1. Introduction
Daron Acemoglu and James Robinson (Reference Acemoglu and Robinson2012) identify political participation as a cause of the rise of developmental institutions. This causality may explain cases from historical Europe, but not those from East Asia, where the quality of institutions improved under political authoritarianism. The explanation of East Asian institutional progress requires a different model. Atul Kohli (Reference Kohli2004) attributes the development in South Korea to a legacy of the colonial period, when the traditional institutions were coercively dismantled and replaced with Japanese institutions. This account, however, generates another question: how did developmental institutions evolve in Japan?
The institutional improvement in late 19th-century Japan (henceforth Meiji Japan) remains a void in the literature on institutions and development. This article presents a model that explains Meiji Japan's rapid modernization that it calls ‘elite redeployment’. Instead of ‘elite bargains’ in participatory polities in historical Europe, the traditional elites in Japan were coercively deprivileged by the revolutionary authoritarian regime, which redeployed those with financial or human capital among them in productive institutions. By doing so, Meiji Japan's authoritarian change agents dismantled the incumbent institutions in an irreversible manner and swiftly built new institutions such as modern administrative, educational, financial, and commercial sectors.
This redeployment policy resolved the major dilemma facing authoritarian change agents, namely, a trade-off between reform and stability. Authoritarian regimes have coercive power to remove underproductive institutions, but at the risk of inviting political instability, as politically influential deprivileged elites may engage in counteraction in order to recover what they perceive as their entitlement. Thus, many authoritarian regimes coopt elites by allowing them access to rent, either conventional or alternative, but this inevitably compromises institutional improvement since fiscal and administrative spaces are, then, not sufficiently liberated for productive players who generate wealth and increase the size of the economic pie for society.
Under Japan's elite redeployment model, the old elites were selectively reincorporated so that they would not hinder growth. This was with the calculation that reincorporating those with financial prowess or administrative skills would suffice for maintaining stability as those were politically influential agents who were capable of leading reactionary campaigns. In other words, their absence significantly weakened opposition. These elites, under normal circumstances, may obstruct reform that will lead to the loss of their conventional rent (Acemoglu et al., Reference Acemoglu, Johnson, Robinson, Aghion and Durlauf2005), but in the case of Meiji Japan, where the revolution politically deprived them of strong bargaining positions, the uncertainty induced them to accept the change as long as they found themselves in a better-treated group among the deprivileged previous elites. The model was also predicated upon the productive nature of the change agents: the Meiji revolutionaries, with a lower-class samurai background, were from local states that had succeeded in internal reform. On the eve of the Meiji Restoration, the leeway of these peripheral elites was already on the rise but was still limited: the revolution and the national unification brought them about political opportunities to build productive institutions at the national level.
2. The Japan puzzle
In America, whether in the government or in society, those who lead are wise in a manner commensurate with their position. On this point, I see that country utterly opposite of ours.
– Kaishū Katsu
Japan's greatest curse for ages has been an excess of officials and lazy rice-eaters who do not work.
– William Elliot Griffis
When does the quality of institutions improve and spur economic development?Footnote 1 Many consider developmental institutions as a historical product (Evans, Reference Evans1989) – a legacy of political changes in the past (Acemoglu et al., Reference Acemoglu, Johnson, Robinson, Aghion and Durlauf2005), or ‘critical junctures and path-dependency’ (Capoccia and Kelemen, Reference Capoccia and Kelemen2007). Such past political changes were the moments when agency, or creative actors with a strategic mind (Herrigel and Zeitlin, Reference Herrigel and Zeitlin2010) and their coalitions (Leftwich, Reference Leftwich2010), acquired high degrees of leeway from (or within) the structure, resulting in the shift in the balance of power between different institutions in society and new political settlements between them (Khan, Reference Khan2017). In particular, what is integral to early development is political settlements that facilitate the transformation of patrimonial institutions replete with rent-seeking actions by traditional elites into productive institutions that enlarge the overall economic pie. Productive institutions that enhance, and are also enhanced by, human capital (Acemoglu et al., Reference Acemoglu, Gallego and Robinson2014; Faria et al., Reference H., Montesinos-Yufa, Morales and Navarro2016; Mokyr, Reference Mokyr2009, Reference Mokyr2017), are known to have attributes such as the rule of law, meritocracy (Evans and Rauch, Reference Evans and Rauch1999), civic-mindedness (Bowles and Gintis, Reference Bowles and Gintis2011), and being impersonal or disinterested (Bourdieu, Reference Bourdieu2004; Greif, Reference Greif2006). Rent-seeking occurs in any economy (Kang, Reference Kang2002), but productive institutions that crosscut the state and society (via bureaucracy and regulation) allocate greater public resources to economic sectors that generate wealth for society than patrimonial institutions that reward underproductive rent recipients.
Acemoglu and Robinson (Reference Acemoglu and Robinson2012) trace such institutional transformation to the beginning of greater political participation. However, as a matter of historical fact, such a democratization thesis is Eurocentric and cannot explain early economic development in East Asia well. On the other hand, Atul Kohli (Reference Kohli1994) advances a colonial legacy thesis, attributing the development in South Korea to Japanese colonial governance, which thoroughly dismantled the traditional monarchical institutions and replaced them with transplanted Japanese institutions, even at a high cost of doing so, in contrast to European empires, which tended to preserve local vested interests and patronize them. This colonial legacy thesis, nevertheless, generates a new question: how did Japan go through its own institutional transformation? Without its own preceding experience, the Japanese colonial regime in Korea would not have shaped such a preference for establishing developmental institutions or had the capacity to do so.
Following the revolution (the Meiji Restoration) in 1867/8, rapid institutional change occurred in Japan in the 1870s and the 1880s. Within this short period of time, the samurai, the previous dominant class, were deprivileged; modern bureaucracy and education was installed; a new financial system was formed; entrepreneurship for modern industries (such as railway and textile) was encouraged; and rules for recruiting civil servants were established. This series of reform, culminating in the proclamation of the constitution in 1889, enabled Japan to start building a modern industrialized economy and expanding as a regional empire in the 1890s. Before the final quarter of the century, however, the country was still lagging behind in many institutional aspects. As shown by the above-quoted remark by Kaishū Katsu, an official of the ancien régime, the Shogunate, which was made after his return from his visit to the United States in 1860, the country's bureaucratic and social order was only partially meritocratic and remained largely hierarchical then (Yoshimoto, Reference Yoshimoto1898); likewise, as William Elliot Griffis, an American educator who stayed in Japan after the Meiji Restoration, wrote in his diary in 1871, the country's samurai-dominated public sector was bloated and far from being efficient (Griffis, Reference Griffis1876). Such views are quite opposite the image of efficient bureaucracy presented by Chalmers Johnson in his seminal work on Japan's ‘developmental state’ in the 20th century, MITI and the Japanese Miracle (Reference Johnson1982), which indicates the speediness and thoroughness of the early change.
Japan achieved such institutional upgrading without a parliament, which (the Diet) was launched only in 1890. Although Acemoglu and Robinson (Reference Acemoglu and Robinson2012) include Japan in their narratives, their analysis is factually problematic: they focus on the liberalizing leadership of Toshimichi Ōkubo, but historians usually view the five years of Ōkubo's leadership (between the exodus of other early revolutionary leaders from the government in 1873 and his assassination in 1878) as yūshi sensei, or greater despotism than the period before, and after, it. Nor did Japan fit the well-known, but Eurocentric, bellicist thesis that identifies war as a catalyst for state formation (Tilly, Reference Tilly1990). Japan did not fight interstate war until 1894 (the First Sino–Japanese War). Some presume that even without war, the perceived menace of Western imperialism urged Japan's patriotic modernization. However, this assumption cannot explain why Japan was different from other non-Western countries that failed to reform themselves despite similar external threats and continued to be plagued by the pathology of patrimonialism and arrested productivity.
Indeed, Meiji Japan remains a void in the literature on institutions and development. To resolve this Japan puzzle, we need to incorporate the literature written by historians well. There are inspiring works on the Meiji Restoration by early-generation historians such as Thomas Smith, William Beasley, Henry Rosovsky, Harry Harootunian, and Bernard Silberman, which were cited by early political economy comparativists such as Reinhard Bendix (Reference Bendix1964), Barrington Moore (Reference Moore1966), and Ellen Trimberger (Reference Trimberger1972, Reference Trimberger1978). This article, however, will also refer to more updated recent works by Japanese historians, most of which are available only in the Japanese language.
3. Theory and model
A distinctive characteristic of the institutional improvement in Meiji Japan is that it took place under the revolutionary authoritarian regime and, therefore, did not occur through elite bargains, that is, horizontal negotiations between reformist and conservative elites that characterize the institutional progress in historical Europe. In major Western European countries, early economic development occurred under existing dynastic regimes, either counterrevolutionary–restorative (England and France) or incumbent–nationally unifying (Prussia/Germany and Piedmont/Italy), in which the privileges of traditional elites, in particular of landed nobles, were at the same time preserved and challenged (Bush, Reference Bush1983; Mayer, Reference Mayer1981). With such a political setting, a key research question has been about identifying the dynamic of elite bargains, in particular when the incumbent elites accept institutional changes that cause the loss of their rent (Acemoglu et al., Reference Acemoglu, Johnson, Robinson, Aghion and Durlauf2005). The changes occurred gradually, such as through political coalitions formed by change agents (Mahoney and Thelen, Reference Mahoney and Thelen2009), while many other countries under existing dynastic regimes, such as Qing China, failed to overcome the institutional sclerosis created by strong status quo-oriented coalitions that strived to preserve their vested interests (Olson, Reference Olson1982).
Meiji Japan was different from such elite bargains cases as the revolutionary regime born in the winter of 1867/8 persisted, with the revolutionary oligarchs collectively ruling for half a century. These revolutionaries, the change agents with the bottom elite background – lower-class samurai – coercively deprivileged the previous elites, the samurai and their local lord (daimyo). Unlike the conservative agents in Western Europe, where elite resilience or return was the norm, the revolution deprived the old elites in Japan of strong bargaining positions. The authoritarian change agents were, nevertheless, careful about steps of dismantling the incumbent institutions, taking into consideration their political repercussions. Although they had coercive power to remove underproductive institutions, doing so would risk inviting political instability in the long run, which inevitably takes public resources away from modernization programs and may even challenge the regime's legitimacy. This is a major dilemma facing authoritarian change agents – a trade-off between reform and stability – as politically influential deprivileged elites may engage in counteraction in order to protect and recover what they perceive as their entitlement.
Thus, many authoritarian regimes coopt elites by allowing them access to rent, either conventional or alternative. Such buying-off, however, occupies major fiscal and administrative spaces and does not liberate sufficient space for productive players who generate wealth and increase the overall size of the economic pie for society. The cooptation aimed at political stability, thus, inevitably compromises long-term institutional efficiency and improvement. This indicates that developmental political settlements can be precarious and exposed to the risk of reversal. Breaking the reform–stability dilemma to achieve permanent institutional upgrading, therefore, requires the conservative agents’ acceptance of the new order, so that the authoritarian change agents can use public resources for economic growth rather than for costly cooptation or repression with reinforced security apparatus.
This article argues that Meiji Japan achieved this acceptance through ‘elite redeployment.’ The Meiji revolutionary regime reincorporated and redeployed the deprivileged elites in newly established productive institutions so as to make reform and stability compatible with each other. How does elite redeployment work? How does it differ from normal cooptation? First, while elite cooptation prioritizes political logic over economic logic, elite redeployment intends to achieve both political and economic objectives as much as possible. Thus, while elite cooptation stagnates the process of institutional improvement by maintaining the underproductive groups of rent recipients, elite redeployment fosters improvement by removing these stumbling blocks in the way of change. Elite deployment harnesses the productivity of the old elites by utilizing their financial and human capital in newly established institutions such as administrative, educational, financial, and commercial sectors so that they will not hinder but assist development; in contrast, the productivity of the coopted elites usually remains at low levels. In Meiji Japan, the conversion of the old elites into productive players even compensated for the society's major shortcoming for modern capitalism: the absence of a competitive bourgeois class equivalent to that in Europe.
Second, elite redeployment is, thus, more selective than elite cooptation. In principle, only those who have good financial or human capital among the old elites were reincorporated. This was with the calculation that doing so would suffice for maintaining political stability as these were politically influential agents who could lead reactionary campaigns. In other words, their absence significantly weakened opposition. At the same time, these reincorporated elites were redeployed with their old patronage ties being severed, which made it hard for them to engage in collective counteraction. The Meiji regime also shrewdly excluded the army from the destinations for elite redeployment. This selective process inevitably led to large numbers of previous elites losing income too, but without an alliance with the most capable group of the previous elites, these impoverished old elites basically lacked the capacity to effectively rebel. In this way, the deprivileged old elites were divided between new winners and losers. Because of the post-revolution uncertainty, those who might obstruct reform, leading to the loss of their conventional rent under normal circumstances, accepted redeployment as long as they found themselves in the better-treated group among the deprivileged previous elites. Moreover, the regime was able to maintain policy options of lower-cost cooptation toward the loser group so as to buy time until greater opportunities were created for them through economic growth.
The rest of this article will empirically demonstrate the above discussion. The first section will highlight how the Meiji Restoration was brought about. This systematic understanding of the revolution is indispensable as the elite deployment in Japan was predicated upon the Meiji change agents already being part of productive institutions by the eve of the revolution. Thus, only by highlighting the dynamic of local-level institutional change in the pre-Meiji period can the nature of the Meiji change agents and their relationship to the existing institutions be rightly understood. The second section will analyze the modus operandi of elite redeployment, focusing on how the Meiji regime redeployed the deprivileged elites with financial or human capital in new institutions such as modern administrative, educational, financial, and commercial sectors so that they would not block institutional progress.
4. The growth of productive institutions and the political rise of change agents
The change agents in Meiji Japan rose to take their places on the political center stage through the Meiji Restoration, the transition from the country's fragmented political system to the unified state headed by the Emperor Meiji (r. 1867–1912). In the new order, these revolutionary change agents enjoyed a high level of political autonomy from the old elites, not only because of their armament with imported Western weaponry, but also because the Imperial House's political connections with these previous elites had been almost absent in the fragmented national polity.
The political power of the Japanese Imperial House significantly declined after the late 12th century, with its political function being reduced to endorsing the legitimate governing mandate of the hegemonic samurai (warrior) lord, conferring upon him the title of Shōgun (commander in chief; henceforth ‘Shogun’). There were around 250–300 local states ruled by the samurai lord known as the daimyō (henceforth ‘daimyo’), and after 1603, the Tokugawa family assumed the title of Shogun and maintained its hegemony for more than two centuries, under which the Emperor's political power remained almost nil (Fujita, Reference Fujita2018). However, in the mid-19th century, this old order began to unravel through a series of new developments. First, in some local states, productive institutions grew and facilitated vibrant agricultural, commercial, and some proto-industrial activities. This partly owed to the fragmented polity itself, where the Tokugawa family pacified but did not unify the nation, leaving some leeway for local states to engage in internal reform. Reform was initially economic but then spilled over to the military field, particularly after China's defeat by Britain in the Opium War (1839–42). Second, following the beginning of US gunboat diplomacy toward Japan in 1853, the political authority of the Shogunate declined fast as a result of its negotiations with Western powers, which were perceived by many samurais as defeatist. Third, as a consequence of the foregoing two developments, change agents from some rising local states, particularly Satsuma and Chōshū, challenged the Shogunate by embracing the authority of the long-marginalized Imperial House. This led to the Meiji Restoration, in which the Tokugawa family was ostracized from the new centralized regime and the local states were eventually disbanded.
The productive institutions established by the Meiji regime were, thus, founded upon those of the local states where the change agents came from. These rising local states, known as yūhan, or ‘prodigious domains’, successfully invested in agricultural, commercial, and proto-industrial production and reaped revenues from exporting goods, first to other local states and then to the world, after the end of Japan's isolationism in the late 1850s (Banno and Ōno, Reference Banno and Ōno2010). Satsuma and Chōshū amassed extra wealth by harnessing their geographical advantages: Satsuma, the southernmost local state, exported tropical agricultural products, especially sugar, and also profited from illicit trade with China (when the Tokugawa Shogunate formally monopolized the country's official foreign trade); Chōshū in western Japan developed as an inter-local state trading and financial hub, banking on its strategic location within sea-lanes of communication in the age of new local state mercantilism (Yamagata, Reference Yamagata2010).
4.1 The political economy of the pre-modern economic growth in some local states
How did productive institutions grow in these local states, from which the revolutionary change agents emerged? It was an end result of the long-term political economy shift in response to the challenges facing these local states, but the emergence of such challenges was already embedded in the patterns of their initial state formation. These local states were formed largely in the late 16th century, when Japan was in a long civil war period. At that time, the samurai, who had traditionally been landed warriors, were incorporated into the daimyos’ armies. Meanwhile, due to the constantly high demand for warriors for war, becoming a samurai created opportunities of upward mobility for farmers. This resulted in these local states increasingly having an unbalanced fiscal structure, with growing numbers of salaried warriors and decreasing numbers of agricultural producers. To prevent further samuraization of the latter, daimyos thus demarcated a class line, coercively moving rural samurais to urban centers and directly ruling and taxing agricultural villages (Ikegami, Reference Ikegami2012: 16–29).Footnote 2 When the civil war period ended in the early 17th century, the now-unlanded samurai were converted into civil servants. As the borders between local states were fixed and the state–society relationship in these states was reshaped, daimyos transformed from what Olson (Reference Olson1993) calls ‘roving bandits’ to ‘stationary bandits’, i.e. long-staying authoritarian rulers facing an incentive to invest in economic growth rather than confiscate assets. Farmers were deprived of their weaponry and placed under strict state surveillance to ensure production and tax revenues for the state.
Because of the conversion of the army into the civilian bureaucracy, the public sector in these local states was already bloated from the outset, and the hierarchy of the old army remained, disallowing even those with civilian merit upward mobility in status (Ochiai, Reference Ochiai2015: 19–23). Such frozen hierarchy, in the long run, generated a paradoxical phenomenon in that samurais of the lower class became de facto, but not mandated, policy makers. These bottom elites acquired and accumulated governance skills and expertise as administrators and began formulating policies for a wide range of affairs. They were, however, unable to attain a legal status or de jure mandate commensurate with the scope of their expanding policy fields. Their input was, in principle, communicated to the decision-making circle of the daimyo and his ministers through their superiors from among the upper class (Mitani, Reference Mitani2017: 56–58).
Nevertheless, in some local states, the policy-making capacity of these bottom elites rose rapidly in the early 19th century. By then, these local states had suffered from grave fiscal unsustainability and were in need of structural economic reform. Following the separation of the samurai and farmer classes, their fiscal structure was based on extraction from rice-producing farmers and distribution to samurai civil servants. However, as market economies matured and household expenditures increased, this extraction–distribution equilibrium became increasingly unsustainable. Daimyos faced the difficult dilemma of securing state revenues and maintaining samurai loyalty, as increasing the level of extraction from farmers often resulted in the fleeing of the farmers to neighboring local states, inviting further fiscal difficulty (Ochiai, Reference Ochiai2015: 26). While many states withstood the crisis only with intermittent piecemeal austerity (thus accumulating debt), some found a solution in altering the rice-based fiscal structure by diversifying the local economy to host production of other crops and manufactured goods. Such structural economic reform required a closer collaboration of the daimyo and capable civil servants, creating greater decision-making power for skilled samurais of a lower-class background (Yamagata, Reference Yamagata2010: 68). Successful reform tended to occur in states known as tozama (outer) that had fought against the Tokugawa family in the early 17th century. The fiscal problem was particularly severe among them as major tozama states with large armies had been destined to suffer fiscally as the size of their territory was reduced or capped in the postwar settlements with the Tokugawa family, while their internal distributive commitments to their civilianized loyal samurais remained high (Ravina, Reference Ravina1999). After the Opium War put daimyos on high alert to Western imperialism, some of them also began to invest economic gains in military reform, which provided capable samurais, especially those who studied imported Western knowledge (yōgaku; also called rangaku, or ‘Dutch studies’) with strong upward mobility (Sugimoto, Reference Sugimoto1989).
Nevertheless, as a whole, the balance of power between change agents and conservative elites did not shift radically in favor of the former at this point. Reform initiatives were often carried out through productive administrative enclaves such as specialized agencies for nurturing new products and sectors (buiku-gata), with projects often being financed with funds owned by the daimyo or managed under his direct supervision (hōzōgin) instead of with regular state budgets (Itō, Reference Itō2014: 12; Yamagata, Reference Yamagata2010: 68). Moreover, political settlements were often unstable. Sharing the sense of crisis and perceiving the need for reform, conservative elites accepted some changes promoted by the reform coalition that bypassed their mandate. However, when they saw that their interests or mandates had been challenged too far, they often resorted to collective counteraction to resist or reverse the reform. Mito was a prominent case of such reform reversal: although it initially pioneered local state reform and served as a model for other local states, the menaced upper-class samurais were solidified and succeeded in abdicating the daimyo (Mitoshi-shi Hensan Iinkai, 1976, 1982).
4.2 The revolution by peripheral elites
After the 1850s, these local states that succeeded in internal reform substantially increased their political presence when the political authority of the Shogunate rapidly declined as a result of its negotiations with Western powers. The Shogunate abandoned its long-standing isolationist policy and accepted unequal treaties such as those applying extraterritorial rights to Western citizens in Japan and disallowing Japan the autonomy to set tariffs; this was perceived by many daimyos and samurais as defeatism against aliens. In particular, young samurais from Chōshū and Satsuma, both ruled by the tozama daimyo, played a key role in new political activism. Radicals from Chōshū waged anti-Shogunate campaigns within the Imperial Court by allying with some aristocrats, while Satsuma initially aimed to reform the Shogunate and reinforce their joint alliance with the Imperial Court: Satsuma, thus, helped the Shogunate in ostracizing the Chōshū activists from the Imperial Court and sent joint expeditionary forces against Chōshū. However, disillusioned by the Shogunate's intransigence and the low prospect for change it desired, Satsuma eventually entered a clandestine alliance with Chōshū (Mitani, Reference Mitani2017: 229–235).
By the eve of the revolution, the de facto decision-making power in these two local states was captured by anti-Shogunate activists who later formed the core of the Meiji regime, either through a coup (in Chōshū) or internal coalition politics (in Satsuma). In both cases, these activists benefited from the greater political space opened by the previous internal reform, and each eventually mobilized the army of their respective local state against the Shogunate (Tanaka, Reference Tanaka1963: 13–14). Thus, the nature of the revolution was somewhat unique. The Meiji Restoration is often described as an ‘aristocratic revolution’ (Smith, Reference Smith1960), an ‘elite revolution’ (Trimberger, Reference Trimberger1972), or a ‘revolution from above’ (Trimberger, Reference Trimberger1978) because it is different from ‘social revolution’, which accompanies the participation of the bourgeoisie and/or the masses in the overthrow of the ancien régime (Skocpol, Reference Skocpol1979), while it is still revolutionary in light of the resultant fall of the previous elite class. To be terminologically more precise, it was a revolution by peripheral elites – some lower-class samurais from some local states – whose political power had been on the rise but was still facing major constraints within the existing hierarchical order.
By forming the core of the new state headed by the Emperor, who, despite the demise of the Shogunate, continued to remain a nominal political power, accepting the decisions made by the revolutionary leadership, these samurai activists transcended their respective lords in power. If some analogy is to be made to the European context, then the Meiji Restoration was akin to the low-level servants of some powerful nobles becoming directly connected to the centralized monarchical regime and depriving those nobles of their political power. Such pattern of political shift did not occur in Europe, where the monarchies were traditional power holders and their survival and stability rested on the political and fiscal support of landed nobles (and these nobles did not host large bureaucracies comprised of middle- and bottom-level elites like the samurai); it did occur in Japan because Japan's Imperial House had long been politically marginalized in the fragmented polity and lacked substantial political connections with the incumbent elites.
Japan was, however, once heading closer to the European model of relationship between the monarch and the nobles. In response to the formation of the anti-Shogunate Satsuma-Chōshū alliance, the Shogun Yoshinobu (r. 1867) preemptively returned his governing mandate to the Imperial House so as to nullify the alliance's casus belli and rebuild his political leadership by becoming the most influential internal player within the Imperial Court. Daimyos of other major local states supported this transition, too (Mitani, Reference Mitani2017: 257–259). Yoshinobu and these daimyos expected a transition to an arrangement similar to the political system in reactionary Europe, where nobles collectively made decisions through the elitist assembly under the dynastic regime. This transition was halted by a coup in January 1868 led by activists from Satsuma and Chōshū who aimed at the ostracism of the hegemonic power from the new order. The new Imperial Army, consisting of the armies of Satsuma, Chōshū, and Tosa, defeated the previous Shogunate in the subsequent civil war and achieved their goal. The revolutionaries formed the core of the regime and implemented centralization by ordering the local states to repatriate land and people to the Imperial House. Through this process, they eliminated the constraints that had long been imposed on them – the hierarchy consisting of the Shogunate, the daimyo, and the upper-class samurai – which would have lasted if they had not taken revolutionary action (Tanaka, Reference Tanaka1963: 23).
5. The modus operandi of elite redeployment
After its creation in the winter of 1867/68, the Meiji regime dismantled the country's existing hierarchical distributive institutions almost within a decade. It first abolished the status of the daimyo and disbanded their local states in 1871, and then canceled the privileges of the samurai, most importantly hereditary stipends, but also some symbolic prerogatives such as the right to carry swords, in 1876. While the revolutionary change agents did so to liberate the fiscal and administrative spaces for building new productive institutions emulating those in the West (which they studied through, and after, the Iwakura Mission to the US and Europe between 1871 and 1873), they were also aware of the potential repercussions of such a drastic transition. They thus shored up the Imperial Army for potential rebellions before announcing the decision, but also prepared strategies to reincorporate the deprivileged previous elites. They prioritized the reincorporation of the politically influential among them, such as the daimyo, the upper-class samurai, and samurais with expertise who were capable of taking up a reactionary leadership. Without these agents, rebellions and opposition were weak and divided. At the same time, their redeployment was designed to turn these elites into productive players in new institutions, such as modern administrative, education, financial, and commercial sectors, so that their reincorporation would not hinder the process of modernization and even assist it in the absence of a competitive bourgeois class equivalent to that in historical Europe.
5.1 Daimyos dismantled
In 1871, as the first step of elite redeployment, the Meiji regime abolished the local states and replaced the daimyos with centrally appointed governors. The regime reassured the previous local lords by preserving their private assets and providing pensions. Furthermore, it offered them an alternative dignity by incorporating them into a newly created central peerage, conferring upon them a new source of honor as the kazoku, or noblemen who supported the Emperor. Still, the change agents, such as Toshimichi Ōkubo, expected the daimyos to resist the termination of the inherited lordship. The regime shored up a centralized army prior to the announcement of the decision by transferring the command of the armies of the three local states – Satsuma, Chōshū, and Tosa – to the Imperial Court in preparation for possible rebellion (Sonoda, Reference Sonoda1993: 121).
The decision was, however, peacefully accepted by all daimyos. According to Hiroki Ochiai, a Japanese historian who conducted extensive research on the end of the samurai class, this rather unexpected collective compliance was attributable to the daimyos’ perception that it was an opportunity to be released from the distressing fiscal management of their respective states. As discussed in the previous section, many local states by then had come to struggle with the dire dilemma of extraction and distribution. This fiscal dilemma was further aggravated by the civil war that followed the ostracism of the Shogun. Prior to 1871, more than three-quarters of the local states had suffered from debt larger than their annual revenues, and for around one-fifth, the debt was more than three times larger than the annual revenues (Ochiai, Reference Ochiai2015: 73). For many daimyos, the lordship was already painful – and no good prospects existed for this conundrum. Thus, they saw accepting the decision as a reasonable option as long as it released them from this cul-de-sac and provided them with honorable roles in the new order. Only Hisamitsu Shimazu, the ruler of Satsuma, expressed disgruntlement publicly, seeing the decision as a betrayal by his servants. The regime offered him a position as cabinet adviser to keep him at arm's length.
5.2 Abolishing samurai privileges
In 1876, five years after the removal of the daimyos, the Meiji regime moved on to abolishing the hereditary stipends for the samurai, which it had taken over from the local states. These stipends, representing more than 30% of the annual state expenditures, severely constrained the new state's capacity to invest in economic and military modernization (Ochiai, Reference Ochiai2015: 3). The regime pensioned off the samurai with government bonds, the amount of which corresponded with rank. Interest gains from these bonds were, however, insufficient for most of the approximately 400,000 samurai households (accounting for nearly 5% of the population). Some reacted with anger; four major revolts took place within a year. The last one, the Satsuma Rebellion in 1877, was particularly large in scale, with more than 30,000 former samurais joining the rebel army headed by one of the revolutionary leaders who had left the regime earlier. Nevertheless, while the lack of coordination between these rebels due to previous fragmentation across local states made it relatively easy for the regime to quell them in short order (Ochiai, Reference Ochiai2015: 210–211), they were also neither joined nor supported by the vast majority of the samurai. As with the case of the daimyos, the regime had devised strategies to reincorporate the samurai.
A major way to productively redeploy samurais was to employ them in the new state. According to an estimate made by Saburō Yasuda, around 30% of the former samurai were incorporated into the new unified state. Such redeployment of former civil servants made sense from an administrative viewpoint: the reincorporated ex-samurais were experienced in everyday local governance and possessed good knowledge of local economy and society. Thus, despite the class-egalitarian policy proclaimed by the Meiji regime, nearly 80% of the (now much streamlined) public sector jobs were initially held by former samurais (Yasuda, Reference Yasuda1971: 308): the new state thus maintained skilled local administrators while significantly reducing the scale of the bloated and costly bureaucracy. This also helped alleviate frustration among capable samurais regarding the loss of their privileges by creating for them a new source of pride as leaders of the new society (Sonoda, Reference Sonoda1993: 181–182), and also contributed to the shift of their loyalty to the new centralized state, when their continued identification with the old local states had been one of the major concerns of the nationalist Meiji regime (Shimizu, Reference Shimizu2009: 196).
Moreover, when the Meiji regime's early statist industrialization attempts failed because of the lack of capital, the focus of its economic policy shifted to earning foreign currency by increasing the export of tea and silk to Western markets. This led to the empowered role of local technocrats. While the central government's market regulatory capacity only incrementally grew, the local governments practically filled in the regulatory gaps in an ad hoc manner through this shift (Nakamura, Reference Nakamura2010: 27–29). Furthermore, even the previous Shogunate, which was transferred from Tokyo to Shizuoka after the Meiji Restoration and became only one of the local states (until 1871), supplied a number of administrators for mid- and low-level positions within the central government. Along with Satsuma, Chōshū, and other reformed local states, the Shogunate was home to a number of bureaucrats who studied Western knowledge. They made a return to Tokyo to serve as junior policy makers under the officials of the new regime (Andō, Reference Andō2020: 142).
The education sector was another major field that absorbed the previous samurai. This was also a natural consequence given that the samurai's human capital endowment was higher than that of the rest of the population. Former samurais opened private schools and taught Japanese, Chinese, and Western texts, in addition to mathematics. In 1879, there were 677 private middle schools in the country, compared to 107 public ones (Amano, Reference Amano1992: 33). As the state standardized the school system and curricula, the majority of these samurai-run private schools were then transferred to the government (Ibid.: 50). The ex-samurai bureaucrats were also eager to support their fellows by recruiting them into the public education sector (Kumazawa, Reference Kumazawa2007: 4). Such broad redeployment of the samurai in the education sector resulted in the country's high primary school enrollment rates, which reached 95% by 1900 (Pyle, Reference Pyle1996: 156). Moreover, when the Meiji regime considered the infusing of nationalism into society as a prerequisite for building modern industries and a strong army (Makihara, Reference Makihara1998), it found that the samurai's ethics of loyalty and filial piety were useful to apply. Thus, although the public education curricula were initially Westernized with translated textbooks in the 1870s, the emphasis turned to traditional Confucius morals in the 1880s when the ex-samurai were redeployed in the education sector: they taught pupils the importance of ‘rewarding the nation’, which was symbolically represented by the Emperor (Karasawa, Reference Karasawa1956).
The Meiji regime also encouraged former samurais to engage in economic activities by issuing them commercial loans. Such loans, referred to as shizoku jusan (granting production to former samurais), continued until 1890, causing a number of samurai entrepreneurs to emerge. The regime saw a need to support the commercialization of the ex-samurai to prevent instability, as many, still not knowing how the market worked, had lost their bond-pensions through speculative activities (Ochiai, Reference Ochiai2015: 166). Some jobs were also created for the former samurai and their family members in state projects such as the development of the country's northern frontiers (which also had a national security objective of preparing for Russia's expansion) and the production of raw silk in state-owned factories (where a number of daughters from samurai households found jobs), but the regime, in principle, encouraged self-help in the market (Ibid.: 156). As the samurai were traditionally (before the 17th century) landed in rural agricultural areas, becoming farmers again after centuries was one of the options for them, but not many took up this path because most arable pieces of land were already owned by farmers and migration to remote areas was required to develop new ones (Ibid.: 165).
Not all of the state-sponsored samurai projects were successful. According to one survey, among the 342 state-supported commercial projects targeting samurais, only 46% produced financially positive outcomes, with 19% recording negative outcomes and 35% recording mixed outcomes. Seiichi Andō argues that this was partly because the Japanese economy itself suffered from a recession caused by the deflation policy in the 1880s. There were also sectoral divergences in outcomes: sericultural businesses tended to make more profits because of the growing global demand for raw silk (Andō, Reference Andō1988: 16–20). According to Yukio Okamoto, the projects managed by former upper-class samurais were generally more successful because of their higher capital endowment and greater experience in administrative management in the previous local states (while capable lower-class samurais had captured upward mobility in these local states, they were a minority in the overall lower-class samurai group) (Okamoto, Reference Okamoto2006: 282). Using their new economic prowess, some of these old elites even reestablished their clout as local politicians with the beginning of parliamentary politics in the 1890s, forming an alliance with local technocrats (many of whom were also ex-samurai) and commoner capitalists (Nakamura, Reference Nakamura2010: 315; Okamoto, Reference Okamoto2006: 282). However, even those projects that failed commercially, Andō argues, played a vital role in maintaining political stability and allowing former samurai to soft-land into the world of market competition (Andō, Reference Andō1988: 206; Harootunian, Reference Harootunian1960: 444). It was also a much lower-cost policy of cooptation than continuing with the traditional distribution, and it also ended without extension, thereby not incurring additional cost, when its political objective of maintaining stability was achieved, with new groups of winners and losers becoming clear – thereby the possibility of collective counteraction by the former dominant class disappeared. Furthermore, these sometimes-costly initial investments left some positive legacy for industrial activities at later stages such as infrastructure, the transfer of knowledge and technology, and the lessons from failure for entrepreneurial latecomers to heed (Andō, Reference Andō1988: 20–21).
According to one estimate, more than half the samurai households eventually had access to income either through state employment or commercialization (Sonoda, Reference Sonoda1993: 184). These previous elites were those with financial or human capital who could otherwise have led strong reactionary campaigns if they had not been reincorporated. A number of former samurais, particularly those without financial prowess or administrative skills, still failed to find alternative means of income and were impoverished. However, poverty itself was widespread in Japanese society at this time, and former samurais were still more advantaged than the rest owing to their higher human capital endowment. The children of these former samurais tended to perform better in education and, thus, enjoyed better chances for upward mobility (Smith, Reference Smith1960: 379–380). Thus, in the words of Hiroki Ochiai, the same players continued winning under the different rule (Ochiai, Reference Ochiai2015: 234). In this way, the change agents began developing modern institutions by terminating the wealth-squandering distribution and productively redeploying old elites. The Meiji regime, whose core was largely staffed by those from Chōshū and Satsuma, was often criticized as being exclusive and relying on individual networks of the revolutionary group, but at least a high degree of meritocracy was maintained within, and the regime's reliance on the individual networks was, to some extent, inevitable given that formal state institutions were still under development – and, indeed, the regime's key duty in the first two decades was to build such institutions and their exam-based recruitment systems (Kitaoka, Reference Kitaoka2020; Yamanaka, Reference Yamanaka1974: 17).
5.3 Bargains with agricultural landlords
It is worth adding that the Meiji regime's redeployment of the unlanded samurai occurred in parallel with its bargains with agricultural landlords. This agricultural upper-class began to amass wealth in the pre-Meiji period through the local states’ economic diversification away from rice discussed above. To maintain fiscal and political stability, the regime found itself in need of striking a bargain with this growing class. On the former, taxation on land owned by large numbers of landlords was a major state revenue channel in a pre-industrial Japan, in contrast to some non-Western economies based on small numbers of large estates, where regimes tend to tax on trade (Grabowski, Reference Grabowski2008); on the latter, the regime had to deter landlords’ potential support of samurai rebels. Thus, the regime elicited the cooperation of landlords by establishing property rights, enabling land collateral, ensuring a free land market, and reducing the land tax rate by 0.5% in 1877 in the wake of the samurai revolts (Sorenson, Reference Sorenson2010; Yoo and Steckel, Reference Yoo and Steckel2016).
After the mid-1880s when the samurai redeployment programs entered their last stage, the regime even more proactively sought alliances with these landlords. It was because the regime faced another threat to the stability: rebellions by poor tenant farmers. Tenant farmers, squeezed by landlords, were the major victims of the new regime–landlord relations. The deflation policy in the 1880s hit this exploited class hard, and also accelerated rural polarization as insolvencies deprived many mid-sized landlords of their newly mortgaged land (Matsuzawa, Reference Matsuzawa2018: 9–12). Some of this disillusioned group in destitution came to be organized by breakaways from the regime who formed opposition groups. The regime, therefore, found itself in need of shoring up its ties with the landlords to contain the emerging rural activism (Moore, Reference Moore1966: 257–258).
The regime incorporated the landlords by attracting them first to local policy-making councils (Yamanaka, Reference Yamanaka1974: 316) and then to the Seiyū-kai (Political Fraternity Association), a conservative party formed after the launch of the Diet with the aim to counter opposition parties (Suetake, Reference Suetake and Ōtsu2014). This coincided with the abovementioned shift in the economic policy from premature statist industrialization to increased production of export-oriented commodities to earn foreign currency first. Thus, greater political and economic power was bestowed upon the concentrated landlord class: despite their (in)famously turning into status quo-oriented political interests in later periods, the liberalized agricultural landlord class, at least initially, productively contributed to development (Chang, Reference Chang2011: 482) and played a key role in the regime's maintaining the reform-stability compatibility.
5.4 Creating a modern financial sector
Another noteworthy issue is Japan's effective modernization of its financial sector. In particular, the relevant agenda is how Japan's financial sector eschewed the path of falling into populist rent-seeking institutions by instead being founded on productive coalitions (Calomiris and Haber, Reference Calomiris and Haber2014). Indeed, Japan's financial modernization was tied to the elite redeployment discussed above. The pensioning-off of the samurai with government bonds necessitated that the regime develops a regulated banking sector since, to maintain the stability, the regime had to prevent the fall of the price of these bonds through the pensioned ex-samurai converting them to gold. The regime instead directed these bonds to modern banks by issuing inconvertible note (Ishii, Reference Ishii2012: 41).
Moreover, the regime utilized the previous daimyos’ ample assets for modernization. In 1877, the Fifteenth National Bank was created, with Motonori Mōri as president – the son of Takachika, the previous lord of Chōshū, who died in 1871 prior to the abolishment of the local states. This bank, which was dubbed as kazoku ginkō (the bank for noblemen), aimed at turning the previous daimyos into modern financiers, solving the ‘trilemma’ facing Japan's burgeoning modern financial sector at that time: lack of capital; the Meiji regime's preference for eschewing foreign capital; and domestic capitalists’ hesitance to invest in new industries unknown to them.
First, despite the early focus of the state economic policy being the development of state-owned modern industries, this attempt was significantly constrained by lack of capital when the regime simultaneously was spending to modernize its military. There was, thus, a need to nurture private industries through private capital. In the pre-Meiji period, local merchants were major financiers, many of them being crony-capitalists profiting from the state-directed economic diversification projects discussed above. With the demise of the local states, many of these traditional merchants, however, declined, being unable to recover their loans to their patron state and losing major economic opportunities provided by them (Nakamura, Reference Nakamura2010: 20; Ochiai, Reference Ochiai2015: 92). Second, the Meiji regime prohibited the financing of economic projects with foreign capital. This was based on the revolutionary change agents’ understanding of the experience of other non-Western countries: they saw that the colonization by Western empires began with financial penetration, which led to the gradual deprivation of sovereignty via insolvency. Thus, except for some limited issuance of sovereign bonds, the Meiji regime avoided borrowing from abroad and banned private industries from inviting foreign investment (Ishii, Reference Ishii2012: 42–43). Third, while the burgeoning industries, thus, relied on domestic investors, including the would-be zaibatsu, many in the early days remained hesitant to invest in modern sectors because of their lack of knowledge about the imported new technologies. The regime's early negotiations with private investors to finance railway projects failed, for instance (Nakamura, Reference Nakamura2010: 26).
The daimyo money solved this trilemma, providing capital to nascent industries such as railway and textile and creating a ‘signaling effect’ to commoner investors, helping the regime achieve its goal of industrializing with domestic capital. It also had other positive side effects. These previous local lords, now incorporated into the central peerage and residing in Tokyo, helped build networks of money, people, and knowledge between the capital and the regions of their origin. They also often invited their previous vassals with expertise to work on the projects they financed (Nakamura, Reference Nakamura2010: 22).
Led by the capital of the daimyo and the samurai, the country's modern financial sector grew rapidly, attracting merchants, both those who survived the political transition among the traditional ones and emergent ones who profited from the country's new international opening through trade and exchange, as well as the growing agricultural landlords who began to rise as major players in local finance (Nakamura, Reference Nakamura1966: 437–439). This made the structure of Japan's financial sector dualist, consisting of large banks founded on urban-based elites and a number of local banks financed by local capital, both serving modern development in a productive manner.
6. Comparative perspective
This article has demonstrated how the quality of institutions improved in Meiji Japan through the coercive deprivileging of the old elites and the redeployment of those with financial or human capital among them in productive institutions, which enabled the regime to implement reform while maintaining stability, even without the causal paths for development and state building observed in Europe, such as political participation and war. Japan's different path of institutional upgrading suggests the heterogeneity and causal equifinality in development before the 20th century. While early parliamentary politics and regional interstate wars shifted the patterns of elite bargains to the detriment of vested-interest actors such as landed nobles in historical Europe, the revolutionary authoritarian regime deterred reactionary rollbacks in Japan by making reform and stability compatible with each other through the elite redeployment. It is worth adding that despite the difference, both European and Japanese models were elitist and products of the age prior to the rise of socialism and Keynesianism: their modernization process was, thus, less affected by pressure for popular distribution (Vu, Reference Vu2010; Waldner, Reference Waldner1999), which emerged in the following century and possibly made keeping open the fiscal and administrative spaces, liberated from the old rent-seeking elites for new productive players, not as easy as before because the legitimacy of regimes now hinged upon distribution beyond the narrow circle of elites.
Japan's elite redeployment model was repeated in colonial Korea. Under the Japanese colonial regime, Korea's ruling Yi family was incorporated into the Japanese peerage and the aristocratic yangban elites were pensioned off from the state. Because the yangban were, unlike the samurai, landed elites, they were turned into productive players primarily as agricultural producers, especially producers of rice, through the transfer of modern agricultural technologies to them and through granting them access to the Japanese market (Kohli, Reference Kohli1994, Reference Kohli2004). Productive institutions installed in the colonial period that replaced the traditional monarchical patrimonial institutions were largely preserved in post-1945 South Korea by the US occupation forces and the Rhee Syngman regime, handed over to the Park Chung-hee regime, under which South Korea saw an industrial takeoff (Vu, Reference Vu2010). Colonial Taiwan, on the other hand, did not follow the same model. Or, more precisely, there was no need for the elite redeployment model there because powerful local vested interests were absent. The island had long been a peripheral territory of China under the Qing Dynasty and was inhabited only by tiny populations of Chinese migrants who had formed egalitarian agricultural and commercial communities, in addition to the island's small indigenous communities. The colonial regime, thus, replicated Japanese institutions there without much coping with incumbent institutions (Hsieh, Reference Hsieh2011). Such a so-to-speak ‘institutional tabula rasa’ model deriving from the émigré nature of the society has also been observed in other Chinese peripheries and offshoots such as Manchuria and Singapore.
Acknowledgements
The author is very grateful to the following people for their kind support: Atul Kohli, Eunsook Jung, Eriko Kumazawa, Nobuhiro Kiyotaki, Keiko Kiyotaki, Hiroshi Mitani, Takuya Onoda, Koji Hirata, Faris Al-Sulayman, Steffen Hertog, Bernard Haykel, Masashi Haneda, his colleagues at the Tokyo College, the editors of the Journal of Institutional Economics, and the anonymous peer-reviewers.
Open access
