Concern about the rise of lazy welfare queens? An empirical explanation of the underdevelopment of the redistributive welfare system in South Korea

Abstract

This study investigates why South Korea has maintained a minimalist welfare state with little redistribution of income. Inspired by the behavioral/attitudinal approach of Alesina and his colleagues, this study focuses on the perception that people who do not work become lazy. This belief is related to the anti-welfare sentiment that non-working benefits encourage laziness. This study shows that perceptions of work and laziness are associated with preferences for redistribution, not only among South Koreans, but also among individuals in Organization for Economic Cooperation and Development (OECD) member countries. Further, such perceptions are also associated with the redistribution policies chosen in a country. This study provides evidence that societal beliefs about work and laziness, along with several other political and economic factors, such as pre-tax income inequality, political institutions, and union density, may explain the small-scale redistribution in South Korea.

Keywords:

• Redistribution
• Welfare policy
• Income inequality
• Social belief

Notes

1 Existing theories have considered many other determinants of the size of welfare and redistribution, such as pre-tax income inequality (Lindert, 1994, 1996; Meltzer & Richard, 1981), trade openness (Rodrik, 1998), electoral rules (Persson & Tabellini, 2002, 2005), and the form of government (Persson & Tabellini, 2002, 2005).

2 It has often been argued that Confucianism has played a major role in the pro-work orientation of South Koreans (Hahm, 2003; Park & Kim, 1999).

3 Europe refers to Western European countries that have been known to spend large amounts of money on welfare. AA chose 14 countries for their Europe category: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Portugal, Spain, Sweden, and the United Kingdom. This study chose a slightly wider set of countries: European countries that also belong to Persson and Tabellini's definition of traditional OECD countries (Persson & Tabellini, 2005). Following this criterion, the Europe category in this study additionally includes Iceland, Luxembourg, Norway, and Switzerland. Note that switching to AA's set of 14 countries does not substantially change the results throughout this paper.

4 Note that there is a slight difference in computing a country-level belief between the left and right panels. In the former, AA use a mean value for a country, measured on a 10-point scale ranging from 1 = “in the long run, hard work usually brings a better life” to 10 = “hard work doesn’t generally bring success — it's more a matter of luck and connections.” AA then divide the mean value by 10 and interpret it as a proxy for the percentage of people who believe that luck determines income. This computation gives a range of 0.1–1 instead of 0–1, and thus, it should not be interpreted as a fraction. This is fixed in this study so that the representative value for each country is actually 0–1 in the latter panel.

5 A welfare queen is a derogative expression used to describe an individual who lives high on government assistance.

6 Educational attainment is classified according to the following categories: “high school drop-out,” “high school graduates,” “some college education,” and “college degree and above.”

7 Several robustness checks were conducted in this study. First, the estimations were repeated after replacing the linearity-assumed educational attainment and household income decile variables by their dummies. Second, the regressions were rerun with nonlinear terms of age and household income up to third-order polynomials. The estimated coefficients on the perception of work and laziness variables did not change.

8 The remaining categories in the welfare spending variable include family cash benefits, family services expenditures, unemployment compensation, housing allowances and rent subsidies, non-categorical cash benefits to low-income households, and non-categorical social services such as food subsidies, health support programs, and active labor market program expenditures. All of these are more or less related to non-working income.

9 Power resource theory emphasizes (1) class conflict in society as the direct effect of the labor power in the market and (2) partisan politics as the indirect effect through the state. Accordingly, empirical studies have employed union density and government partisanship as independent variables (Bradley, Huber, Moller, Nielsen, & Stephens, 2003). The cross-country regressions in this study control only for union density because government partisanship is likely an outcome of this study's main independent variable, the perceptions of work and laziness—voters who believe that people who do not work become lazy are more likely to vote for the rightist parties. Variables that are caused by the main independent variable of interest are bad controls that lead to bias. For the most intuitive explanation of this bias, see Angrist and Pischke (2008). In addition, note that the unemployment rate is not included in the models for the same reason—a society's perception that people who do not work become lazy is likely to create a social pressure to work.

10 Europe is chosen for comparison because of an obvious reason: it is the archetype of a region with large welfare.

11 According to Persson and Tabellini (2005), parliamentary regimes tend to spend more on welfare than do presidential regimes because of the confidence requirement. Parliamentary cabinets can remain in power only when they continuously enjoy the confidence of a majority in the legislature, and thus, they have strong incentives to control their party members. This then encourages stable majorities supporting the cabinet that tend to strive for the joint interest of their voters. As a result, the members’ spending decisions tend to be geared toward broad programs benefitting a majority of voters such as social transfers or public services. In contrast, legislators in a presidential system do not require such stable majorities, which then leads to the pursuit of the interests of different minorities against each other for different issues. Consequently, spending tends to be targeted toward powerful minorities.

12 Perhaps it is possible that social insurance benefits such as old-age pension, disability benefits, and survivors’ benefits would increase. These benefits are not related to the blame laid on people who do not work.