 2009. Beyond Continuity: Institutional Change in Advanced Political Economies. USA: Oxford University Press

Alvin Almendrala Camba is a Commissioning Editor for e-IR.

Please Consider Donating

Before you download your free e-book, please consider donating to support open access publishing.

E-IR is an independent non-profit publisher run by an all volunteer team. Your donations allow us to invest in new open access titles and pay our bandwidth bills to ensure we keep our existing titles free to view. Any amount, in any currency, is appreciated. Many thanks!

Donations are voluntary and not required to download the e-book - your link to download is below.


Get our weekly email