class: center, middle, inverse, title-slide .title[ # Far-right Electoral Success <br><br> Exacerbates Administrative Discrimination <br><br> Against Minorities ] .subtitle[ ## Evidence from a Field Experiment in Italy ] .author[ ### Krzysztof Krakowski, Dina Rosenberg, Alessio Romarri, and Merlin Schaeffer ] .date[ ### 2024-06-13 ] --- layout: true # Far-right Electoral Wins .font70[Increase Racism] .left-column[ <img src="https://image.cnbcfm.com/api/v1/image/107409158-17146029902024-05-01t222932z_1159614980_rc2xh7a5m73o_rtrmadp_0_usa-election-trump.jpeg?v=1714650301&w=1920&h=1080" width="80%" style="display: block; margin: auto;" /> <img src="https://pbs.twimg.com/media/FOH1FlhWUAIqbN1.jpg:large" width="80%" style="display: block; margin: auto;" /> ] --- .right-column[ <img src="./media/Bursztyn.png" width="80%" style="display: block; margin: auto;" /> .backgrnote[.center[*Source:* Bursztyn, Egorov, and Fiorin (2020)]] ] --- .right-column[ <img src="https://pbs.twimg.com/media/FZIn0ymXgAAOopA.jpg:large" width="80%" style="display: block; margin: auto;" /> ] --- .right-column[ <img src="./media/Romarri.png" width="75%" style="display: block; margin: auto;" /> .backgrnote[.center[*Source:* Romarri (2020)]] ] --- class: clear layout: false # Public Administrators .font70[Entrusted with Impartiality, but ...] .push-left[ <img src="./media/Olsen.png" width="70%" style="display: block; margin: auto;" /> .backgrnote[.center[*Source:* Olsen, Kyhse-Andersen, and Moynihan (2022)]] ] .push-right[ <img src="./media/Anselm.png" width="96%" style="display: block; margin: auto;" /> .backgrnote[.center[*Source:* Hemker and Rink (2017)]] ] --- class: inverse middle center # Are Public Administrators <br> also Susceptibile to the Influence of <br> Far-Right Electoral Successes? --- class: clear # Three Mechanisms .font70[(beyond scope the of this study)] .font120[ 1. **Recruitment**: Appointment of like-minded individuals to the administration, 2. **Cream Skimming**: Strategically allocate resources to force administrators to prioritize which citizens to serve (Guul, 2023; Bell and Jilke, 2024), 3. **Erosion of Norms**: Electoral success emboldens administrators to exercise discretion in ways that disadvantage minorities (Bursztyn, Egorov, and Fiorin, 2020). ] --- layout: false # A Field Experiment .push-left[ <img src="https://www.cos.io/hs-fs/hubfs/badges_stacked.original.png?width=834&name=badges_stacked.original.png" width="35%" style="display: block; margin: auto;" /> <img src="media/OSF.png" width="90%" style="display: block; margin: auto;" /> ] -- .push-right[ .center[.font110[Research assistants call 200 <br>Italian "Azienda Sanitaria Locale"<br>to ask for a general medical practitioner.]] .push-left[ <img src="media/Black_Female_Caller2.png" width="60%" style="display: block; margin: auto;" /> <img src="media/Black_Male_Caller.png" width="60%" style="display: block; margin: auto;" /> .center[.font90[Two students with <br> West-African accent]] ] .push-right[ <img src="media/White_Female_Caller.png" width="60%" style="display: block; margin: auto;" /> <img src="media/White_Male_Caller.png" width="60%" style="display: block; margin: auto;" /> .center[.font90[Two students with <br> Native-Italian accent]] ] ] --- # Interview Guideline <img src="media/healthcare_scenarios_ENG.png" width="100%" style="display: block; margin: auto;" /> --- class: clear # Discrimination .font70[in Access to Basic Healthcare] <img src="media/Figure_Present_1.png" width="100%" style="display: block; margin: auto;" /> .backgrnote[.center[ *Note:* Raw data and associated densities by caller accent alongside OLS estimates of their mean difference `\(\Delta\)` (95% CIs in brackets), covariate-adjusted for municipality fixed effects, gender of the caller, as well as date and time of the calls. n = 760 telephone calls (resp. 372 & 388) to 247 municipalities. ]] --- # Electoral Wins of Far-Right Mayors <img src="media/Map_Italy.png" width="80%" style="display: block; margin: auto;" /> .backgrnote[.center[*Source*: Romarri (2020)]] --- class: clear # Far-Right Electoral Wins .font70[and access to basic healthcare] <img src="Crete_files/figure-html/Figure2a-1.png" width="79%" style="display: block; margin: auto;" /> .backgrnote[.center[ *Note:* Denials alongside quadratic fits for n = 114 municipalities that lie within a "bandwidth" of `\(\pm 5\%\)` of the electoral threshold of a far-right victory ]] --- class: clear # Exacerbated Discrimination .font70[A Diff-in-RD design] .left-column[ <img src="Crete_files/figure-html/Figure2b-1.png" width="90%" style="display: block; margin: auto;" /> ] .right-column[ <img src="media/Figure_Present_2.png" width="100%" style="display: block; margin: auto;" /> .backgrnote[.center[ *Note:* Bias-corrected non-parametric regression-discontinuity estimates with optimized bandwidth (Calonico, Cattaneo, and Farrell, 2020) by caller accent alongside their `\(\Delta(\text{RD})\)` estimates (95% BCa CIs based on 5,000 bootstraps in brackets). Estimates are covariate-adjusted for municipalities' %foreign born, economic prosperity, logged population density, educational qualification of the population, altitude, as well as gender of the caller and date and time of the calls. Effect sizes are expressed in standard deviations of the respective outcome in the control group.]]] --- # Conclusion <br> .push-left[ .content-box-blue[.center[ Far-right electoral success threatens equal access to essential public services and thus potentially undermines public trust in government institutions. ]] .center[.font110[**Further research:**]] + **Test mechanisms** through which far-right politics influence discriminatory practices. + **Explore the generalizability** of these findings to other sectors and countries. + **Develop and evaluate interventions** to mitigate discrimination in public services under far-right leadership. ] .push-right[ .font50[ Bell, E. and S. Jilke (2024). "Racial Discrimination and Administrative Burden in Access to Public Services". In: _Scientific Reports_, p. 1071. Bursztyn, L., G. Egorov, and S. Fiorin (2020). "From Extreme to Mainstream: The Erosion of Social Norms". In: _American Economic Review_, pp. 3522-3548. Calonico, S., M. D. Cattaneo, and M. H. Farrell (2020). "Optimal bandwidth choice for robust bias-corrected inference in regression discontinuity designs". In: _The Econometrics Journal_, pp. 192-210. Guul, T. S. (2023). "Political Solutions to Discriminatory Behavior". In: _American Political Science Review_, pp. 719-733. Hemker, J. and A. Rink (2017). "Multiple Dimensions of Bureaucratic Discrimination: Evidence from German Welfare Offices". In: _American Journal of Political Science_, pp. 786-803. Olsen, A. L., J. H. Kyhse-Andersen, and D. Moynihan (2022). "The Unequal Distribution of Opportunity: A National Audit Study of Bureaucratic Discrimination in Primary School Access". In: _American Journal of Political Science_, pp. 587-603. Romarri, A. (2020). _Do Far-Right Mayors Increase the Probability of Hate Crimes? Evidence From Italy_. SSRN Scholarly Paper ID 3506811. Rochester, NY: Social Science Research Network. ]] --- class: clear inverse middle center # Appendix --- # Post Pre-reg data <img src="media/Figure_1_PostPrereg.png" width="95%" style="display: block; margin: auto;" /> --- # Online as "Success" or "Don't Know" <img src="media/Figure_1_Online.png" width="95%" style="display: block; margin: auto;" /> --- # Second round Elections <table style="NAborder-bottom: 0; color: black; width: auto !important; margin-left: auto; margin-right: auto; color: black; margin-left: auto; margin-right: auto;" class="table table"> <thead> <tr> <th style="empty-cells: hide;border-bottom:hidden;" colspan="1"></th> <th style="border-bottom:hidden;padding-bottom:0; padding-left:3px;padding-right:3px;text-align: center; " colspan="2"><div style="border-bottom: 1px solid #ddd; padding-bottom: 5px; ">Decline</div></th> <th style="border-bottom:hidden;padding-bottom:0; padding-left:3px;padding-right:3px;text-align: center; " colspan="2"><div style="border-bottom: 1px solid #ddd; padding-bottom: 5px; ">Questions</div></th> </tr> <tr> <th style="border-bottom:hidden;padding-bottom:0; padding-left:0px;padding-right:0px;text-align: center; font-weight: bold; " colspan="1"><div style="border-bottom: 1px solid #ddd; padding-bottom: 5px; ">Diff-in-RD:</div></th> <th style="border-bottom:hidden;padding-bottom:0; padding-left:0px;padding-right:0px;text-align: center; font-weight: bold; " colspan="2"><div style="border-bottom: 1px solid #ddd; padding-bottom: 5px; ">2.461</div></th> <th style="border-bottom:hidden;padding-bottom:0; padding-left:0px;padding-right:0px;text-align: center; font-weight: bold; " colspan="2"><div style="border-bottom: 1px solid #ddd; padding-bottom: 5px; ">-0.346</div></th> </tr> <tr> <th style="text-align:left;"> </th> <th style="text-align:center;"> Ital. </th> <th style="text-align:center;"> W. African </th> <th style="text-align:center;"> Ital. </th> <th style="text-align:center;"> W. African </th> </tr> </thead> <tbody> <tr> <td style="text-align:left;"> RD-Robust </td> <td style="text-align:center;"> 0.631* </td> <td style="text-align:center;"> 3.092* </td> <td style="text-align:center;"> 0.360 </td> <td style="text-align:center;"> 0.014 </td> </tr> <tr> <td style="text-align:left;box-shadow: 0px 1.5px"> </td> <td style="text-align:center;box-shadow: 0px 1.5px"> (0.345) </td> <td style="text-align:center;box-shadow: 0px 1.5px"> (1.601) </td> <td style="text-align:center;box-shadow: 0px 1.5px"> (0.443) </td> <td style="text-align:center;box-shadow: 0px 1.5px"> (0.444) </td> </tr> <tr> <td style="text-align:left;"> Bandwidth </td> <td style="text-align:center;"> 4.796 </td> <td style="text-align:center;"> 3.052 </td> <td style="text-align:center;"> 3.627 </td> <td style="text-align:center;"> 4.524 </td> </tr> <tr> <td style="text-align:left;"> Effective.cases </td> <td style="text-align:center;"> 43 vs 51 </td> <td style="text-align:center;"> 25 vs 44 </td> <td style="text-align:center;"> 33 vs 44 </td> <td style="text-align:center;"> 43 vs 57 </td> </tr> <tr> <td style="text-align:left;"> Overall.cases </td> <td style="text-align:center;"> 89 vs 113 </td> <td style="text-align:center;"> 104 vs 118 </td> <td style="text-align:center;"> 95 vs 118 </td> <td style="text-align:center;"> 107 vs 119 </td> </tr> </tbody> <tfoot><tr><td style="padding: 0; " colspan="100%"> <sup></sup> * p </td></tr></tfoot> </table> --- # Binary decline outcome .push-left[.font80[ `$$\text{Binary version 1, "(Probably) Yes"}= \begin{cases} 1, & 1 \text{ or } 2 \\ 0 & 3 \text{ to } 5 \end{cases}$$` <table class="table table" style="color: black; width: auto !important; margin-left: auto; margin-right: auto; color: black; margin-left: auto; margin-right: auto;"> <thead> <tr> <th style="text-align:left;"> (Probably) Yes </th> <th style="text-align:left;"> </th> <th style="text-align:right;"> Native-Italian accent </th> <th style="text-align:right;"> West-African accent </th> </tr> </thead> <tbody> <tr> <td style="text-align:left;"> 0 </td> <td style="text-align:left;"> N </td> <td style="text-align:right;"> 1 </td> <td style="text-align:right;"> 50 </td> </tr> <tr> <td style="text-align:left;"> </td> <td style="text-align:left;"> % col </td> <td style="text-align:right;"> 0.3 </td> <td style="text-align:right;"> 12.3 </td> </tr> <tr> <td style="text-align:left;"> 1 </td> <td style="text-align:left;"> N </td> <td style="text-align:right;"> 390 </td> <td style="text-align:right;"> 356 </td> </tr> <tr> <td style="text-align:left;"> </td> <td style="text-align:left;"> % col </td> <td style="text-align:right;"> 99.7 </td> <td style="text-align:right;"> 87.7 </td> </tr> </tbody> </table> ]] .push-right[.font80[ `$$\text{Binary version 2, "(Probably) No"}= \begin{cases} 1, & 5 \text{ or } 4 \\ 0 & 1 \text{ to } 3 \end{cases}$$` <table class="table table" style="color: black; width: auto !important; margin-left: auto; margin-right: auto; color: black; margin-left: auto; margin-right: auto;"> <thead> <tr> <th style="text-align:left;"> (Probably) No </th> <th style="text-align:left;"> </th> <th style="text-align:right;"> Native-Italian accent </th> <th style="text-align:right;"> West-African accent </th> </tr> </thead> <tbody> <tr> <td style="text-align:left;"> 0 </td> <td style="text-align:left;"> N </td> <td style="text-align:right;"> 391 </td> <td style="text-align:right;"> 394 </td> </tr> <tr> <td style="text-align:left;"> </td> <td style="text-align:left;"> % col </td> <td style="text-align:right;"> 100.0 </td> <td style="text-align:right;"> 97.0 </td> </tr> <tr> <td style="text-align:left;"> 1 </td> <td style="text-align:left;"> N </td> <td style="text-align:right;"> 0 </td> <td style="text-align:right;"> 12 </td> </tr> <tr> <td style="text-align:left;"> </td> <td style="text-align:left;"> % col </td> <td style="text-align:right;"> 0.0 </td> <td style="text-align:right;"> 3.0 </td> </tr> </tbody> </table> ]] --- # Binary decline outcome .font70[Discrimination] <table style="NAborder-bottom: 0; color: black; width: auto !important; margin-left: auto; margin-right: auto;" class="table"> <thead> <tr> <th style="text-align:left;"> </th> <th style="text-align:center;"> (Probably) Yes </th> <th style="text-align:center;"> (Probably) No </th> </tr> </thead> <tbody> <tr> <td style="text-align:left;"> West-African Accent </td> <td style="text-align:center;"> −0.096*** </td> <td style="text-align:center;"> 0.023** </td> </tr> <tr> <td style="text-align:left;"> </td> <td style="text-align:center;"> (0.020) </td> <td style="text-align:center;"> (0.009) </td> </tr> <tr> <td style="text-align:left;"> Gender: Male </td> <td style="text-align:center;"> 0.042* </td> <td style="text-align:center;"> −0.001 </td> </tr> <tr> <td style="text-align:left;box-shadow: 0px 1.5px"> </td> <td style="text-align:center;box-shadow: 0px 1.5px"> (0.018) </td> <td style="text-align:center;box-shadow: 0px 1.5px"> (0.009) </td> </tr> <tr> <td style="text-align:left;"> Num.Obs. </td> <td style="text-align:center;"> 760 </td> <td style="text-align:center;"> 760 </td> </tr> <tr> <td style="text-align:left;"> R2 </td> <td style="text-align:center;"> 0.407 </td> <td style="text-align:center;"> 0.471 </td> </tr> <tr> <td style="text-align:left;"> BIC </td> <td style="text-align:center;"> −327.6 </td> <td style="text-align:center;"> −1524.3 </td> </tr> </tbody> <tfoot><tr><td style="padding: 0; " colspan="100%"> <sup></sup> + p </td></tr></tfoot> </table> --- # Binary decline outcome `\(\Delta_{\text{RD}}\)` <table style="NAborder-bottom: 0; color: black; width: auto !important; margin-left: auto; margin-right: auto; color: black; margin-left: auto; margin-right: auto;" class="table table"> <thead> <tr> <th style="empty-cells: hide;border-bottom:hidden;" colspan="1"></th> <th style="border-bottom:hidden;padding-bottom:0; padding-left:3px;padding-right:3px;text-align: center; " colspan="2"><div style="border-bottom: 1px solid #ddd; padding-bottom: 5px; ">(Probably) Yes</div></th> <th style="border-bottom:hidden;padding-bottom:0; padding-left:3px;padding-right:3px;text-align: center; " colspan="1"><div style="border-bottom: 1px solid #ddd; padding-bottom: 5px; ">(Probably) No</div></th> </tr> <tr> <th style="border-bottom:hidden;padding-bottom:0; padding-left:0px;padding-right:0px;text-align: center; font-weight: bold; " colspan="1"><div style="border-bottom: 1px solid #ddd; padding-bottom: 5px; ">Diff-in-RD:</div></th> <th style="border-bottom:hidden;padding-bottom:0; padding-left:0px;padding-right:0px;text-align: center; font-weight: bold; " colspan="2"><div style="border-bottom: 1px solid #ddd; padding-bottom: 5px; ">-0.129</div></th> <th style="empty-cells: hide;border-bottom:hidden;" colspan="1"></th> </tr> <tr> <th style="text-align:left;"> </th> <th style="text-align:center;"> Ital. </th> <th style="text-align:center;"> W. African </th> <th style="text-align:center;"> W. African </th> </tr> </thead> <tbody> <tr> <td style="text-align:left;"> RD-Robust </td> <td style="text-align:center;"> 0.002 </td> <td style="text-align:center;"> −0.127* </td> <td style="text-align:center;"> 0.060 </td> </tr> <tr> <td style="text-align:left;box-shadow: 0px 1.5px"> </td> <td style="text-align:center;box-shadow: 0px 1.5px"> (0.001) </td> <td style="text-align:center;box-shadow: 0px 1.5px"> (0.073) </td> <td style="text-align:center;box-shadow: 0px 1.5px"> (0.037) </td> </tr> <tr> <td style="text-align:left;"> Bandwidth </td> <td style="text-align:center;"> 5.521 </td> <td style="text-align:center;"> 3.358 </td> <td style="text-align:center;"> 2.841 </td> </tr> <tr> <td style="text-align:left;"> Effective.cases </td> <td style="text-align:center;"> 55 vs 135 </td> <td style="text-align:center;"> 32 vs 88 </td> <td style="text-align:center;"> 28 vs 81 </td> </tr> <tr> <td style="text-align:left;"> Overall.cases </td> <td style="text-align:center;"> 110 vs 262 </td> <td style="text-align:center;"> 126 vs 262 </td> <td style="text-align:center;"> 126 vs 262 </td> </tr> </tbody> <tfoot><tr><td style="padding: 0; " colspan="100%"> <sup></sup> * p </td></tr></tfoot> </table> --- # The Controls ``` r # PCA to reduce dimensions of covariates and prohibit multicollinearity ####################################################################### ## Economy (pca_econ <- cntrls %>% prcomp( formula = ~ sh_unemployed_2011 + sh_agriculture_epl_2011 + sh_industry_epl_2011 + sh_commerce_epl_2011 + sh_service_epl_2011 + sh_cars_2017, data = ., na.action = na.exclude, scale = TRUE)) cntrls$pca_econ <- -pca_econ$x[ , "PC1"] %>% scale() %>% as.numeric() ## Educational qualification of pop (pca_edu <- cntrls %>% prcomp( formula = ~ sh_illiterate_2011 + sh_sec_education_2011 + sh_students_2011 + sh_university_enrolled_2016, data = ., na.action = na.exclude, scale = TRUE)) cntrls$pca_edu <- -pca_edu$x[ , "PC1"] %>% scale() %>% as.numeric() ## Combine with further controls cntrls <- cntrls %>% mutate( sh_foreing_16 = sh_foreing_16 %>% scale() %>% as.numeric(), # Share foreign Born in 2016 pop2011 = (pop2011 / Superficieterritorialekmqal) %>% log() %>% scale() %>% as.numeric(), # logged population density Altitudinedelcentrometri = Altitudinedelcentrometri %>% scale() %>% as.numeric()) # Altitude ``` --- class: clear # Balanced Confounders at the Threshold <table style="NAborder-bottom: 0; color: black; width: auto !important; margin-left: auto; margin-right: auto; font-size: 18px; color: black; margin-left: auto; margin-right: auto;" class="table table"> <thead> <tr> <th style="text-align:left;"> </th> <th style="text-align:center;"> % Foreign born </th> <th style="text-align:center;"> Population </th> <th style="text-align:center;"> Altitude </th> <th style="text-align:center;"> Economy </th> <th style="text-align:center;"> Educational distribution </th> </tr> </thead> <tbody> <tr> <td style="text-align:left;"> RD-Robust </td> <td style="text-align:center;"> 0.002 </td> <td style="text-align:center;"> −0.155 </td> <td style="text-align:center;"> 0.304 </td> <td style="text-align:center;"> −0.710 </td> <td style="text-align:center;"> −0.809 </td> </tr> <tr> <td style="text-align:left;box-shadow: 0px 1.5px"> </td> <td style="text-align:center;box-shadow: 0px 1.5px"> (0.568) </td> <td style="text-align:center;box-shadow: 0px 1.5px"> (0.422) </td> <td style="text-align:center;box-shadow: 0px 1.5px"> (0.435) </td> <td style="text-align:center;box-shadow: 0px 1.5px"> (0.407) </td> <td style="text-align:center;box-shadow: 0px 1.5px"> (0.475) </td> </tr> <tr> <td style="text-align:left;"> Bandwidth </td> <td style="text-align:center;"> 4.057 </td> <td style="text-align:center;"> 3.806 </td> <td style="text-align:center;"> 6.374 </td> <td style="text-align:center;"> 2.846 </td> <td style="text-align:center;"> 5.442 </td> </tr> <tr> <td style="text-align:left;"> Effective.cases </td> <td style="text-align:center;"> 28 vs 66 </td> <td style="text-align:center;"> 28 vs 63 </td> <td style="text-align:center;"> 43 vs 100 </td> <td style="text-align:center;"> 19 vs 54 </td> <td style="text-align:center;"> 38 vs 94 </td> </tr> <tr> <td style="text-align:left;"> Overall.cases </td> <td style="text-align:center;"> 80 vs 186 </td> <td style="text-align:center;"> 80 vs 186 </td> <td style="text-align:center;"> 80 vs 186 </td> <td style="text-align:center;"> 80 vs 186 </td> <td style="text-align:center;"> 80 vs 186 </td> </tr> </tbody> <tfoot><tr><td style="padding: 0; " colspan="100%"> <sup></sup> * p </td></tr></tfoot> </table> .backgrnote[.center[*Note:* Results are based on non-linear sharp RD estimates with triangular kernel, bandwidth selected by mserd, cluster-robust bias-corrected inference asu suggested by Calonico, Cattaneo, and Farrell (2020).]]