Скачать презентацию THE LOSERS PARADOX A LEGISLATOR SURVEY EXPERIMENT ON Скачать презентацию THE LOSERS PARADOX A LEGISLATOR SURVEY EXPERIMENT ON

275e22d62a923387db19e4627ab9821a.ppt

  • Количество слайдов: 16

THE LOSERS’ PARADOX A LEGISLATOR SURVEY EXPERIMENT ON WHY LOSERS ULTIMATELY WIN PROTECTION Megumi THE LOSERS’ PARADOX A LEGISLATOR SURVEY EXPERIMENT ON WHY LOSERS ULTIMATELY WIN PROTECTION Megumi Naoi, Assistant Professor, Dept. of Political Science University of California, San Diego

LEGISLATORS IN THE GLOBAL ECONOMY Legislators’ Position-taking on Globalization • The search for “electoral LEGISLATORS IN THE GLOBAL ECONOMY Legislators’ Position-taking on Globalization • The search for “electoral connection” (Mayhew 1974): How “constituents’ interests” shape legislators’ policy-positions. • Used district-level data and tested H=O, R=V, and the role of ideology. But legislators are made to do two things: • Constituency service • Law-making aspect, especially how parties build the supra-majority coalition to legislate globalization bills, is generally missing (exception: Gilligan). Problematic, because: (i) Many policy changes relating to globalization, if not all, have to be legislated, (ii) “electoral disconnection” is prevalent.

LEGISLATORS’ POSITION-TAKING IN THE WORLD OF TRADE THEOREMS Point a divides Yea vs. Nay LEGISLATORS’ POSITION-TAKING IN THE WORLD OF TRADE THEOREMS Point a divides Yea vs. Nay Votes: Electoral Connection

LEGISLATORS’ POSITION-TAKING IN THE CONTEXT OF LAW-MAKING Electoral disconnection occurs in districts b/w a LEGISLATORS’ POSITION-TAKING IN THE CONTEXT OF LAW-MAKING Electoral disconnection occurs in districts b/w a and b (supra-majority requirement)

VARIOUS SOURCES OF “ELECTORAL DISCONNECTION” Incentive-based Arguments Party discipline Side-payment from party leaders to VARIOUS SOURCES OF “ELECTORAL DISCONNECTION” Incentive-based Arguments Party discipline Side-payment from party leaders to back-benchers Sectoral & issue-linkages Political Institutions: Party systems, electoral systems and electoral competition (e. g. , Duverger’s law) Ideology and Psychology-based Arguments • • • Legislators’ ideology & polarization (that deviate from the median-voters) • Sympathy & projection (Naoi & Kume, IO, 2011) • This Paper: Framing & Legitimization & Rationalization

THIS PAPER The Goal: (i) Describe how different policy rationales mobilize coalitions among legislators THIS PAPER The Goal: (i) Describe how different policy rationales mobilize coalitions among legislators and (ii) identify the sources of this shift. Research Design: Survey experiment on legislators. Holding the policy’s distributional consequences constant, randomize different policy rationales and solicit legislators’ support for the policy. Subjects: 478 Japanese Lower-House Legislators before 2009 election.

PRE-ELECTION, LEGISLATOR SURVEY EXPERIMENT (2009) Block-randomized LH Legislators by: political parties, seniority, urban-rural district, PRE-ELECTION, LEGISLATOR SURVEY EXPERIMENT (2009) Block-randomized LH Legislators by: political parties, seniority, urban-rural district, electoral systems (SMD & PR), and nonresponses for 2005 post-election survey by Asahi Newspaper & The University of Tokyo. Randomly assigned three versions of the questionnaires to all lower house incumbents via mail survey. Approximated the actual position-taking in the campaign: Told in the cover letter that we will publish their responses on the website of “The Election” (Za-Senkyo), a popular website for political news (and we actually did).

SURVEY INSTRUMENTS: SAME POLICY, DIFFERENT RATIONALES Control Group (No Framing) • Q: Food imports SURVEY INSTRUMENTS: SAME POLICY, DIFFERENT RATIONALES Control Group (No Framing) • Q: Food imports from abroad have been increasing in recent years. What do you think about the opinion that we should limit food imports from abroad? Treatment 1 (Producer/Private Goods Framing) • Q: Food imports from abroad have been increasing in recent years. What do you think about the opinion that we should limit food imports from abroad in order to protect farmers’ jobs and income? Treatment 2 (Consumer/Public Goods Framing) • Q: Food imports from abroad have been increasing in recent years. What do you think about the opinion that we should limit food imports from abroad in order to protect consumer safety?

26% RESPONSE RATE & (ALMOST) BALANCED GROUPS Farmers’ Jobs Consum Safety Control Total Responses 26% RESPONSE RATE & (ALMOST) BALANCED GROUPS Farmers’ Jobs Consum Safety Control Total Responses 44 41 41 LDP 51% (50. 5) 41% (49. 9) 50% (50. 6) DPJ 38% (49. 0) 32% (47. 1) 36% (48. 5) Senior 31% (46. 8) 29% (46. 1) 33% (47. 7) Freshmen/women 24% (43. 6) 9. 5% (30. 1) 8% (35. 4) Urban 52% (50. 9) 60% (50. 3) 56% (50. 6) High Unemploym 60%(49. 5) 46. 3%(48. 1) 42. 9%(50. 1) Competitive Dist. 51% (50. 6) 68. 6% (47. 1) 55% (50. 4) Standard deviations in parentheses. Balance tests: Difference-in-means tests & Cochran (1968)’s rule of thumb (a mean difference should not differ more than a quarter of a standard deviation).

RESULTS IN AGGREGATE: % PROTECTIONIST RESPONSES % Protectionist Responses Consumer-safety framing mobilizes additional support RESULTS IN AGGREGATE: % PROTECTIONIST RESPONSES % Protectionist Responses Consumer-safety framing mobilizes additional support for protectionism by 23 percentage points from the control group (significant at 5%). Farmer’s income/employment framing has no systematic effects.

BUT SUB-GROUP ANALYSIS REVEALS: URBAN REPRESENTATIVES ARE SYMPATHETIC TO PROTECTING FARMERS’ JOBS & INCOME BUT SUB-GROUP ANALYSIS REVEALS: URBAN REPRESENTATIVES ARE SYMPATHETIC TO PROTECTING FARMERS’ JOBS & INCOME Framing Effects on Rural vs. Urban Representatives ( % Protectionist Responses) Farmers’ Jobs Control Consumer Safety Rural Districts 45. 5% 50% Urban Districts 46. 2%** 7. 7% 66. 7%** Note: ** significant at 5% level. Rural Representatives: Rationales don’t matter; hard core protectionists. Urban Representatives: Rationales matter a lot. Farmers’ income framing increased the support for protectionism by 33 percentage points to the level of rural representatives in the control group. Producer framing (protecting farmer’s jobs and income) can make urban representatives think like rural representatives.

RESULTS OF PROBIT ANALYSES Dependent Variable: Protectionist Responses (Y=|PR: 0, 1|) Model 1 Model RESULTS OF PROBIT ANALYSES Dependent Variable: Protectionist Responses (Y=|PR: 0, 1|) Model 1 Model 2 Model 3 Model 4 Farmers’ Jobs 0. 57(1. 51) -0. 03(0. 05) -0. 47(0. 78) -0. 37(0. 49) Consum Safe 0. 98(2. 49)** 0. 09(0. 15) 0. 19(0. 31) 0. 29(0. 42) Urban -1. 33(2. 10)** -1. 82(2. 54)** -1. 89(2. 28)** Farm*Urban 1. 37(1. 67)* 1. 55(1. 72)* 1. 68(1. 68)* Consu*Urban 1. 79(2. 07)** 1. 78(1. 94)* 1. 74(1. 64)* High Unemp 1. 16(1. 49) 1. 31(1. 59) Farm*Unemp -0. 31(0. 32) -0. 37(0. 37) Cons*Unemp 0. 01(1. 01) 0. 002(0. 00) LDP, Senior, Freshm/w LDP, Senior, Fresh, Competitive Controls LDP Urban representatives increased the support with the farmer’s jobs & income framing (i. e. , private goods framing)

WHY URBAN LEGISLATORS CARE ABOUT “FARMERS’ JOBS & INCOME”? Four Possibilities (1) Opinion Responsiveness WHY URBAN LEGISLATORS CARE ABOUT “FARMERS’ JOBS & INCOME”? Four Possibilities (1) Opinion Responsiveness (Urban voters are also sympathetic) Matched survey after election. (1) Party discipline Yes, to some extent. Legislators who are more dependent on party nomination (freshmen/women and PR list) increased support w/ farmer-framing. (3) Coalition of losers (Naoi & Kume 2011): high unemployment support for farmers Rejected. See next table. (4) Social desirability bias in competitive district? Rejected; desirability bias is stronger in uncompetitive districts, and only exists for “consumer safety framing”.

(2) Party discipline (Dependence on Party Nomination) Yes Producer Control Consumer Freshmen/women & PR (2) Party discipline (Dependence on Party Nomination) Yes Producer Control Consumer Freshmen/women & PR 50%* 12. 5% 53. 8%* Mid & Senior & SMD 34. 6% 39. 4% 57. 1% (3) Coalition of losers (Unemployment rate) Producer Control Consumer Low Unemployment 29. 4% 37. 5% 54. 5% High Unemployment 29. 4% 48. 1% 57. 9%* (4) Social desirability bias (Competitiveness) Producer Control Consumer Competitive districts 40% 45. 5% 50% Uncompetitive districts 38. 8% 23. 5% 72. 7%***

DISCUSSION & NEXT STEPS (1) Different policy rationales can mobilize diverse legislators’ coalitions for DISCUSSION & NEXT STEPS (1) Different policy rationales can mobilize diverse legislators’ coalitions for protectionism, even with a policy with identical distributional consequence. (2) Framing was most effective to change minds of swing legislators, i. e. , urban representatives for the issue of agricultural protectionism. (3) Public goods framing > private goods framing. Yet, private goods framing also had substantive effects on urban legislators. (4) Sources of this shift: urban & freshmen/women & PR list. Story about “swing” legislators, party discipline, or, lack of experience?

REPRESENTATIVENESS OF LEGISLATOR SURVEY: PROBIT DV: 1: Responded, 0: No response. REPRESENTATIVENESS OF LEGISLATOR SURVEY: PROBIT DV: 1: Responded, 0: No response.