Bootstraps, Buddies, and Bribes: Perceived Meritocracy in the United States and China

ABSTRACT

We examine the extent to which Americans and Chinese perceive their countries as meritocracies. Chinese and Americans share a strong belief in the importance of education, ambition, and hard work for getting ahead. The Chinese, however, also believe strongly in nonmeritocratic elements, especially social ties. They are thus much more likely than Americans to have a dual-consciousness about getting ahead. Perceptions also vary within each country. In the United States, minorities perceive less meritocracy than whites and men perceive less meritocracy than women. These factors, however, do not explain variations in Chinese perceptions, which are more closely tied to China-specific predictors (e.g., hukou status). We also find that Chinese perceptions reflect the degree of contact with nonrelatives and people in occupations more prestigious than one’s own. Together, these findings suggest that American beliefs about the role of merit in social mobility are not unique. They also highlight a need for studies of stratification beliefs in nondemocratic, non-Western countries and research that examines how life experiences influence stratification beliefs.

KEYWORDS:

• China
• CGSS
• GSS
• ISSP
• meritocracy
• mobility
• stratification beliefs

Notes

1. Other scholars have examined the extent to which people want their society to be meritocratic and/or what they think it might take to achieve that goal, but we do not examine those issues in this article.

2. Research on beliefs about the causes of wealth and poverty are most closely related to our work, but even they differ from ours in two important respects. First, people may have different beliefs about meritocracy (i.e., what it takes to get ahead) than they do about the more extreme outcomes of wealth and poverty. Second, those studies examine whether people think wealth and poverty are the result of individual or structural elements, and that approach does not map neatly onto the distinction between meritocratic and nonmeritocratic elements. Some individual factors are meritocratic (e.g., hard work), but others are not (e.g., a willingness to cheat). Likewise, although structural explanations are often nonmeritocratic (e.g., racism), some promote meritocracy (e.g., antidiscrimination laws).

3. There are also many studies that rely on the logic of self-interest to explain stratification beliefs (Bolzendahl and Myers 2004), but that makes most sense when examining how people want the stratification system to work. Poor respondents, for instance, may be especially likely to favor redistribution of income because they would benefit. Focusing on exposure, in contrast, is more common and logical in work like ours that examines perceptions of how the stratification system actually works. Previous studies have not always distinguished clearly between these two types of beliefs, but doing so promotes theoretical clarity (Duru-Bellat and Tenret 2012; Marshall et al. 1999).

4. The “instruction hypothesis” suggests that education will have the opposite effect: it will weaken perceived meritocracy because education increases awareness of inequality and the factors that cause it. While this hypothesis is supported in some countries, it is not supported in the United States (Duru-Bellat and Tenret 2012).

5. For details about the ISSP, see https://dbk.gesis.org/dbksearch/sdesc2.asp?no=5400. For information about the CGSS, see http://www.chinagss.org/index.php?r=index/index&hl=en. For information about the GSS, see http://gss.norc.org/.

6. Formal tests show that the standard deviation in perceptions of nonmeritocratic elements is similar in the two countries, but the standard deviation in perceptions of meritocratic elements is larger in China.

7. A closer examination shows that women place less weight than men on coming from a wealthy family and race. Also, the coefficient for female is only significant at the .05 level after controlling for education. This is because women in the sample have less education than men and because education is associated with greater perceived meritocracy. Controlling for that negative indirect connection reveals the positive effect of being female.

8. We tried adding age squared to Models 1 and 4, but it was not statistically significant.

9. An insightful reviewer wondered whether the Chinese results were weak because Chinese respondents avoid extreme responses like “essential” and choose more moderate responses. To test this idea, we dichotomized the items in our scales to indicate when Chinese respondents chose above the median. We then used the count of such answers to create percentile scores like the original dependent variables. Rather than drawing out subtle patterns, this change led to fewer significant coefficients and a drop in the R2. We also compared the standard deviations for all the items in our scales across the two countries and found that they are larger in China for 8 of the 12 items. Both of these supplemental analyses suggest that the weak Chinese results are not an artifact of a bias toward moderate responses among the Chinese.

10. We also tried two alternate ways of coding the Communist Youth League respondents. First, we tried grouping them with Communist Party members. Second, we tried coding them with respondents who are not Communist Party members. Neither approach produced statistically significant results.

11. To check the robustness of the results in Table 4, we tried alternate coding schemes for several independent variables. First, we replaced the minority variable with two variables to distinguish between minorities that tend to speak Mandarin and those who do not. The variable for non-Mandarin speakers showed the same pattern of significance as in Table 4. The other variable was not significant. These results cast doubt on the idea that non–Mandarin-speaking minorities in China would have low perceived meritocracy like minorities in the United States. Given the similar implications of the results, we retained the original coding, which does not require us to make assumptions about language usage based on ethnicity. This also maintains the correspondence between Tables 3 and 4. We also tried a variable identifying people employed by state-owned enterprises, but it was not significant.

Additional information

Notes on contributors

He Xian

He Xian is a lawyer at Butzel Long, P.C. and a graduate of the Michigan State University College of Law. His research examines perceptions of stratification in the United States and abroad. His current projects focus on the way such perceptions are shaped by tax, welfare, and immigration policy.

Jeremy Reynolds

Jeremy Reynolds is Professor of Sociology at Purdue University. His work focuses on work-family issues and subjective aspects of inequality. He is especially interested in preferences regarding paid work schedules and unequal access to schedules that allow people to meet both financial and caregiving needs. His work has been published in American Sociological Review, Social Forces, Work & Occupations, ILR Review, Journal of Marriage & Family, and Journal of Family Issues. He is a former winner of the Rosabeth Moss Kanter Award for research on work-family issues.