Breaking with Dogma: Unorthodox Consumption Patterns and Women's Labor Market Outcomes in Turkey

ABSTRACT

Using information about household consumption data from TURKSTAT's Household Surveys for 2007–13 as a sign of religious unorthodoxy, this study explores the effect of religion on women's labor force and educational participation in a Muslim-majority country, Turkey. A household is categorized as “unorthodox” if its members report that they consume goods that contradict conservative Sunni practices, such as alcohol. This information is then used in female labor force participation estimations. Results show that living in an unorthodox household has a positive and highly significant effect on the probability of married women's labor market participation. For single women, the estimations provide weaker evidence regarding the positive effect of unorthodoxy on the probability of participation in education and the labor force. The study concludes that protection of the rights to follow unorthodox practices in society may bear positive implications with regard to women's agency.

KEYWORDS:

• Consumption
• female labor force participation
• religion

JEL Codes:

• J16
• Z12

ACKNOWLEDGMENTS

Earlier versions of this paper were presented at the XXVIII National Conference of Labour Economics, Libera Università Internazionale degli Studi Sociali “Guido Carli” Roma, September 27–28, 2013; METU Gender and Women's Studies Conference Ankara, October 9–11, 2015; and the UNDP Session of EY International Congress of Economics Ankara, November 5–6, 2015. I would like to thank the participants of each session for their valuable suggestions. Many thanks also go to Emel Memiş, Todd Goehle, three anonymous reviewers, and the Associate Editor of Feminist Economics for their constructive comments.

NOTES ON CONTRIBUTOR

F. Kemal Kızılca is lecturer at the Economics Department of Ankara University (Turkey). He previously worked as a postdoctoral researcher at the Economic Policies Research Unit of Minho University (Portugal). Kızılca's research interests include gender economics, labor economics, and post-Keynesian economics.

Notes

1 Florence Jaumotte (2003) and Becky Pettit and Jennier Hook (2005) provide multicountry analyses of FLFP in Western and developed countries. For the MENA region, see Valentine M. Moghadam (1998) and Niels Spierings, Jeroen Smits, and Mieke Verloo (2010).

2 For a socio-anthropological account of seafood consumption and its relation with religion in Turkey, see Ståle Knudsen (2006).

3 Statistics are not reliable as to the number of followers of each religious denomination in Turkey. According to estimates, Alevis compose 10 to 25 percent of the total population of Turkey (David Shankland 2003; Tahire Erman and Emrah Goker 2000).

4 Figures in section 3.2 indicate higher rates because of truncation of age at 64 years.

5 For more information, see http://www.turkstat.gov.tr/PreTablo.do?alt_id=1059.

6 Almost no household in the surveys declare consumption of pork. Pork is only available in certain supermarkets of the largest Turkish cities.

7 The total estimated value of declared beer consumption in 2013 is 750,000,000 liras, which corresponds to 93,750,000 liters at the 2013 price of 4 lira per half-liter bottle. According to the data of the Tobacco and Alcohol Market Regulatory Authority of 2013, the total amount of beer sold in 2013 was actually 880 million liters, around nine times as high as the reported amount in the household survey. http://www.tapdk.gov.tr/piyasa_duzenlemeleri/alkollu_ickiler_piyasasi/kamuoyuna_yonelik_bilgiler/2004-2014_yillari_litre.xlsx.

8 I obtained these figures from the institutions’ websites, www.worldvaluessurvey.org and www.worldchristiandatabase.org.

9 For detailed information on the Alevis in Turkey, see Shankland (2003).

10 A recent health survey conducted by Belgin Ünal and Gül Ergör (2013) suggests that the prevalence of alcohol consumption in Turkey is 13 percent (23 percent and 4 percent among men and women, respectively).

11 For example, see the studies summarized in Mark R. Killingsworth and James J. Heckman (1986), Jacob Mincer's important study (1962), as well as Murphy (1995), Lehrer (1995), Heineck (2004), Libertad Gonzales (2004), and Gündüz-Hoşgör and Smits (2008).

12 For husbands’ influence over women's workforce participation, see also Gul Ozyegin (2010) and Beşpınar (2010).

13 This corresponds to $6,700 approximately at the mean exchange rate of 2003.

14 Choosing a narrower (wider) range results in a slight increase (decrease) in coefficient estimates.

15 See the studies cited in section 2.4 as well as the results on education provided in Table 3.

16 http://www.oecd.org/ctp/consumption/Table-4.A4.3-Taxation-of-alcoholic-beverages-Dec-2014.xls.2015.

17 In the dataset, there also exist some other items that are purchased by a small share of the households; however, I opt to exclude them from the analysis as they are all pooled definitions such as “other food n.e.c.,” “other sugar products,” “other milk products,” and so on.

18 See Jennifer C. Olmsted (2005), Beşpınar (2010), Seçil A. Kaya Bahçe and Emel Memiş (2013), and Bahramitash and Olmsted (2014) for related discussions.

19 In current Turkey, over 85 percent of the citizens declare themselves as “religious”; see http://www.bahcesehir.edu.tr/icerik/1725-turkiye-degerler-atlasi-2012-yayinlandi.