1 URBAN INFORMALITY AND ECONOMIC VULNERABILITY THE CASE OF TURKEY SUMMARY Based on four different national surveys in Turkey (household, shopkeepers and bazaaris and street vendors), the paper proposes that informality can be better understood as an activity resembling a continuum rather than arbitrary formal/informal dichotomies. Measuring the prevalence of informal practices through six indicators suggests that shopkeepers tend to be more formal than household while bazaaris and street vendors appear more informal. Many factors determine the degree of informality, but degree of economic vulnerability is one of the most important in capturing why and how individuals adopt informal practices. As degree of economic vulnerability increases so does degree of informality. Keywords—informality, Turkey, economic vulnerability THE AUTHORS Ali Çarkoglu Mine Eder Sabanci University Bogazici University Faculty of Arts and Social Sciences Department of Political Science and IR Orhanli, Tuzla 34956 Istanbul-Turkey Bebek-Istanbul 34342 Tel: 90 216 483 9297 Tel: 90 212 3596535 or 6502 Fax: 90 216 483 9250 Fax: 90 212 287 24 55 [email protected][email protected]CONTACT ADDRESS TO BE USED Mine Eder Visiting Professor (September 2006-June 2007) Yale University The MacMillan Center for International and Area Studies Box: 208206 New Haven Connecticut, USA 06520-8206 Tel: 203 432 5596 Fax: 203 432 5963 ACKNOWLEDGEMENT • Authors would like to thank Middle East Research Competition (MERC) of Ford Foundation for the initial grant, Bogazici University, Istanbul and Sabanci University, Istanbul for their matching research grants without which these surveys would not have been possible. We also would like to thank the Bellagio Residency Program of Rockefeller Foundation which made the joint writing of this paper possible.
43
Embed
URBAN INFORMALITY AND ECONOMIC …research.sabanciuniv.edu/28/1/stvkaf01670.pdf2 URBAN INFORMALITY AND ECONOMIC VULNERABILITY THE CASE OF TURKEY SUMMARY Based on four different national
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
URBAN INFORMALITY AND ECONOMIC VULNERABILITY
THE CASE OF TURKEY
SUMMARY Based on four different national surveys in Turkey (household, shopkeepers and bazaaris and street vendors), the paper proposes that informality can be better understood as an activity resembling a continuum rather than arbitrary formal/informal dichotomies. Measuring the prevalence of informal practices through six indicators suggests that shopkeepers tend to be more formal than household while bazaaris and street vendors appear more informal. Many factors determine the degree of informality, but degree of economic vulnerability is one of the most important in capturing why and how individuals adopt informal practices. As degree of economic vulnerability increases so does degree of informality.
THE AUTHORS Ali Çarkoglu Mine Eder Sabanci University Bogazici University Faculty of Arts and Social Sciences Department of Political Science and IR Orhanli, Tuzla 34956 Istanbul-Turkey Bebek-Istanbul 34342 Tel: 90 216 483 9297 Tel: 90 212 3596535 or 6502 Fax: 90 216 483 9250 Fax: 90 212 287 24 55 [email protected][email protected] CONTACT ADDRESS TO BE USED Mine Eder Visiting Professor (September 2006-June 2007) Yale University The MacMillan Center for International and Area Studies Box: 208206 New Haven Connecticut, USA 06520-8206 Tel: 203 432 5596 Fax: 203 432 5963 ACKNOWLEDGEMENT
• Authors would like to thank Middle East Research Competition (MERC) of Ford Foundation for the initial grant, Bogazici University, Istanbul and Sabanci University, Istanbul for their matching research grants without which these surveys would not have been possible. We also would like to thank the Bellagio Residency Program of Rockefeller Foundation which made the joint writing of this paper possible.
2
URBAN INFORMALITY AND ECONOMIC VULNERABILITY
THE CASE OF TURKEY
SUMMARY
Based on four different national surveys in Turkey (household, shopkeepers and bazaaris and street
vendors), the paper proposes that informality can be better understood as an activity resembling a
continuum rather than arbitrary formal/informal dichotomies. Measuring the prevalence of informal
practices through six indicators suggests that shopkeepers tend to be more formal than household
while bazaaris and street vendors appear more informal. Many factors determine the degree of
informality, but degree of economic vulnerability is one of the most important in capturing why and
how individuals adopt informal practices. As degree of economic vulnerability increases so does
Those who indicated that when buying goods with long-term payments only a handshake is enough to finalize the
deal 28.60 36.30 32.10
Those who indicated that when buying goods 75% or
more of the payments are made in cash 61.20 82.40 96.00
Those who indicated that when selling goods 75% or
more of the payments of their customers are made in cash 80.80 98.40 100.00
Those who indicated that 50% or more of the goods they
are selling are obtained without receipt 13.40 28.10 59.30
Informality index (add from 1 to 4) 0 8.1 1.0 0.0 1 21.3 7.8 1.8 2 36.2 38.9 23.2 3 24.7 37.9 60.6 4 2.7 9.8 14.4 Mean 1.9 2.5 2.9 Standard Deviation 1.0 0.8 0.7 Cronbach's Alpha N* 1221 619 304 327
*Missing values in some of the questions leads to different sample sizes for different target groups that are smaller than the total number of interviews conducted.
27
Table 2. Measuring Vulnerability across 4 Target Populations
Household Shopkeepers Bazaaris Street
Vendors Variable Description % % % %
1 Those who do not own a car 76.8 61.4 62.4 89.9
2 Those who do not own a washing machine 11.7 8.9 22.9 37.6
3 Those who do not own a house/apartment 41.7 32.1 47.7 61.2
4 Per capita household income less than 100 YTL 24.0 20.7 41.8 57.5
5 Those who declare that their income is not at all
satisfactory 13.8 9.9 23.5 28.4
6 Those who declare that they can only a month or less if
their income is somehow cut 69.8 58.5 82.4 88.4
Vulnerability index (add from 1 to 6) 0 6.3 11.1 2.6 0.3 1 17.5 26.0 19.9 6.1 2 29.9 27.6 19.6 11.3 3 23.1 16.5 23.9 27.5 4 11.6 8.9 16.7 27.2 5 5.4 2.1 10.1 17.4 6 1.9 0.6 4.9 8.9 Above four sample's average vulnerability (3-6) 42.0 28.1 55.6 81.0 Mean 2.4 1.9 2.8 3.7 Standard Deviation 1.3 1.3 1.5 1.3 Cronbach's Alpha 0.52 0.44 0.55 0.44
28
Table 3. Reported length of time that could be managed, if household income was cut off for some reason
Household Shopkeepers Bazaaris Street
Vendors Could not manage a single day 19.2 15.3 14.4 20.8
Less than 1 month 50.6 43.1 68.0 67.6 1 to 3 months 9.4 17.3 10.5 7.0 3 to 6 months 7.7 8.9 2.6 1.5
6 to 12 months 2.8 0.6 0.0 0.0 1 year and longer 7.9 13.4 4.2 2.1
NR 2.4 1.3 0.3 0.9
29
Table 4. Ordered Logit Estimates for the Determinants of Informality Index
Household Shopkeepers Bazaaris Street Vendors Threshold Estimate Sig. Estimate Sig. Estimate Sig. Estimate Sig.
*Kurdish origin is determined by whether or not the respondent reports to have spoken to their parents in their childhood in Kurdish or Zaza. **Reference category is Kurdish origin university graduates with the highest level of vulnerability.
30
Figure 1. Predicted Probabilities of Different Levels of Informality, Household Sample
Non-Kurdish, Illiterate, lived in the city for 50% his life
Informal 5
Informal 2
Informal 4
Informal 3
Informal 6
31
Figure 2. Impact of Education on Predicted Probabilities of Different Levels of Informality, Household Sample without Kurdish background taken into account
0,22
0,55
0,82
0,26
0,12
0,21
0,51
0,270,27 0,27
0,29
0,04
0,14
0,32 0,32
0,28
0,16
0,0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
Illiterate Literate but noprimary school
diploma
Primary school Junior high school High school University +
Non-Kurdish, non-vulnerable, lived in the city for 50% his life
Informal 0
Informal 1
Informal 2
32
Figure 3. Impact of Education on Predicted Probabilities of Different Levels of Informality, Household Sample with Kurdish background taken into account
0,25
0,57
0,18
0,25
0,13
0,070,07
0,10
0,22
0,13
0,14
0,27 0,28
0,33 0,330,35
0,320,30
0,0
0,1
0,2
0,3
0,4
0,5
0,6
Illiterate Literate but noprimary school
diploma
Primary school Junior high school High school University +
Kurdish, non-vulnerable, lived in the city for 50% his life
Informal 0
Informal 2
Informal 1Informal 3
33
Appendix1: Sampling Methodology
The nature of the informal economic activity impedes one to set a well-defined target population
from which to sample. Rather we chose to focus on the households for which there are well-tested
sampling procedures in Turkey.32 This sampling procedure first creates regions of provinces on the
basis of socio-economic criteria and allocates target populations across these regions. We chose to
focus on urban populations in our work and thus created regions according to urban populations of
provinces. Then the total sample size is distributed across these regions according to their shares of
the total urban population. Then from within these regions representative provinces are chosen
according to their respective target population sizes. Once the provinces within each region and
their respective sample sizes are determined then districts and neighborhoods within each chosen
district is selected in a similar fashion. Within each neighborhood streets are randomly selected.
However, at this stage since the number of households in each street is not known the principle of
allocating the same equal probability of being chosen in the sample is broken. Each street is given
the same weight and equal numbers of interviews are conducted in each chosen street.
We followed the same principles in selecting our urban sample. We then adopted these
principles in the selection of our samples for three additional target populations. One of these
populations was the shopkeepers, the other was bazaar tradesmen and finally the street vendors. For
the shopkeepers we took the selected streets as our starting point. In every street we aimed at
interviewing one shopkeeper. In every street a list of shops were made and one shop is selected
randomly. Since most streets would predominantly have similar types of shops we also restricted
our interviewers to reach different types of shops in every street. In other words, we did not want to
obtain data only on grocery stores, sellers of fruits and vegetables, pharmacies and the like, which
can be found on most streets. If in one street in a neighborhood yields one type of shop for
interview in the other street a different type is selected if it at all existed.
34
For the bazaars we took the chosen districts for our household survey as our starting point
and went to the municipality administrations in these to obtain a full list of neighborhood bazaars.
In every bazaar we aimed at reaching at most 10 stall-keepers. Half of these interviews are then
reserved for fruit, vegetables etc. and the other half is reserved for sellers of dried or manufactured
foodstuff, textiles and other household items.
We followed a similar procedure for selecting the street vendors. The total numbers of
interviews are first distributed to the selected districts according to their population shares.
Obviously street vendors are not present in every street in every neighborhood. Rather they cluster
in specific central localities. Thus in every district locations wherein street vendors are expected to
be present are determined. Then we applied a quota for selecting the vendors. At most one third of
the interviews could be of one type in any given district.
As such our samples are not random in the sense of making the probability of selecting
every targeted unit equal to one another but they are comparable to one another since their socio-
economic environments are all controlled for in our procedure. All four samples come from same
localities and are comparably selected. A total of 1232 interviews were completed with the
household, 619 with shopkeepers, 306 with bazaaris and 327 with the street vendors.
35
Appendix2: Survey questions INFORMALITY Do you have anyone of the following insurance schemes (SSK, Bag-Kur, Retirement Fund etc)? 1> Yes, I do 2>No, I do not How frequently do you collect receipts in your shoppings for the house?
1>Never 2>Sometimes 3>Frequently 4>All the time How much of your due taxes do you actually pay? Not paying Paying taxes any taxes fully 1 2 3 4 5 6 7 8 9 10 If someone in the family becomes ill, how would you meets, the associated costs such as doctors, check-ups, medicine and hospital? 1) We would use the insurance of one of the family member (SSK; Bag-Kur, Retirement Fund etc) 2) We would use green card 3) We would pay out of our own pocket 4) We would borrow from our relatives 5) We would borrow from friends 6) Employer helps
Other Which of the following items do you or your wife/husband have/own?(answer obtained for each item) a) Mobile phone b) Bank account c) Credit card When you buy goods from a producer or a wholesaler, do you ever buy with a credit,(long term payment)?
1>Yes, I do 2>No, I do not (If he/she buys on credit) What is the maximum length of time can you buy on credit? (open ended) When you buy goods on credit with long term payment, how do you do that?
1) Shaking hands and getting an oral promise 2) Making them sign a promisory note 3) Having them sign a check with a longer term date Other ( writing in a book, creating an open account, credit cards)
What is the percentage of the goods and services that you buy on credit in our entire purchase? (open ended) In your purchases of goods and services for your business, how much of your payments do you do in cash, and/or check/note and/or credit card? Cash (open ended) Checks/Notes (open ended) Credit cards (open ended) How much of the payments your customers make is done in cash, checks/notes and credit cards? Cash (open ended) Checks/notes (open ended) Credit cards (open ended) How much of the goods you buy for your business is without any receipts? (open ended) What is the percentage of your shopping customers who go wihout receipts? (open ended) What is the percentage of customers who bargain with you? (open ended) How much of a price reduction would you agree as a result of this bargaining? (open ended)
36
(Asked only to shopkeepers) Do you ever sell with long term payment to your customers?
1>Yes, I do 2>No, I do not If yes, what the maximum length of time you would allow customers to have long term payment? (open ended) What is the percentage of sales with long term payment in your total sales? (open ended) Whom do you sell goods on long term payment the most?
a) permanent customers b) old customers I like c) If the market is bad, almost to everyone d) To those who come on the recommendation of friends and colleagues Other
How do you sell your customers on a long term payment basis? 1> Shaking hands and getting an oral promise 2> Making them sign a promisory note 3> Having them sign a check with a longer term date Other ( writing in a book, creating an open account)
(Only to the household members) During the previous month, have you ever gone to a bazaar?
1>Yes, I did 2>No, I did not If yes, how many times did you go? (open ended) How much of the total household expenditures is spent in the bazaars? (open ended) Have you ever shopped from a street vendor during the previous month?
1>Yes, I did 2>No, I did not If yes, how many times? (open ended) How much of your total household expenditures is spent with the the street vendors? (open ended) VULNERABILITY QUESTIONS If you take into account the previous six months, could you tell us which one of the following range is closest to your total average monthly household income including wages, rents, pensions etc of all the members living in your house?
With this income to what extent are you able to meet your household needs? Not at all Meet them fully 1 2 3 4 5 6 7 8 9 10 If your income was cut off for some reason, how long would you be able to manage? (open ended) At the end of this time, whom would ask for help?
1. Family members living outside home 2. Relatives 3. Neighbors and those living in the neighborhood 4. Friends 5. Hemseris (local friends coming from the same original cities) Other (Noone, ask from state, look for a new job)
37
Do own the house you are living in, somebody owns buy you don’t pay any rent, you live in employer/state financed housing or do yo pay rent?
1. I own 2. House belongs to another but I dont pay any rent 3. Live in employer or state financed housing 4. Pay rent
Do you or your wife/husband own a car? If you have one, can you tell us its brand and model? ( if more than one, please write the more expensive car)
1) I Don’t have any car 2) Owns a commercial vehicle 3) Lower brand and models 4) Normal brand and models 5) Luxury brands and models
Which one of the following items exist in your house and/or owned by a household member? 1) Exists 2)Does not exist House phone (fixed line) Washing machine Dish washing machine Computer Microwave oven Satellite dish Air conditioner Summer house
38
REFERENCES Adaman, F, Çarkoglu A. and Senatalar, B. (2005).Toplumun Kamu Yönetimine, Kamu Hizmetlerine ve Reforma Bakisi (Turkish Society’s Evaluation of Public Administration, Public Services and Reform). Istanbul: Türkiye Ekonomik ve Sosyal Etüdler Vakfi (TESEV) publications. Adaman, F, Çarkoglu, A. and Senatalar, B. (2002). Household View on the Causes of Corruption in Turkey and Suggested Preventive Measures. Istanbul: TESEV Publications Aktar, A. (1970). Kapitalizm, Az Gelismislik ve Türkiye’de Küçük Sanayi (Kapitalizm, Underdevelopment and Small-scale production). Istanbul: AFA. Atauz, S. and Atauz, A. (1992). Enformel sector, Kentsel Isgücü Pazarlari, Sosyal ve Ekonomik Yapilanmalar Üzerine Betimsel Tartismalar (Informal sector, urban labor markets and discussions on social and economic structuring), Planlama 9.(2/1-4), 4-21. Ayata, S (1987). Kapitalizm ve Küçük Üreticilik: Türkiye’de Hali Dokumaciligi (Capitalism and small-scale production: Carpet Weaving in Turkey). Ankara: Yurt Yayinciligi. Ayata, S. (1991). Industrial Change and Social Stratification in the urban context. In M. Kiray ed. Structural Change in Turkish Society, Bloomington: Indiana University Press. Bairoch, P. (1973). Urban Employment in developing Countries: The nature of the problem and proposals for its solution. Geneva: International Labor Organization, ILO. Bulutay, T. (1995). Child Labor in Turkey. Geneva and Ankara: ILO and SIS. Charmes, J (1998). Informal Sector, Poverty and Gender: A Review of Empirical Evidence. Washington DC: The World Bank. Chen, M and Carr, M (2001). Globalization and the Informal Economy: How global trade and investment impact on the working poor. Women in Informal Employment globalizaing and organizing WIEGO, http://www.wiego.org/papers/carrchenglobalization.pdf Chen, M. (2005). “Rethinking the Informal Economy” EDGI/WIDER Research Paper No. 2005/10 http://www.wider.unu.edu/publications/rps/rps2005/rp2005-10.pdf Çinar, E.M.(1994). Unskilled urban migrant women and disguised employment: Home working women in Istanbul-Turkey. World Development, 22(3), 369-380. De Soto, H. (1989). The Other Path: The informal revolution. New York: Harper and Derdiyok, T. (1993). Türkiye’de Kayitdisi ekonominin tahmini. (The forecast of informal economy in Turkey). Türkiye Iktisat, 14; 54-63 (May). Eder, M. (director), Yakovlev,A., Çarkoglu, A., and Chaudry, K. (2002). Redefining Contagion: Political Economy of Suitcase Trade between Turkey and Russia. IREX Project, Final Report, August. Erder S. (1996). Istanbul’da bir Kent Kondu: Ümraniye. (A city shanty town in Istanbul: The case of Ümraniye). Istanbul: Iletisim Yayinlari.
39
Hart, K. (1973). Informal Income Opportunities and Urban Employment in Ghana. Journal of Modern African Studies, 11(1), 61-89. ILO (1972). Employment, Incomes and Equality: A Strategy for Increasing Productive Employment in Kenya. Geneva: ILO. Kartal, S.K (1992). Ekonomik ve Sosyal Yönleriyle Türkiye’de Kentililesme. (Urbanization in Turkey with it social and economic dimensions). Ankara: Adim Yayincilik. Kasnakoglu, Z. and Yayla, M (1998). Türkiye’de Kayitdisinin boyutlari (The size of informal economy in Turkey). Istihdam ve Egitim Projesi Piyasasi Bilgisi. Ankara: State Institute of Statistics, SIS now TUIK. Kiray, M (1982). Metropoliten Kent Olgusu Toplumbilim Yazilari. (Metropolitan City Phenomenon: Sociology essays). Gazi Üniversitesi, Ankara: IIBF no.7. Köksal, S.E and Lordoglu, K. (1993).Geleneksel Çirakliktan Çocuk Emegine bir alan arastirmasi. (From traditional apprenticeship to child labor: A field research). Istanbul: Frederich Ebert Foundation. Long, S. J. (1997). Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage Publications. Maloney, W. (2004) Informality revisited. World Development, 32(7),1159-1178. Özar, S. (1996). Kentsel Kayitdisi Kesimde Istihdam Sorununa Yaklasimlar ve bir Ön saha çalismasi. (Approaches to the employment problem in the urban informal sector and work on a preliminary fieldwork).Unpublished paper. Istanbul Özbay, F. (1990). Kadinlarin eviçi ve evdisi ugraslarindaki degismeler” (Changes in women’s affairs inside and outside the home. In Sirin Tekeli (Ed.), Kadin Bakis açisindan 1980’lerde Türkiye’sinde Kadin. (Women in the 1980s Turkey: A women perspective). Istanbul: Iletisim Yayinlari. Portes, A. and Schauffler, R. (1993). Competing perspective on the Latin American Informal Sector. Population and Development Review, 19(1), 33-60 Portes, A. Castells, M and Benton, L.A. (Eds.), (1989). Informal Economy: Studies in Advanced and Less Developed Countries. Baltimore and London: University of John Hopkins Press. PREALC (ILO’s Regional Employment Program, known with its Spanish acronym) (1981). Sector informal: Funcionamento y politicas. Santiago de Chile: PREALC. Senyapili, T. (1992). A new stage of Gecekondu Housing in Istanbul. In I. Tekeli, T. Senyapili, A. Türel and E. Acar (Eds.), Development of Istanbul Metropolitan Area and Low Cost Housing. Istanbul: Turkish Social Science Foundation.
40
Temel A., Simsek, A. and Yazici K. (1994). Kayitdisi Ekonomi, Tanimi, Tespit Yöntemleri ve Türk Ekonomisindeki Büyüklügü (Informal Economy, Its definition, measurement methods and its size in Turkish economy). Ankara: State Planning Organization. Thomas, J. (1993). Informal economic activity. Michigan: Michigan University Press. Tokman, V.E. (1989). Policies for a heterogeneous informal sector in Latin America. World Development, 17(7),1067-1076. Yenal, H. D. (2000). Weaving the Market: The informal economy and gender in a transnational trade network between Turkey and the Soviet Union Unpublished Dissertation. Binghamton University. Binghamton: State University of New York.
41
END NOTES
1 The term bazaari is used differently in the Turkish context. A bazaari in our sample does not refer
to an individual who has a fixed shop in a huge city bazaar as would the case be in many developing
countries. We refer to bazaaris as stall holders in open-air neighborhood bazaars. The municipality
allocates a fixed area for certain days of the week so that bazaaris can open their stalls.
2 Intentional informality is very difficult to differentiate from one that follows from necessity. We
leave this distinction intentionally out of our analysis for the purposes of this article.
3 De Soto (1989).
4 See Bairoch (1973).
5 See Portes and Schauffler (1993) p.38.
6 See Hart, (1973).
7 PREALC (1981).
8 De Soto (1989).
9 See Maloney (2004).
10 See Portes and Schauffler (1993).
11 See International Labor Organization ILO (1972).
12 See Thomas (1993) p.5.
13 See, for instance, Erder (1996), Kartal (1992), Kiray (1970, 1981), and Senyapili (1992).
14 See, for instance, Aktar, (1990), Ayata (1987, 1991).
15 For good examples see Çinar (1994), Özbay (1990), Senyapili (1992).
16 See for example Bulutay (1995) and Köksal and Lordoglu (1993).
17 See Atauz and Atauz (1992), Özar (1996).
18 See Eder et al (2002) and Yenal (2000) for further elaboration.
19See, Derdiyok (1993), Kasnakoglu and Yayla (1998), Temel, Simsek and Yazici (1994).
42
20 In fact, this percentage directly parallels Turkish Statistics Institution (TUIK)’s data on 2000
health expenditure statistics where the public resources account for only 62 percent of the total
health expenditures in the country. See www.tuik.gov.tr
21 The government has recently eliminated the requirement for the pensioners to file for the three
monthly VAT tax returns. The VAT (KDV in Turkish) rates vary between 1% - 18% but with the
exception of basic staples it is generally applied as 18%. VAT is payable on all local purchases.
22 76% of the households bargain when they shop receiving an average of 11% reduction as a result.
23 When ask whom they would ask for help after depleting the emergency savings, a significant
group answered as “nobody”. This reflects lack of any security network backing. Among the street
vendors, 22% indicated that they would not ask anybody for help. Among shopkeepers this ratio is
only half while for bazaaris it is 8.5% and amongst the household respondents about 7%.
24 These numbers are considerably higher than the poverty and economic vulnerability
measurements of TUIK primarily because TUIK measurements are based on household
consumption surveys rather than income. Economic vulnerability is measured based on the local
cost of basic needs basket including non-food. The numbers on economic vulnerability range
between 26-36%. See www.tuik.gov.tr
25 Although we count the number occurrences that we take as representations of informality for our
respondents the differences between these counts can not be assumed to be equal. Although a given
respondent scoring 0, therefore no informality according to our index, has lower level of
participation in informal activities than someone scoring 1, we do not know how high the
informality of that individual scoring one is. Moreover, we also do not know whether the larger
informality of someone scoring 4 as opposed to 3 is of equal magnitude comparing someone who
got 1 as opposed to 0. In other words, equal differences in our scores do not necessarily correspond
to equal differences in the underlying informality of our respondents.
26 See Long (1997) pp.114-148, for an accessible treatment of the analysis of ordered outcomes.
43
27 Shopkeepers, bazaaris and street vendors in Turkey are overwhelmingly male. In our household
sample, the gender differences do not emerge as a significant determinant of informal behavior.
Most questions, however, were asked on household basis rather than gender, which is why the
survey would not capture the differences in income or occupation.
28 Determining one’s Kurdish origin is difficult. For respondents reporting to have spoken with their
parents in their childhood in Kurdish, or Zaza, we attributed a Kurdish ethnic background.
29 See for instance Hart (1970).
30 Due to space constraints we do not show similar figures for our other three samples but results
therein concerning the expected level of informality at zero level of economic vulnerability is
similar to the one reported about the household sample.
31 See Carr and Chen (2001), and Charmes (1998).
32 For details of these procedures, see Adaman, Çarkoglu and Senatalar (2005, 2002).