Voting locally abstaining nationally: descriptive representation, substantive representation and minority voters’ turnout

ABSTRACT

Do disadvantaged minorities turnout less when they have descriptive yet not substantive representation? When minorities have descriptive representation, their representatives’ activities and preferences might not necessarily increase minorities’ public goods’ access. With substantive representation, minority representatives pursue policies that improve minorities’ public goods’ provisions. We claim that voters’ calculus might decrease turnout in the first case and increase it in the latter case. We examine this claim in the Israeli context focusing on the Arab-Muslim minority in Israel and comparing it to the Jewish majority and Druze minority. We show that different minority representation qualities in different institutions induce different turnout behaviours across institutions within the same minority. Specifically, Muslim Arabs decreased their turnout in national elections and increased it in local elections.

KEYWORDS:

• Descriptive representation
• substantive representation
• electoral turnout
• second-order election
• Arab Israelis
• ballot-box data

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 We use this term to refer to the part of Israel which lies between the Mediterranean Sea and the Green Line. This part does not include Palestinian inhabitants of the West Bank or the Gaza Strip.

2 http://www.cbs.gov.il/publications11/local_authorities09/pdf/112_113.pdf (last viewed 11 August 2016).

3 We use 2010 data as these data were measured immediately after the local and national elections taking place between 2003 and 2008, and hence are the most relevant for our tests.

4 http://www.cbs.gov.il/shnaton67/st02_02.pdf (viewed 12 September 2016).

5 It should be noted that as a large part of this population resides in East Jerusalem and a smaller part of the Druze population resides in the Golan Heights and therefore are not eligible to vote in national elections. Thus, we did not include these voters in our analysis. Yet, we present it here as a part of the general statistics regarding the Arab population in Israel. http://www.cbs.gov.il/www/statistical/arab_pop08.pdf (viewed 12 September 2016).

6 http://www.cbs.gov.il/www/statistical/arab_pop08.pdf (viewed 12 September 2016).

7 http://www.cbs.gov.il/reader/newhodaot/hodaa_template.html?hodaa=201524029 (viewed 14 August 2016).

8 The variance between the two national elections and the two local elections in ballot boxes number stems from demographic changes, the formation of new towns and population growth. Yet, the variance between local and national elections in terms of ballot-box numbers, stems from an intriguing phenomenon where many ballot-boxes that exist in national elections are gone come local elections.

9 The complete statistical analysis and results are obtainable in the following URL: https://www.researchgate.net/profile/Maoz_Rosenthal/contributions.

10 http://www.cbs.gov.il/reader/newhodaot/hodaa_template.html?hodaa=200824160 (viewed 14 September 2016).

11 http://www.cbs.gov.il/reader/newhodaot/hodaa_template.html?hodaa=201324087 (viewed 15 May 2013).

12 For the matching process we described above we filtered out 600 ballot boxes in local elections and about 500 ballot boxes in national elections.