Waseda Institute for Advanced Study (WIAS)早稲田大学 高等研究所


Organizational votes: A possible cause of the weakening of center-left political parties?
IKEDA Fumi, Assistant Professor

IKEDA Fumi, Assistant Professor

Volunteering at an election office opened my door to political science

My research theme is political parties and party bases. Currently my main focus is the organizational votes of labor unions that support political parties.

Although I studied in the Faculty of Letters as an undergraduate, my experience as a volunteer election worker, in jobs such as campaign truck announcer, gave me an interest in elections that led me to pursue political research. My starting point was my on-site election experience, and I still maintain that real-world perspective in my research.

It is said that today center-left political parties around the world, including those in Europe, are losing momentum. Japan is no exception. Center-left political parties are the parties slightly to the left of center, between conservatives such as the Liberal Democratic Party and leftists such as the Communist Party. I generally position the Constitutional Democratic Party of Japan and the Democratic Party for the People (which were formed after the break-up of the former Democratic Party of Japan) as center-left parties. The Democratic Party of Japan took power from the Liberal Democratic Party of Japan in 2009 and became the ruling party, but in 2012 it became the opposition party, rapidly lost support, and broke up. That evolution is an example of how Japanese center-left political parties are weakening.

The literature on party politics often cites the decline in union membership rate (number of workers in unions/total number of workers) as one cause of the weakening of center-left political parties. The logic is simple: if a union’s membership declines, the mobilization power of its organizational votes also declines, which in turn leads to the weakening of the center-left. However, I wondered if a decrease in a union’s membership would in fact lead to a decrease in the mobilization power of that union’s organizational votes, so I decided to investigate the relationship between the union membership rate and the mobilization power of organizational votes of labor unions that support center-left political parties.

2. Exploring the relationship between union membership and vote mobilization power

There are six representative unions that support Japan’s center-left parties. Table 1 shows the number of members of each labor union in 2016, the number of votes gained by the candidates they sponsored in the Upper House election in the same year, and the average number of votes gained by the candidates they sponsored in Upper House elections from 2004 to 2016 (Table 1).

Table 1. Number of union members and number of votes for sponsored candidates

Labor union by industry Number of union members in 2016 Number of votes in the Upper House election Average value of votes in the Upper House election from 2004 to 2016
 UA Zensen

(Japanese Federation of Textile, Chemical, Food, Commercial Service and General Workers’ Union)

1,534,354 196,023 176,404

(The Confederation of Japan Automobile Worker’s Unions)

770,067 266,623 242,541

(All-Japan Prefectural and Municipal Workers Union)

806,987 184,187 245,713
 Denryoku Sōren

(The Federation of Electric Power Related Industry Worker’s Unions of Japan)

214,413 270,285 241,765

(Japan Teachers’ Union)

246,011 176,683 183,910

(The Federation of Information and Communication Technology Service Workers of Japan)

213,413 171,486 203,184

Source: Asahi Shimbun

The number of members of each labor union supporting center-left political parties (2016); the number of votes gained by sponsored candidates in the Upper House election (2016); the average of votes gained by sponsored candidates in the Upper House elections (2004-2016) are listed.

In this table, for example, the union JAW had about 770,000 members in 2016, but the number of votes gained by their sponsored candidate in the Upper House election that year was about 260,000. In contrast, the union UA Zensen had about 1.5 million members in 2016, while their sponsored candidate gained only about 190,000 votes in the 2016 Upper House election. Clearly UA Zensen has more members than JAW, but the mobilization of UA Zensen’s organizational votes was small. Comparing across labor unions, it cannot be said that a large number of union members results in a large number of organizational votes, or that a small number of union members results in a small number of organizational votes. In other words, we have no conclusive evidence that the number of members in an organization is directly linked to the number of organizational votes.

Figure 1. Organization rate of labor unions in Japan, 1990–2015 The organization membership rate is decreasing year by year.

In recent years, the number of labor union members in Japan has been declining, as has the organization rate for those unions (Fig. 1). However, the number of votes gained by sponsored candidates in the 2016 Upper House elections, shown in Table 1, has not decreased for all labor unions relative to the figures for 2004–2016. Moreover, there is no conclusive evidence that the decrease in the number of organizational votes is the result of the organization rate, even though it too is declining year by year (Fig. 1).

What determines the mobilization power of organizational votes?

Then, what is the driver of the mobilization power of organizational votes? To investigate the contrast between JAW and UA Zensen, I examined five aspects: the growth rate of the industry; the type of enterprise that the union represents; the size of the enterprise union; whether or not there is a merger of enterprise unions; and the presence or absence of a company town centered in the region. The results are shown in Table 2.

Table 2. Comparison of JAW and UA Zensen

JAW UA Zensen
Growth rate of the industry Tendency to grow Not all industries are growing.
Industry of member organization Only the automobile industry Member industries include sectors such as textiles, services, and distribution.
Organization Enterprise union is of a considerable size. Most of enterprise unions are relatively small.
History of merges Never merged. Has experienced merger several times.
Company town (geographic concentration of resources) There are company towns and the resources are concentrated. Some companies have company towns, but the number is small and the resources are not concentrated.


First of all, JAW is in a growing industry; it consists of only enterprise unions of large automobile companies. In addition, there have been no union mergers, and the companies have company towns rooted in their regions. UA Zensen, on the other hand, is made up of small enterprise unions for various industries in sectors such as textiles, chemicals and transportation. Not all of them are growth industries, and the unions have been merged several times. Even if there are company towns rooted in some regions, the number is declining. Therefore, it can be assumed that the factors that determine the mobilization power of organizational votes are also characteristics of JAW; these include unified structure representing multiple organizations; company towns rooted in regions; representation of large companies; absence of mergers; and current growth in the industry.

Directions for future study: Comparison of center-left political parties in Japan and Europe

From the above results, I conclude that even if labor union membership numbers are decreasing, the ability to mobilize organizational votes is not necessarily in decline—and it appears that the unions’ electoral clout remains strong. That leaves one question unanswered: what is the cause of the weakening of Japanese center-left political parties? I aim to continue pursuing these issues, and in the future I am going to compare the causes of a comparable center-left weakening in Europe with those in Japan, in order to identify similarities and differences and elucidate the factors that drive these phenomena.

Interview and composition: AIMONO, Keiko
In cooperation with: Waseda University Graduate School of Political Science J-School

Page Top
WASEDA University