The deterrent effect of an anti-minaret vote on foreigners’ location choices

Abstract

In a national ballot in 2009, Swiss citizens surprisingly approved an amendment to the Swiss constitution to ban the further construction of minarets. The ballot outcome manifested reservations and anti-immigrant attitudes in regions of Switzerland which had previously been hidden. We exploit this fact as a natural experiment to identify the causal effect of negative attitudes towards immigrants on foreigners’ location choices and thus indirectly on their utility. Based on a regression discontinuity design with unknown discontinuity points and administrative data on the population of foreigners, we find that the probability of their moving to a municipality which unexpectedly expressed stronger reservations decreases initially by about 40%. The effect is accompanied by a drop of housing prices in these municipalities and levels off over a period of about 5 months. Moreover, foreigners in high-skill occupations react relatively more strongly highlighting a tension when countries try to attract well-educated professionals from abroad.

This is a preview of subscription content, log in to check access.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Notes

  1. 1.

    A notable exemption is the study by Elsayed and De Grip (2018) who find that Islamist terrorist attacks worsen attitudes towards Muslims in the Netherlands and go hand in hand with an increase in Muslims’ intention to re-migrate. Moreover, Muslims’ attitudes towards integration are also negatively affected by the terrorist attacks. Both effects are relatively larger for the highly educated immigrants.

  2. 2.

    The Swiss direct democratic system, among other things, allows the electorate to propose amendments to the federal constitution by launching a so-called popular initiative. Initiators have to collect 100,000 valid signatures amounting to about 2% of the population of Swiss citizens. Once this hurdle is overcome, voters decide in a national referendum on whether the constitution should be changed. This is the case if the initiative is approved by the majority of voters overall as well as in a majority of cantons. If this holds, the federal government is obliged to implement the will of the majority. For an overview about the Swiss political system, see, for example, Linder (2010). Swiss citizens are accustomed to expressing their opinions at the poll (Stutzer et al. 2018). Direct democratic decision-making is very common in Switzerland, even at the national level. Citizens openly discuss the proposals and can rely on being provided with ample coverage of the issues in the media. Newspapers, for example, publish the statements and voting recommendations of political parties and opinion leaders. Frey (1994) analyses the role of public discussion in the process of direct democratic decision-making, with an emphasis on the Swiss experience.

  3. 3.

    The campaign poster was, for example, forbidden in Basel, Freiburg, Lausanne, Morges, Neuenburg, Nyon, and Yverdon.

  4. 4.

    Figures A.2 and A.3 in Slotwinski and Stutzer (2018) provide examples of maps displaying the outcome where approval is marked green and rejection is marked red.

  5. 5.

    These votes include (i) the initiative “For the regulation of immigration” (“Initiative für eine Regelung der Zuwanderung”) held on September 24, 2000, (ii) the initiative “Against the abuse of the asylum law” (“Initiative gegen Asylrechtsmissbrauch”) held on November 24, 2002, (iii) the referendum on “The federal decree regarding the regular naturalization and the easier naturalization of young, and second-generation foreigners” (“Referendum zum Bundesbeschluss über die ordentliche Einbürgerung sowie über die erleichterte Einbürgerung junger Ausländerinnen und Ausländer der zweiten Generation”) held on September 26, 2004, and (iv) the referendum on “The federal decree about the acquisition of citizenship rights by third-generation foreigners” (“Referendum zum Bundesbeschluss über die ordentliche Einbürgerung sowie über erleichterte Einbürgerung junger Ausländerinnen und Ausländer der dritten Generation”) also held on September 26, 2004.

  6. 6.

    We test this supposition in a supplementary empirical analysis (see Section 7.3) and do not find systematic reactions in the decisions to move.

  7. 7.

    For example, regions of the brain that are important for pain detection are activated. Moreover, individuals are highly sensitive in detecting situations of exclusion (see, e.g., Williams 2009; Williams and Nida 2011). Related evidence based on panel data in Germany shows that higher vote shares of right-wing parties are associated with lower reported life satisfaction for immigrants (Knabe et al. 2013).

  8. 8.

    Please note that the two conditions are close to the identifying assumptions in a RDD context. Condition I is close to the the conditional independence assumption, and condition II is close to the local randomization assumption.

  9. 9.

    Our identification strategy deviates from related strategies based on close election outcomes (see, e.g., Folke 2014). In particular, we are not studying the consequences of the vote outcome on the people who voted as it is in some form usually the case in studies using election results. Instead, we study the consequences of the newly revealed information for a group of people, i.e., the foreigners outside of the municipality, who are not themselves involved in generating the vote outcome. These foreigners face the same (or a more) limited set of information as we the statisticians. Given the publicly available information, the deviations from the past voting records were news.

  10. 10.

    Appendix B in the longer working paper version (Slotwinski and Stutzer 2018) presents the details of the analysis.

  11. 11.

    Card et al. (2008) are—to our knowledge—the first who try to estimate an unknown threshold value in order to estimate effects at this point. In their application, they are interested in the presence of a minority population in a municipality and the share after which the majority starts to leave the municipality; i.e., a tipping point based on the Schelling model about dynamics in segregation (Schelling 1971).

  12. 12.

    A discussion of the fuzzy case and an in-depth discussion of the theoretical derivations and the method are presented in the original paper (Porter and Yu 2015).

  13. 13.

    Appendix C in Slotwinski and Stutzer (2018) offers additional details on the discontinuity estimation in the context of our analysis.

  14. 14.

    When moving within Switzerland foreigners have to register in the new municipality of residence, as well as to de-register in the former municipality of residence within 14 days. In addition, they have to provide a copy of the rental contract. This data is then systematically collected by the FSO. As this registration is an obligation, i.e., a deviation is an offense, and the period is rather close to the moving date, the registered moving dates should provide us with an accurate representation of the actual individual moving behavior.

  15. 15.

    Our empirical strategy draws on the following cantons: Zürich (ZH), Bern (BE), Luzern (LU), Schwyz (SZ), Nidwalden (NW), Glarus (GL), Zug (ZG), Fribourg (FR), Solothurn (SO), Basel-Stadt (BS), Basel-Landschaft (BL), Schaffhausen (SH), Appenzell-Ausserrhoden (AR), St. Gallen (SG), Graubünden (GR), Aargau (AG), Thurgau (TG), Ticino (TI), Vaud (VD), Valais (VS), Neuchâtel (NE), and Jura (JU).

  16. 16.

    Regarding the necessary bandwidth choice in the applied nonparametric procedure, we abstain from using a technique that either depends on the variance of the assignment variable or cross-validation. First, our running variable is a date and the variation of such a variable makes no sense. Cross-validation would be theoretically appealing. However, it is computationally intensive, given the large number of observations and the simulation-based approach. We rather rely on eye-balling and group cantons according to size when choosing the bandwidth that is as small as possible but which makes the graphs reasonably smooth. We start with a bandwidth size of 45 days for the largest cantons ZH, BE, AG, and VD. This bandwidth size ensures that each estimation includes one beginning and one end of the month, when relocations are frequent. The next group contains the cantons LU, FR, SO, BL, SG, TG, TI, and VS for which a bandwidth of 60 days is applied. We use 75 days for the cantons SZ, ZG, BS, SH,GR, and NE. For the smallest ones, i.e., NW, GL, AR, and JU, a bandwidth of 90 days is chosen. These bandwidths are used for the RDD graphs, the RDD estimates and the threshold search. As the bandwidth for the testing has to be considerably smaller than for these estimates, we choose to use half of it in the testing procedure.

  17. 17.

    These cantons are ZH, BE, SZ, NW, GL, SO, AR, GR, TG, TI, VD, and VS. They cover about 66% of the moves within our sample period.

  18. 18.

    The canton of Zug is a special case where switching municipalities have not changed their relative position, and thus we are not surprised by not detecting a reaction.

  19. 19.

    In our specific setting, the division between treated and untreated individuals is probably not perfect at the threshold value, as there is most likely no perfect temporal separation between those people who decided where to move after the new information became available (treated group) and those people who decided where to move beforehand (control group). Consequently, our estimates should rather be interpreted as a local intent-to-treat effect.

  20. 20.

    The discontinuity estimates and RDD graphs for single cantons are reported in Table 10 and Fig. 8 in Appendix 3. As addressed before, we use the residual of a regression on time indicators to control for periodical patterns, thus in “normal” times it varies around zero.

  21. 21.

    Note that a decreasing estimate of the discontinuity with increasing bandwidths is consistent with the fading of the effect over time.

  22. 22.

    We have no information about an individuals’ religious affiliation. Instead, we infer that individuals originating from countries where Islam is the main religion are more likely to be Muslim. The classification on the main religion of countries is based on the Cross-National Socio-Economic and Religion Data, 2011 and was downloaded from the Association of Religion Data Archives, www.TheARDA.com.

  23. 23.

    We code somebody as high-skilled if his or her occupation corresponds to an ISCO skill level of 4, i.e., the second stage of tertiary education or first stage of tertiary education (medium duration). Upper-medium skilled, i.e., ISCO skill level of 3 and thus the first stage of tertiary education (short or medium duration). Medium skilled people refer to an ISCO skill level of 2, i.e., lower to post-secondary education. Foreigners considered low-skilled are in an occupation of ISCO skill level 1 which requires primary education.

  24. 24.

    Travel time is collected using Stata’s Traveltime3 command (Bernhard 2013), which retrieves travel time between two locations from Google Maps.

  25. 25.

    We again follow, for example, Davis (2008) in controlling for these periodical patterns by using a battery of indicator variables. We first run the specification

    $$ y_{t}=\beta_{0} + \beta x_{t} +u_{t}, $$

    where the dependent variable is yt, the number of moving individuals at a certain date t, and xt includes indicator variables for months, the day of the month and weekdays. We estimate over the period between 2006 and 2008, thus before the event, in order to control for moving behavior in normal times. We then predict the residual ut for the entire time period and use it as our dependent variable. For convenience, we still refer to this variable as the number of moving individuals.

  26. 26.

    More specifically, we choose those municipalities that changed their position in the ranking by only up to two steps.

  27. 27.

    As addressed before, we use the residual of a regression on time indicators to control for periodical patterns, thus in “normal” times it varies around zero.

References

  1. Akay A, Constant A, Giulietti C, Guzi M (2017) Ethnic Diversity and Well-being. J Popul Econ 30(1):265–306

    Article  Google Scholar 

  2. Akerlof G, Kranton R (2000) Economics and identity. Q J Econ 115 (3):715–753

    Article  Google Scholar 

  3. Algan Y, Bisin A, Manning A, Verdier T (2012) CUltural integration of immigrants in Europe. Oxford University Press, Oxford

    Google Scholar 

  4. Barone G, D’Ignazio A, de Blasio G, Naticchioni P (2016) Mr. Rossi, Mr. Hu and Politics. The role of immigration in shaping natives’ voting behavior. J Public Econ 136:1–13

    Article  Google Scholar 

  5. Bernhard S (2013) Stata Command TRAVELTIME3: Stata Command to Retrieve Travel Time and Road Distance Between Two Locations. Statistical Software Components. Boston College Department of Economics

  6. Branscombe N, Schmitt MT, Harvey R (1999) Perceiving pervasive discrimination among African Americans: implications from group identification and well-being. J Pers Soc Psychol 77(1):135–149

    Article  Google Scholar 

  7. Card D, Mas A, Rothstein J (2008) Tipping and the dynamics of segregation. Q J Econ 123(1):177–218

    Article  Google Scholar 

  8. Damm AP (2009) Determinants of recent immigrants’ location choices: quasi-experimental evidence. J Popul Econ 22(1):145–174

    Article  Google Scholar 

  9. Davis LW (2008) The effect of driving restrictions on air quality in Mexico City. J Polit Econ 116(1):38–81

    Article  Google Scholar 

  10. Dill V (2013) Ethnic concentration and extreme right-wing voting behavior in West Germany. SOEP Paper No. 565

  11. Dustmann C, Preston I (2001) Attitudes to ethnic minorities, ethnic context and location decisions. Econ J 111(470):353–373

    Article  Google Scholar 

  12. Dustmann C, Preston I (2007) Racial and economic factors in attitudes to immigration. The B.E. Journal of Economic Analysis and Policy 7(1):Article 62

    Article  Google Scholar 

  13. Elsayed A, De Grip A (2018) Terrorism and the integration of muslim immigrants. J Popul Econ 31(1):45–67

    Article  Google Scholar 

  14. Ettinger P, Imhof K (2014) Qualität der medienberichterstattung zur minarett-initiative. In Abstimmungskampagnen. Springer, pp 357–369

  15. Folke O (2014) Shades of brown and green: party effects in proportional election systems. J Eur Econ Assoc 12(5):1361–1395

    Article  Google Scholar 

  16. Freitag M, Rapp C (2013) Intolerance toward immigrants in switzerland: diminished threat through social contacts? Swiss Polit Sci Rev 19(4):425–446

    Article  Google Scholar 

  17. Frey BS (1994) Direct democracy: politico-economic lessons from Swiss experience. Am Econ Rev 84(2):338–342

    Google Scholar 

  18. Gerdes C, Wadensjö E (2008) The impact of immigration on election outcomes in Danish municipalities. IZA Discussion Paper No. 3586

  19. Gorinas C, Pytliková M (2017) The influence of attitudes toward immigrants on international migration. Int Migr Rev 51(2):416–451

    Article  Google Scholar 

  20. Hahn J, Todd P, van der Klaauw W (2001) Identification and estimation of treatment effects with a regression-discontinuity design. Econometrica 69(1):201–209

    Article  Google Scholar 

  21. Hainmueller J, Hiscox MJ (2010) Attitudes toward highly skilled and low-skilled immigration: evidence from a survey experiment. Am Polit Sci Rev 104(1):61–84

    Article  Google Scholar 

  22. Hainmueller J, Hopkins DJ (2014) Public attitudes toward immigration. Ann Rev Polit Sci 17:225–249

    Article  Google Scholar 

  23. Halla M, Wagner A, Zweimüller J (2017) Immigration and voting for the far right. J Eur Econ Assoc 15(6):1341–1385

    Article  Google Scholar 

  24. Henry R (2009) Does racism affect a migrant’s choice of destination? IZA Discussion Paper No. 4349

  25. Imbens GW, Lemieux T (2008) Regression discontinuity designs: a guide to practice. J Econ 142(2):615–635

    Article  Google Scholar 

  26. International Labour Office (2012) ISCO-08. International standard classification of occupations: structure Group Definitions and Correspondence Tables 1

  27. Knabe A, Rätzel S, Thomsen SL (2013) Right-wing extremism and the well-being of immigrants. Kyklos 66(4):567–590

    Article  Google Scholar 

  28. Kuhn A, Brunner B (2018) Immigration, cultural distance and natives’ attitudes towards immigrants: evidence from Swiss voting results. Kyklos 71(1):28–58

    Article  Google Scholar 

  29. Lee D, Lemieux T (2010) Regression discontinuity designs in economics. J Econ Lit 48(2):281–355

    Article  Google Scholar 

  30. Linder W (2010) SWiss democracy: possible solutions to conflict in multicultural societies. Palgrave Macmillan, Basingstoke

    Google Scholar 

  31. Liu RY (1988) Bootstrap procedures under some non-I.I.D. Models. Ann Stat 16(4):1696–1708

    Article  Google Scholar 

  32. McFadden D (1974) Conditional logit analysis of qualitative choice behavior. In: Zarembka E, McFadden D (eds). Academic Press, New York, pp 105–142

  33. Méndez Martínez I, Cutillas IM (2014) Has immigration affected spanish presidential elections results? J Popul Econ 27(1):135–171

    Article  Google Scholar 

  34. Paola Md, Scoppa V, Falcone M (2012) The deterrent effects of the penalty points system for driving offences: a regression discontinuity approach. Empir Econ 45(2):965–985

    Article  Google Scholar 

  35. Parchet R (2014) Are local tax rates strategic complements or strategic substitutes? IdEP Economic Papers, University of Lugano (2014/07)

  36. Porter J (2003) Estimation in the regression discontinuity model mimeo. Department of Economics, Harvard University

  37. Porter J, Yu P (2015) Regression discontinuity design with unknown discontinuity points: testing and estimation. J Econ 189(1):132–147

    Article  Google Scholar 

  38. Rudert S, Janke S, Greifeneder R (2017) Under threat by popular vote: naturalistic social exclusion due to the Swiss vote against mass immigration. PLOS ONE 12(8):e0182703

    Article  Google Scholar 

  39. Schelling TC (1971) Dynamic models of segregation. J Math Sociol 1(2):143–186

    Article  Google Scholar 

  40. Schmitt MT, Spears R, Branscombe NR (2003) Constructing a minority group identity out of shared rejection: the case of international students. Eur J Soc Psychol 33(1):1–12

    Article  Google Scholar 

  41. Slotwinski M, Stutzer A (2018) The deterrent effect of an anti-minaret vote on foreigners’ location choices. WWZ Working Paper 2018/28

  42. Stutzer A, Baltensperger M, Meier AN (2018) Overstrained citizens? The number of ballot propositions and the quality of the decision process in direct democracy. WWZ Working Paper 2018/25

  43. Tajfel H (1978) The social psychology of minorities. Minority Rights Group, London

    Google Scholar 

  44. Thiriet M (2009) Schweizer Muslime sind wütend, enttäuscht und vor den Kopf gestossen. Tages Anzeiger

  45. Thistlethwaite DL, Campbell DT (1960) Regression-discontinuity analysis: an alternative to the ex post facto experiment. J Educ Psychol 51(6):309

    Article  Google Scholar 

  46. Tolnay SE, Beck EM (1992) Racial violence and Black migration in the American South, 1910 to 1930. Am Sociol Rev 57(1):103–116

    Article  Google Scholar 

  47. Waisman G, Larsen B (2016) Income, amenities and negative attitudes. IZA J Migr 5(1):8

    Article  Google Scholar 

  48. Wehrli C (2009) Klares, aber vieldeutiges Nein zu Minaretten. Neue Zürcher Zeitung

  49. Wesselmann ED, Bagg D, Williams KD (2009) I feel your pain: the effects of observing ostracism on the ostracism detection system. J Exper Soc Psychol 45 (6):1308–1311

    Article  Google Scholar 

  50. Williams KD (2009) ADvances in experimental social psychology: ostracism: a temporal need-threat model, vol 41. Academic Press, San Diego

    Google Scholar 

  51. Williams KD, Nida SA (2011) Ostracism: consequences and coping. Curr Dir Psychol Sci 20(2):71–75

    Article  Google Scholar 

  52. Wu CFJ (1986) Jackknife, bootstrap and other resampling methods in regression analysis. Ann Stat 14(4):1261–1295

    Article  Google Scholar 

Download references

Acknowledgments

Data on the foreign resident population was kindly provided by the Federal Statistical Office in cooperation with the State Secretariat for Migration. We are grateful to Bruno Frey, Ulrich Matter, Armando Meier, Reto Odermatt, Markus Roller, Kurt Schmidheiny, Simone Schotte, and Conny Wunsch for detailed discussions and many seminar participants for insightful comments as well as three anonymous referees and the Editor Klaus F. Zimmermann for their guidance in improving this paper. We are further grateful to Jack Porter and Ping Yu for sharing the basic code of their method. Michaela Slotwinski acknowledges financial support from WWZ Forum, the NCCR - On the Move funded by the Swiss National Science Foundation and the Sinergia Grants 130648 and 147668 of the Swiss National Science Foundation.

Funding

This study was funded by the National Center of Competence in Research - The Migration-Mobility Nexus financed by the Swiss National Science Foundation, the WWZ Forum, and the Sinergia Grants 130648 and 147668 of the Swiss National Science Foundation.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Michaela Slotwinski.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Responsible editor: Klaus F. Zimmermann

Appendices

Appendix 1: Exemplary analysis for a single canton

In order to provide an intuition for the application of the methodology used, we exemplify our econometric proceeding for the canton of Thurgau (TG). The dependent variable of interest is an indicator of whether a moving individual chooses to locate in a switcher municipality.Footnote 27

We first visually inspect how the probability that a foreigner chooses to move to one of the switcher municipalities evolves over time, whereby we undersmooth the estimate using the same bandwidth as in the specification testing. As can be seen in Figure 6, the probability is rather stable until the vote, which is marked by the solid vertical line. Some time after the vote has taken place, the probability seems to drop temporarily. For the detection of a possible threshold, we define a time span within which we expect the jump to occur, if there is one. This time span is marked by the two dashed vertical lines.

Fig. 6
figure6

Probability of moving to a switcher municipality in the canton of Thurgau. Local linear smooth of the probability that a foreigner moves to a switcher municipality, using a bandwidth of 30 days. The light gray dots represent raw means within bins of 15 days

We perform the specification test as described in Section 4 (or in Appendix C in Slotwinski and Stutzer (2018) in more detail); i.e., testing the null hypothesis of no treatment effect, within the defined time span. We find that the null hypothesis is rejected for this canton. The test statistic is 2.62 and exceeds the critical value of 0.79 at the 5% significance level. These results are also listed in Table 9 in Appendix 3. The estimated threshold date for the canton of Thurgau is March 2, 2010 and, thus about 3 months after the vote has taken place. We proceed with a classical RDD analysis and plot a local linear smooth separately from both sides of the estimated threshold. In Fig. 7, a negative jump in the probability that a foreigner chooses to locate in a switcher municipality is observed at the estimated threshold date. Moreover, it seems that the effect fades and that after some time the probability that a foreigner moves to a switcher municipality returns to its former level.

Fig. 7
figure7

Probability of moving to a switcher municipality in the canton of Thurgau-RDD graph around the threshold date March 2, 2010. Local linear smooth of the probability that a foreigner moves to a switcher municipality separately from both sides of the threshold, and using a bandwidth of 60 days. The light gray dots represent raw means within bins of 15 days

When we estimate the treatment effect at the detected threshold, we find a statistically significant negative discontinuity at the particular date. The probability of moving to a switcher municipality drops by 16 percentage points at the threshold. The details of the estimation result are listed in Table 10 in Appendix 3.

Appendix 2: Descriptive statistics

Table 8 Descriptive statistics

Appendix 3: Results for the individual cantons

Table 9 Results for the specification test and the estimated threshold dates
Fig. 8
figure8

Probability of moving to a switcher municipality in 12 cantons—RDD plots around the estimated threshold dates. Local linear smooth, using the same bandwidth (h) as in the estimations in Table 10, of the probability that a foreigner chooses to move to a switcher municipality over the moving date, separately from both sides of the threshold for all cantons in which a discontinuity was found. The dashed vertical lines indicate the testing region, the first solid line indicates the vote date, and the second the estimated threshold date. The light gray dots represent raw means within bins of 15 days

Table 10 RDD estimates of the probability that foreigners move to a switcher municipality at the estimated thresholds for 12 cantons

Appendix 4: Additional tables and figures

Table 11 RDD estimates of the probability that foreigners move to different definitions of non-switcher municipalities
Table 12 Individual characteristics of movers around the threshold dates
Table 13 Municipality characteristics before and after the vote on the minaret initiative
Table 14 Number of moving individuals around the threshold dates
Fig. 9
figure9

Number of individuals moving around the threshold in the pooled sample of cantons. Local linear smooth of the number of individuals who move over the moving date, separately from both sides of the threshold, using a bandwidth of 90 days. The light gray dots represent raw means within bins of 10 days

Fig. 10
figure10

Number of individuals moving to a switcher municipality in the pooled sample of cantons. Local linear smooth of the number of individuals who move to a switcher municipality over the moving date, separately from both sides of the threshold, using a bandwidth of 90 days. The light gray dots represent raw means within bins of 10 days

Fig. 11
figure11

Moving-pattern in 12 cantons for which a discontinuity was detected. This graph shows the histograms of moving dates in our estimation sample and for cantons in which the method detects a systematic discontinuity. The share in the parentheses after the canton abbreviation indicates the share of individuals moving at the last or the first day of the month. The date of the minaret initiative is marked by the red solid vertical line, the estimated threshold date by the black solid vertical line, and the typical moving dates (31 March, 30 June, and 30 September) by the gray dashed vertical lines

Fig. 12
figure12

Moving-pattern in 10 cantons for which no discontinuity was detected. This graph shows the histograms of moving dates in our estimation sample for cantons for which no systematic discontinuity was detected. The share in the parentheses after the canton abbreviation indicates the share of individuals moving at the last or the first day of the month. The date of the minaret initiative is marked by the red solid vertical line and the typical moving dates (31 March, 30 June, and 30 September) by the gray dashed vertical lines

Table 15 Development of announced housing prices in switcher municipalities after the vote
Fig. 13
figure13

Raw monthly means of rental and selling prices over time for cantons in which a discontinuity was detected. The dashed line represents the means for non-switcher municipalities in the comparison group and the solid line the treatment group of switcher municipalities. The solid vertical black line indicates the month of the vote

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Slotwinski, M., Stutzer, A. The deterrent effect of an anti-minaret vote on foreigners’ location choices. J Popul Econ 32, 1043–1095 (2019). https://doi.org/10.1007/s00148-019-00729-6

Download citation

Keywords

  • Attitudes
  • Foreigners
  • Location choice
  • Popular initiative
  • Regression discontinuity design

JEL Classifications

  • D83
  • J61
  • R23
  • Z13