Representation by Elites

Week 13 POL837 Issues in Comparative Politics

Daniel Sánchez Pazmiño

Simon Fraser University

April 2024

Key terms

  • Political elite

  • Democratic representation

  • Asymmetric polarization

  • Motivated reasoning

  • Bias

Representation by elites

  • Classic theory posits that in representative democracies, political elites propose policy which is popular among citizens (Broockman & Butler, 2017; Downs, 1957).

  • In a nutshell, the readings investigate how well do elites represent the public in various settings.

  • Broockman & Skovron (2018) presents a fundamental failure of representation by political elites, who cannot accurately gauge public opinion on policy issues.

  • On the other hand, the other articles argue that elites can represent the public well but only with the right incentives.

  • Presenting scientific evidence to elites can change their policy preferences to align with the scientific consensus (N. Lee, 2022).

  • Increasing the number of a minority group in the legislature (Wang, 2023) or the contact with the minority group can heighten the representation of that group’s interests (Washington, 2008).

Bias in Perceptions of Public Opinion among Political Elites - in the American Political Science Review

  • Broockman & Skovron (2018) investigate whether political elites in the United States accurately predict public support for liberal or conservative policy.

  • They find that both Democratic and Republican elites overestimate public support for conservative policies, but Republicans overestimate it more.

  • The most overestimated policy position is gun control. Democratic elites estimate public support at around 50%, Republicans at around 40%, while true support is around 80%.

  • The mechanism behind the bias is that information environments of politicians are not representative of constituents.

    • Both Democratic and Republican elites disproportionately hear from conservatives, who are more likely to contact their representatives.

Broockman & Skovron (2018): Discussion

  • Few candidates report conducting polls to gauge public opinion, and those who do are not more accurate in their public opinion estimates. Why?

  • The silent majority concept is often brought up to justify how extremist can win elections, particularly RRP candidates. Do these findings support this view?

  • The authors repeatedly mention that these biases are pervasive. Under what assumptions would these biases be pervasive for democratic representation? How do these biases would affect policy outcomes, voting behaviour and turnout, if at all?

Do Policy Makers Listen to Experts? Evidence from a National Survey of Local and State Policy Makers - in the American Political Science Review

  • N. Lee (2022) provides U.S. political elites with expert information on policy issues to examine the extent to which they are willing to update their policy preferences in response to facts.

  • It is found that exposing elites to short messages which summarize scientific evidence on policy issues greatly affects their beliefs and policy preferences, aligning them with the scientific consensus.

  • Findings apply to both Republic and Democratic elites, despite the theoretical expectation that Republicans would be less likely to update their beliefs.

N. Lee (2022): Discussion

  • Are the findings by Broockman & Skovron (2018) consistent with N. Lee (2022)’s findings? If not, what potential differences exist that can reconcile the two studies?

  • The author states that not providing additional information about scientific evidence to the political elites biases against the expert information. Is this a reasonable assumption? How does this extend to cases where expert consensus is not as clear?

  • Can agreeing with scientific evidence be seen as socially desirable? How does this affect the interpretation of the results? Would it depend on the issue or partisan alignment?

Female Socialization: How Daughters Affect Their Legislator Fathers’ Voting on Women’s Issues - in the American Economic Review

  • Washington (2008) investigates whether having a female child affects the voting behavior of legislators on women’s issues.

  • Conditional on number of children, each daughter increases the congressperson’s propensity to vote liberally, particularly on reproductive rights issues.

Washington (2008): Discussion

  • Much of the attitudinal shift is from Democratic legislators who had daughters. Relating to Broockman & Skovron (2018), what reasons could explain the lack of support for female-friendly policy among Democratic legislators before they had daughters?

  • The paper’s identifying assumption relies on the inexistence of fertility stopping rules. Washington (2008) does not find evidence of such rules, but if there were, how would that affect the results?

  • Also relating to Broockman & Skovron (2018), would increasing the participation of women in legislator’s “feedback circles” have a similar effect on legislators’ voting behavior, or is a family connection the most important factor in changing attitudes?

Washington (2008): Fertility stopping rules

  • Two couple types: one that has children until having a male child (gender-stopping rule) and another that has children until having the desired number of children (number-stopping rule).

  • The number-stopping rule causes correlation between the number of daughters and the number of children, which could bias the results. Easily fixed by controlling for the number of children.

  • The gender-stopping rule would create a correlation between the (unobserved) gender preference and the number of daughters/sons. Cannot control for this: cannot possibly difference between couple types if gender preference is unobserved.

  • Gender preference likely is related with legislator voting behaviour, and also with the number of children, hence the estimate of the effect of daughters on voting behaviour would be biased.

Do Women Always Represent Women? The Effects of Gender Quotas on Substantive Representation - In Political Behavior

  • Wang (2023) examines the impact of gender quotas on the Taiwanese legislature’s policymaking, focusing on how the quota impacts women differently depending on how they are elected in terms of the social policy is initiated and passed.

  • The gender quota established in the 2008 Taiwanese legislative election increased the share of social policies initiated by female legislators relative to male legislators in the quota tier. It also increased the share of policies which were passed in the quota tier.

  • The quota did not increase the share of social policies initiated by women in the non-quota tier relative to men in the non-quota tier. There was no evidence of a spillover effect to male legislators in either tier.

Wang (2023): Discussion

  • The theoretical mechanism proposed by the author argue that women in quota-elected seats may feel more obliged to pursue policy oriented to women. Does using welfare, education, and health as dependent variables make sense in this context?

  • The author implicitly defines quota-elected female politicians as those women elected in the quota tier. ¿Is it adequate to believe the quota caused to the election of these elites?

  • Washington (2008) is statistically underpowered to find an effect of daughters on mother legislators. With Wang’s (2023) findings in mind, what would be a predictable result of the effect of daughters on mothers?

Activity: Article Criticisms

  • Think about a criticism you have for any of the week’s readings.

  • Frame it in 10 words or less. Include the author and year.

    • E.g. Washington 2008: Too much economics.

  • Go to menti.com and enter the code to write down the criticism.

  • You will have 3 votes to upvote other criticisms.

Criticisms

  • Broockman & Skovron (2018)
    • More discussion is needed on why these biases are necessarily pervasive for democratic representation. Are they large enough to actually influence the policy positions of politicians?
    • Strange wording and terms in some cases (“dyadic”, “intercept shift”)
  • N. Lee (2022)
    • The study does not consider the possibility of social desirability bias in the responses of the political elites.
    • Rent control may have a lesser degree of scientific consensus than the other issues.
    • More discussion should be considered for the implications of politicized issues.
  • Washington (2008)
    • The most important identifying assumption, the lack of gender-biased fertility stopping rules, is poorly explained.
  • Wang (2023)
    • The relationship between social policies and women issues is unclear: welfare, education and health are not women-specific issues, these are much broader, yet the author treats them as such.
    • Institutional context is unclear: a separate section would’ve been helpful to better understand it and thus define clear-cut control and treatment groups for the dif-in-dif.
    • It is impossible to identify women who were elected because of the quota, since observed women legislators could have been elected on a counterfactual world without the quota anyway. Wording should be changed to reflect this.
    • Parallel trends are poorly justified - an event study regression plot should be presented, such as in B. Lee (2023).
    • Unclear why the IV design is needed and how it works in the context of the paper: what is the instrument and how does the exclusion restriction hold?

Thank you! Good luck in your exams!

References

Broockman, D. E., & Butler, D. M. (2017). The causal effects of elite position-taking on voter attitudes: Field experiments with elite communication. American Journal of Political Science, 61(1), 208–221. https://doi.org/10.1111/ajps.12243
Broockman, D. E., & Skovron, C. (2018). Bias in Perceptions of Public Opinion among Political Elites. American Political Science Review, 112(3), 542–563. https://doi.org/10.1017/S0003055418000011
Downs, A. (1957). An economic theory of democracy (First Edition). Harper and Row.
Lee, B. (2023). Educational content, exclusive national identity, and anti-immigrant attitudes. The Journal of Politics, 85(4), 1182–1197.
Lee, N. (2022). Do Policy Makers Listen to Experts? Evidence from a National Survey of Local and State Policy Makers. American Political Science Review, 116(2), 677–688. https://doi.org/10.1017/S0003055421000800
Wang, S. (2023). Do Women Always Represent Women? The Effects of Gender Quotas on Substantive Representation. Political Behavior, 45(4), 1979–1999. https://doi.org/10.1007/s11109-022-09808-z
Washington, E. L. (2008). Female Socialization: How Daughters Affect Their Legislator Fathers. American Economic Review, 98(1), 311–332. https://doi.org/10.1257/aer.98.1.311