The Liberal Democratic Party in Japan: Explaining the Party's Ability to Dominate Japanese Politics
IN THIS ARTICLE
The Liberal Democratic Party’s largely uninterrupted dominance of Japanese politics must be ascribed to processes which transverse electoral systems and periods of economic vigour. This essay proposes that clientelistic behaviour within the Japanese political system best explains the LDP’s dominance of Japanese politics. Clientelism, an exchange of benefit for voter support (Scheiner 2006: 64) evolved from structural factors in the Japanese political system and was harnessed by the LDP to maximise its tenure.
Structural factors conducive to clientelism included fiscal centralisation, the pre-1994 electoral system and electoral malapportionment. Fiscal centralisation provided a context for the commoditisation of votes for material gains. The pre-1994 electoral system, Single Non-Transferable Vote in Multimember Districts (SNTV/MMD), encouraged the proliferation of koenkai networks and money politics, entrenching clientelistic behaviours in elections. Electoral malapportionment was a result of the pre-1994 electoral system and encouraged politicians to appeal to segments of the population through pork-barrel politics and protectionist policies. This structural defect also allowed the LDP to garner a majority of seats, without a majority of votes.
Whilst clientelism evolved from structural factors, its continuing presence in the post-1994 electoral system can be ascribed to the creation of a clientelistic ‘norm’, whereby the Japanese citizenry expect to receive dividends from their votes and candidates’ continue to see clientelism as tantamount to electoral success. Clientelism, therefore, has permeated the fabric of Japanese politics, especially in rural areas. Its presence has conditioned the electorate to prolong the LDP’s tenure in return for material benefit; allowing the LDP to retain power long after it failed to gain a majority of the popular vote.
I. STRUCTURAL FACTORS
A. Fiscal centralisation
The presence of clientelistic behaviours in Japanese politics originated from structural factors in the political system. The primary cause for the emergence of clientelistic behaviour was tight fiscal centralisation in Japan. It is rare for Japanese rural prefectures to have access to substantial fiscal resources, instead relying upon the national government (Fukui & Fukai 1996: 272; Scheiner 2006: 61). Indeed, Ethan Scheiner estimates that local prefectures source 70 per cent of their revenue from the national government (Scheiner 2005: 805). As such, the 47 prefectural governments are engaged in a constant struggle to obtain funds from national coffers. Hence, Diet members are not simply representatives of their constituents; they act as conduits, or “pipelines” between the national treasury and their respective prefectures (Hirano 2006: 57).
The LDP has capitalised significantly on fiscal centralisation. Electoral candidates are not judged qualitatively on policy considerations, rather, judgement is based on quantitative grounds; the ability to facilitate the transfer of financial benefit to constituents. J.A.A. Stockwin describes the perception of politicians as ‘monetary or material distributionists’ (Stockwin 2008: 178). Nowhere is this more salient than the case of Tanaka Etsuzankai, who provided significant (and infamous) public works services to his electorate, including high speed train lines and expressways. Bradley Richardson observes that the prolific talent Tanaka had in securing pork led to his repeated electoral success (Richardson 1997: 29). The reduction in qualitative differentiation of candidates was favourable to the LDP: it led to an erosion of serious political debate as it denies the opportunity for opposition to scrutinize the ruling party and distinguish themselves from the incumbents. In fact, clientelism can account, to a moderate degree, of the inability of the opposition in Japan to usurp the LDP. Although there were key structural deficiencies in the composition of Japanese oppositions, such as their fragmentation (Cox & Niou 1994: 230; Kohno 1997: 120), the opposition had ample opportunity to unseat the LDP. For instance, major corruption scandals affected the elections of 1967, 1983, 1990 and 1993 (Nyblade & Reed 2008: 930) and policy failures following the stagnation of the economy should have seen the removal of the LDP from power longer than a brief 10 months in 1993-94. However, the opposition was disadvantaged because large blocs of votes had been secured on the basis of clientelist appeals (Scheiner 2006: 4). The LDP, therefore, achieved electoral security through clientelism even in the wake of political failures. As the essay describes later, this electoral security was also guaranteed by the opposition’s failure to gain seats in rural prefectures, which were key to forming a national government. Nevertheless, the LDP’s ability to remain in government more than a decade after the Japanese economy began to stagnate is illustrative of the extent to which qualitative differentiation has declined in Japanese politics (McElwain 2008: 39).
Moreover, fiscal centralisation bequeathed further electoral benefits to the LDP’s national governments: constituents had incentives to elect and re-elect those candidates who were influential in the national government to secure resources for their prefecture (Scheiner 2005: 809). The mantle of influence largely fell upon LDP candidates, who generally boasted governmental experience. In fact, a quarter of all new LDP candidates had been a mayor or prefectural assembly member (Scheiner 2006: 137). The structural incentives to vote for an LDP candidate are augmented if the LDP-led national government exhibited clientelistic behaviour. The distinction between pork-barrel politics and clientelistic behaviour is that the latter features the elements of punishment and reward. Under a truly clientelist system, non-LDP prefectures should receive less government funding. Threats to cuts funding for projects in non-LDP prefectures were commonplace (Asashi Shinbun 19 October 1999; Asashi Shinbun [Online] 6 March 2002), however, statistical evidence on the LDP’s “carrot and stick” allocation of fiscal resources is inconclusive (Reed 2001). Nonetheless, there is an academic consensus that regions which were better connected to the LDP receive a greater distribution of resources (Fukui & Fukai 1996: 285; Scheiner 2005: 808). It is arguable that this was also a popular preconception among the electorate. Hence, clientelistic behaviour, (or the threat of) coordinated votes irrespective of voter-party ties in favour of the LDP (Cox & Thies 2000: 39). Prolonged LDP success, therefore, was underpinned by Japan’s fiscal centralisation, which conferred advantages to those politicians who could bring resources to their constituents. These politicians largely hailed from the LDP, generating impetus for their consistent re-election.
B. The effect of SNTV/MMD
Where fiscal centralisation acted as the genesis of vote commoditisation, the SNTV/MMD system’s emphasis on the personal vote and the cultivation of support networks (koenkai) entrenched the process into electoral behaviour. The LDP thrived in the consequent environment of kinken-seiji (money-power politics). The SNTV/MMD system pitted intraparty and interparty candidates against one another. Thus, candidates were invariably pressed to pursue a personal campaign strategy, in order to differentiate themselves from their competition (Reed 1994: 22). Candidates were the cornerstone of Japanese voting behaviour, rather than party-ties or issue-based concerns (Hrebenar 1986: 21). Elections were typified by the ability to attract votes, rather than substantial policy matters (Carlson 2006: 234; Reed 1994: 22; Swindle 2002: 286). This reinforced the decline in qualitative differentiation described earlier. Additionally, the focus on candidates and vote-attraction manifested itself through the creation of personal support networks, called koenkai. Essentially, koenkai are platforms of support manifested in exchange for monetary or lobbying favours (Scheiner 2006: 71). Koenkai were of great use to political candidates because they were more personalised than a party branch, and could focus on the provision of specific goods and services for direct candidate support (Stockwin 2008: 140; Swindle 2002: 282). Particularly startling is the pervasiveness of these networks: in 1989, a national newspaper reported that over half the electorate, around 40 million eligible voters, were involved in koenkai (Richardson 1997: 27). Due to their widespread utilisation, koenkai broadcasted that the exchange of gifts for personal electoral support was acceptable electoral practice and this created a precedent for large-scale clientelism (Cox & Thies 2000: 40).
The chief precedent created by koenkai was the LDP’s recognition that money, or the promise of a similar material benefit, was tantamount to electoral success. By courting constituents with larger material benefits than their interparty or intraparty opponents, LDP candidates’ election or re-election opportunities were buffeted (Carlson 2006: 241). The use of money to coordinate votes is acutely felt amongst Japanese constituents; where party affiliations are seldom strong (Richardson 1997: 23) and roughly half the electorate is undecided (Scheiner 2006: 68). The LDP candidates’ extensive contacts with Japanese business facilitated their ability to provide larger material benefits. Millions of yen donated by business groups in support for LDP candidates are commonly reported (Nadell 1990: 40; Richarson 1997: 180). This provided the LDP with a tremendous advantage, but also increased the temptation for corruption (Stockwin 2008: 176). Matthew Carlson’s analysis of personal support in Japan found that, for the 1966, 1990 and 1993 elections, LDP lower house candidates reported a yearly income of ¥209.4 million (Carlson 2006: 245). Mustering this amount of monetary support was clearly beyond the means of most opposition parties, who were without the extensive business contacts of the LDP. However, recent political parties, such as the Democratic Party of Japan (DPJ) have been established by former LDP politicians. The increased presence former LDP members have in opposition could see opposition parties boasting greater business contacts, and improved fiscal competitiveness with the LDP. Nevertheless, the excessive cost of maintaining koenkai networks and challenging LDP incumbents had a prohibitive effect on most opposition parties; many could not afford to run candidates in all districts, thereby sacrificing Diet seats for financial austerity. Particularly under a SNTV/MMD system, where one electoral district can translate into the acquisition of multiple lower house seats, the opposition’s inability to run some candidates only increased obstacles to attaining national government. The SNTV/MMD system and its resultant emphasis on personalised voting cemented the normality of clientelistic appeals into the consciousness of politicians and constituents alike. The LDP capitalised on this context; the SNTV/MMD system was suited to the LDP’s tactics, and as a result, was far more successful in attracting personal votes than other parties.
C. Electoral malapportionment
In addition to fiscal centralisation and the SNTV/MMD system, disproportionate allocation of seats under the SNTV/MMD system was directly exploited by the LDP’s clientelistic behaviour to gain electoral advantage. By specifically targeting electorally powerful sub-constituencies with clientelistic appeals, the LDP retained power long after it failed to gain the majority vote in 1963. Urbanisation in Japan led to accompanying population shifts from rural to urban areas; around 10 per cent of the Japanese population live on farms (Richardson 1997: 155). The 1947 Electoral Law did not provide for an independent body to review electoral boundaries and reallocate seats (Kishimoto 1988: 16; Stockwin 2008: 177). Instead, electoral districts largely remained unmolested by review and, consequently, rural prefectures contained more Diet seats per capita than urban districts (Scheiner 2006: 57). This increased the value of rural votes to almost three times that of an urban vote (Mulgan 1997: 882).
The LDP realised that this malapportionment could be used to effect a situation where it obtains more seats than votes (Cox & Niou 1994: 221). There was thus little incentive for the LDP to correct the discrepancy. This was, according to J.A.A. Stockwin, a ‘negative gerrymander’ (Stockwin 2008: 177). While some commentators feel that the failure to redraw electoral boundaries was malign (Mulgan 1997; Pempel 1992), Steven Reed advances that the appeal to powerful sub-constituencies within malapportioned constituencies had some legitimacy: ‘[e]lectoral systems structure political competition […] Politicians learn which strategies work best under their electoral system and politicians who use more effective strategies will tend to dominate the politics of the country’ (Reed 1994: 21). With the votes of rural areas overrepresented, the LDP forged a particularly clientelistic relationship with these sectors. The LDP-led national government used restrictions on agricultural imports to shelter farmers from foreign competition in conjunction with direct subsidies to shore up rural support (Gordon 1990: 946). Economic protectionism of rural constituents was blatant: restrictions on agricultural goods, in a time of increased economic and trade liberalisation, were in place far longer than those on industrial goods (Richardson 1997: 157). This created significant incentive for rural constituents to vote for LDP candidates to ensure the retention of the LDP national government (Inoguchi 2005: 103). Thus, the LDP’s fervent clientelistic appeal to the rural prefectures frequently defied economic sense (Richardson 1997: 233) but ensured the continued support of the intended audience.
The LDP was very successful in pursuing this strategy. Indeed, the number of rural seats held by non-LDP parties remained consistently low (Scheiner 2005: 803), the stable support resting on the clientelistic relationship between patron (the LDP-led national government) and client (the rural constituents). This explains the weakness of opposition parties in rural areas. The sustained failure of opposition parties to challenge LDP incumbents has also been ascribed to the LDP’s changes to campaign periods. By reducing the length of the official campaign period from 20 days in 1958 to 12 days in 1994, Kenneth McElwain posits that the LDP reduced the ability of opposition parties to challenge incumbents (McElwain 2008: 33, 41). While McElwain advances a well-contended point, it is imperative to stress that by virtue of koenkai-created networks of personal support, campaigning in Japan is a year-round process (Pempel 1992: 16). As such, the ability for nonincumbents to vie for election in their constituency is not restricted to the official number of days. McElwain’s argument aside, election to rural prefectures, due to their low seat-vote threshold, was central to forming a national government. Moreover, without success in rural elections, opposition parties had little governmental credibility (Scheiner 2005: 803). With consistent support in rural areas, based on clientelist networks, the LDP consistently achieved parliamentary majorities without accruing a majority of the popular vote.
II. NORMATIVE FACTORS
The deficiencies of the SNTV/MMD system, including electoral malapportionment, cannot completely govern the analysis of LDP dominance. In 1994, the lower house electoral system was changed to a mixed-member majoritarian system (MMM), which combined 300 single-member districts (SMD) with 180 proportional representation (PR) seats. Malapportionment of seats was largely corrected, although no system can perfectly ensure an equal distribution of seats per capita (Mulgan 1997: 879). Following the 1994 reforms, the LDP regained power and continued its dominance of Japanese politics until the 2009 elections. The corollary is that further factors exist which contributed to the sustained electoral dominance of the LDP. This essay posits that clientelism continued to exist even after the LDP’s structural advantages had largely been eroded through electoral reform. The continuity of clientelism was due to its place as a norm in Japanese politics. Norms are behavioural precepts (Kramer 2007: 97) which establish a model of behaviour. As the essay has demonstrated, the dominant practice of Japanese politics was consistent support based on social contracts formed through clientelism (Richardson 1997: 46). The model of behaviour created in Japanese politics is summarised diagrammatically below.
Clientelism is embedded in the Japanese political system. It is ingrained in both the expectations of the client and the behaviours of the patron. Although clientelism is certainly not what can be considered a “positive” established practice, it is still expected by the electorate, to a significant degree (Stockwin 2008: 178). This expectation rests on the perception that (1) votes are commodities to be exchanged; and (2) politicians are channels for monetary or material benefit. The behaviour of the patron continues to lend itself to such expectations. Clientelistic behaviours are also embedded in the structure of political parties’ election strategy. Despite the change in lower house electoral system, representation still occurs on an individual level (Swindle 2002: 279). Although there is not a need to differentiate between candidates from the same party, candidates from different parties must still vie for election. Thus, the purpose of koenkai – to provide a base of support for individual candidates – remains unchanged. Koenkai were the precedent for national clientelist behaviours. As these support networks still factor into parties’ electoral strategies, this essay asserts that kinken-seiji has been left untouched by the 1994 reforms. Despite tight financial restrictions imposed through the electoral reforms, Carlson showed that for the 1996, 2000 and 2003 elections, each LDP politician possessed almost six support groups and reported an average yearly income of ¥102.2 million (Carlson 2006: 245).
The yearly income and support networks have effectively been halved from their pre-reform state (ibid); however, stipulations pertaining to annual income reports can be circumvented (Pempel 1992: 17). Nevertheless, this volume of income is improbable in a koenkai-free environment. If anything, the need for koenkai may have increased, despite the removal of intraparty campaigning for electoral seats. Party affiliation has significantly declined following 2000; today, only 20 per cent of the eligible electorate firmly supports the LDP (Scheiner 2005: 801). A steady increase in the desire for appealing and charismatic candidates has widened the number of undecided voters in Japanese politics (Stockwin 2008: 145), and has increased the role of modern audio-visual media in political discourse (Köllner 2006: 252). This More pressingly, in an environment of reduced identification with parties, there is incentive for LDP candidates, and candidates of other parties, to revert to clientelism. In addition, those LDP members who are part of particular zoku (tribe) have an acute incentive to prolong pork-barrelling.
Zoku are groups of politicians who represent a specific policy area (for instance, construction or agriculture) and pander to the interest groups within that area (Kawabata 2008: 102). In return, interest groups provide those politicians with monetary and campaign support during elections (ibid). On a local level, gifts to influential sub-constituencies are continually relied upon to manufacture electoral dividends (Carlson 2006: 250; Hirano 2006: 61-2). Anecdotal evidence, such as public works spending (which has remained over 10 per cent of the total account budget since 1955), illustrates that material benefits are still being exchanged between the national government and rural prefectures (Richardson 1997: 29). We can conclude that money and promise of material benefit, is still, in the eyes of Japanese political parties, paramount to obtaining votes. Thus, money politics and koenkai networks still play a role in shaping the electoral fortunes of politicians in a post-reform environment (Carlson 2006: 245). Clientelism is still alive and well in Japanese politics, due to its normative place in the mindset of the clients and tactics of the patrons. The LDP, as the chief proponent of clientelistic behaviour is therefore best-suited to the continuity of clientelistic electoral dynamics under the MMM system.
III. COMPLEMENTARY FACTORS
The LDP’s dominance of Japan is one of the world’s premier examples of uninterrupted conservative rule in an industrialised democracy (Pempel: 1992: 13). It would be naïve, therefore, to ascribe the LDP’s dominance of Japan’s political realm to a single factor. Although this essay has shown that clientelism affected multiple aspects of the Japanese political system and is the best explanation of the LDP’s dominance, it is not the only explanation. The fragmentation of Japanese opposition parties should be regarded, along with clientelism, as concurrent causes of systemic LDP dominance in the pre-1994 era.
Following the 1958 elections, the Japanese political scene looked to be evenly split between evenly the LDP and the Japan Socialist Party (JSP) (Kohno 1997: 117). However, tumult within the JSP led to the proliferation of parties through the 1960s and 1970s, including the Democratic Socialist Party, Clean Government Party, New Liberal Club and Socialist Democratic Federation. This fragmentation (which continued well after the 1960-1970 period) had reduced the opposition’s effectiveness on a number of levels. First, it ensured that the only way for opposition parties to form a serious effort to unseat the LDP national government was through a coalition (Hrebenar 1986: 7). Effective coalition-formation, an already difficult process was further complicated by the tenuous position of the LDP after the 1976 and 1979 elections. This led the LDP to pursue a strategy of deal-making to secure its Diet position (Mulgan 2000: 75). The offer of power-sharing was an efficient lure to dissuade opposition parties from competition with the LDP (Mulgan 2000: 71; Stockwin 2006: 129). Deal-making politics was not a fleeting process; it was later replaced by coalition-building after the LDP’s brief electoral defeat in 1993. Both deal-making and coalition-formation turned the opposition’s attention from forming a competent challenge to the LDP to siding with it; blunting any initiative to form an alternative government. Secondly, the multiplicity of alternative parties which emerged from the JSP’s division were rendered ineffective because they could not maximise their vote potential. That is, under the SNTV/MMD system, the vote for non-LDP parties was dispersed widely between alternative parties, rather than concentrated in one opposition party, such as the JSP (Cox & Niou 1994: 230). In turn, this reduces the capacity for opposition parties to attain Diet seats, and prolonged the electoral success of the LDP. As a ‘coalition of factions’ (Abe, Shindo & Kawato 1994: 123) the LDP was not a “unitary” party in any sense, however, it did not physically fragment until Ichiro Ozawa’s breakaway in 1993. This allowed it to increase its staying power in the face of an ineffective opposition.
This essay has demonstrated that clientelism, an extension of pork-barrel politics, has been the chief method utilised to manufacture electoral success and prolong the dominance of the LDP. Although many factors have been cited as reason for the LDP’s domination of Japanese politics, clientelism is the only explanation which transverses electoral systems, politicians and periods of economic vigour. Structural factors within the Japanese political system, such as fiscal centralisation and the SNTV/MMD electoral system produced a context conducive to clientelism which the LDP exploited to full effect: during a fifty-year period (1955-2005), the LDP had been out of power a mere 10 months and 20 days. The fragmentation of opposition further increased the LDP’s pre-eminence in the political system. The continuation of the LDP’s dominance following electoral reform in 1994 can be ascribed to the place of clientelistic behaviour as a norm in Japanese political consciousness; the voters expect it, and political candidates employ it without question. So long as clientelistic appeals are effective in attracting and retaining the support of the Japanese electorate, then clientelism’s presence in Japanese politics is a continued possibility.
Browne, Eric and Kim, Sunwoong (2003), ‘Factional rivals and electoral competition in a dominant party: Inside Japan‘s Liberal Democratic Party, 1958-1990.’ European Journal of Political Research, Vol. 42, pp. 107-134.
Carlson, Matthew (2006), ‘Electoral Reform and the Costs of Personal Support in Japan.’ Journal of East Asian Studies, Vol. 6, pp. 233-258.
Christensen, Raymond (1994), ‘Electoral Reform in Japan: How it was enacted and Changes It May Bring.’ Asian Survey, Vol. 34, No. 7, pp. 589-605.
Cox, Gary and Niou, Emerson (1994), ‘Seat Bonuses under the Single Non-transferable Vote System: Evidence from Japan and Taiwan.’ Comparative Politics, Vol. 26, No. 2, pp. 221-236.
Cox, Gary and Thies, Michael (2000), ‘How Much Does Money Matter? “Buying” Votes in Japan, 1967-1990.’ Comparative Political Studies, Vol. 33, pp. 37-57.
Fukui, Haruhiro and Fukai, Shigeko (1996), ‘Pork Barrel Politics, Networks and Local Development in Contemporary Japan.’ Asian Survey, Vol. 36, No. 3, pp. 268-286.
Gordon, Peter (1990), ‘Rice policy of Japan’s LDP: Domestic Trends toward Agreement.’ Asian Survey, Vol. 30, No. 10, pp. 943-958.
Hirano, Shigeo (2006), ‘Electoral Institutions, Hometowns, and Favored Minorities: Evidence from Japanese Electoral Reforms.’ World Politics, Vol. 58, pp. 51-82.
Hrebenar, Ronald (1986), The Japanese Party System: From One-Party Rule to Coalition Government. Colorado: Westview Press.
Inoguchi, Takashi (2005), Japanese Politics: An Introduction. Melbourne: Trans-Pacific Press.
Kohno, Masaru (1997), Japan’s Postwar Party Politics. Princeton: Princeton University Press.
Mori, Kenneth M. (2008), ‘Manipulating Electoral Rules to Manufacture Single-Party Dominance.’ American Journal of Political Science, Vol. 52, No. 1, pp. 32-47.
Mulgan, Aurelia G. (1997), ‘Electoral Determinants of Agrarian Power: Measuring Rural Decline in Japan.’ Political Studies, Vol. 155, pp. 975-999.
____________________ (2000), ‘The dynamics of coalition politics in Japan.’ Asia-Pacific Review, Vol. 7, No. 2, pp. 66-85.
Nyblade, Benjamin and Reed, Steven (2008), Who Cheats? Who Loots? Political Competition and Corruption in Japan, 1947-1993.’ American Journal of Political Science, Vol. 52, No. 4, pp. 926-941.
Pempel, T.J. (1992), ‘Political Parties and Representation: The Case of Japan.’ Political Science and Politics, Vol. 25, No. 1, pp. 13-18.
Reed, Steven (1994), ‘Democracy and the Personal Vote: A Cautionary Tale from Japan.’ Electoral Studies, Vol. 13, No. 1, pp. 17-28.
Richardson, Bradley (1997), Japanese Democracy: Power, Coordination and Performance. New Haven Yale University Press.
Scheiner, Ethan (2005), ‘Pipelines of Pork: Japanese Politics and a Model of Local Opposition Party Failure.’ Comparative Political Studies, Vol. 38, pp. 799-823.
__________________ (2006), Democracy Without Competition in Japan: Opposition Failure in a One-Party Dominant State, New York: Cambridge University Press.
Stockwin, J.A.A. (2008) (4th edition), Governing Japan: Divided Politics in a Resurgent Economy. Victoria: Blackwell Publishing.
Swindle, Stephen (2002), ‘The Supply and Demand of the Personal Vote.’ Party Politics, Vol. 8, No. 3, pp. 279-300.
Thies, Michael (1998), ‘When Will Pork Leave the Farm? Institutional Bias in Japan and the United States.’ Legislative Studies Quarterly, Vol. 23, No. 4, pp. 467-492.