Wages VS Informal Payment in police: comparative study of Russia, Bulgaria, Kazakhstan and Latvia

1. Tatiana Karabchuk , Ruslan Almuhametov National Research University Higher School of Economics tkarabchuk@hse.ru Wages VS Informal Payment in police: comparative study of Russia, Bulgaria, Kazakhstan and Latvia LSES workshop, Moscow 2013

2. Introduction • Problem and motivation • Research goal and tasks • Literature review • Data and methodology • Results • Conclusions

3. 1. Problem and motivation • Fist stage analysis, bases on collected interviews, showed that there is a definite problem with ineffective wage setting in the police system in all four countries. • Policy makers, public authorities and police officers spoke about the same problem – low wage rates of policemen, reforms started (in Russia and Kazahstan). How are wages determined in Police? • From broad literature and mass media we know that both conventional and unconventional economic activities take place in police. Do low wages (non-satisfaction with payment) stimulate the growth of informal payments among policemen?

4. 2. Goals and tasks The main aim of the research is to disclose wage factors in 4 countries and identify the impact of inefficient wage on informal payment/activity (corruption approval). The tasks are: • To trace the wage distribution among policemen and disclose its determinants. • To reveal the impact of legal wage size on the attitudes toward the informal activity (bribe-taking).

5. 3. Literature review Two blocks of literature could be applied: 1) The first block contains papers on wage setting, wage factors, wage distribution as well as fair payments. This part of literature is much bigger and dates to 1950s 2) The second block comprises from all sort of informal economic behavior of policemen: conventional and unconventional. This part of literature is less thicker and dates to early 1960s

6. 3. Literature review Wage efficiency and rent seeking theories Corruption and moonlighting behavior Becker , Journal of Political Economy,(1968) Beck, Paul J., and Michael W. Mahe, Economic Letters, (1986) Bayley and Shearing, Law & Society Review, (1996) Braguinsky, Contemporary Economic Policy, (1996) Tanzi, Staff Papers - International Monetary Fund (1998) Anon , Jakarta Post, (2004) Ayling, and Grabosky, International Journal of Police Strategies and Management, (2006) Ayling and Shearing, Criminology and Criminal Justice, (2008) • Bhagwati and Srinivasan, Journal of Political Economy, (1980). • Shapiro and Stiglitz, American Economic Review, (1984) • Appelbaum and Katz, The Economic Journal, (1987). • Brown, Industrial and Labour Relations Review, (1990) • Nickell and Nicolitsas, Labour Economics, (1997) • Campbell and Kamlani, (1997) • Ichino and Maggi, (2000) • Lazear, The American Economic Review, (2000) • Chang and Ching-chong Lai,The Scandinavian Journal of Economics, (2002)

7. 3. Literature review Empirical research on police formal andinformal payments is not so rich. There are several country studies that demonstrates very interesting and different results. • Bayley, 1996 (The Police and Political Development in Europe) • Zvekic, 1998 (Countries in transition) • Darryl D'Monte, 2000 (Asian societies and corruption) • Beck & Ruth ,2002 (Attitudes to Corruption Amongst Russian PoliceOfficers and Trainees) • Tymoty Frye, 2002 (Police as an obstacle for business) • Shleifer & Treisman, 2004 (Normal Country) • Robertson, 2004 (Police reform in Russia) • Wallace and Latcheva, 2006 (Corruption, Trust in Public Institutions and the Informal Economy in Transition Countries of Central and Eastern Europe) • Ayling and Shearing, 2008 (Australia case of police corruption) • Gerber, Mendelson, 2008 (Russian case of publicabuse by police)

8. Wage setting in police (budget sector) • Wages in police are regulated by the state and budget sector rules. • The wage of a policeman usually consists of: • Formal basic payment according to the position and ranking; • Benefits for number or years; • Benefits for scientific position and rank; • Additional payments (surplus) for good work, difficulties and etc. The hypotheses for the wage determinants are coming from Mincerian wage equation, human capital theory, theory of compensating differences, budget sector mechanism

9. 4. Data and methodology Data: Surveys of policemen conducted in 2011-2012 in Bulgaria, Russia, Kazakhstan and Latvia (each country sample consists of 450-500 respondents). Empirical results: • Descriptive statistics of samples • Wage and informal activity indicators • Wage determinants (OLS regression for 4 countries) • Bribery approval probability (marginal effects of probit regressions for 4 countries)

10. 5. Results: descriptive statistics

11. 5. Results: wage and informal activity indicators

12. Methodology : wage factors Dependent variable: lnwage (corrected for minimum living spending in the region and for dollars PPP) Independent variables: • Gender • Tenure in police • Tenure in police squared • Higher education • Working hours • Working experience (3 dummies) • Number of years living in the city • Level of position (3 dummies) • Supervision of people (4 dummies) • Departments (dummies )

13. 5. Results: lnwage determinants

14. 5. Results: lnwage determinants

15. Family well-being during the last year

16. Corruption approval, probability to take informal payments Hypotheses: • The bigger the family and the poor the family is more chances for approving corruption • The higher the real wage the less policemen approve bribery (wage satisfaction theory) • The higher the increase in wages (within two years of reforms) the less policemen approve bribery • The bigger the difference between real and ideal wage the more policemen approve bribery (A.Mas (2006) “Pay, Reference Points, and Police Performance”)

17. Probit regressions on attitudes and thinking of corruption in police (dependent dummy variables)

18. Probit regressions on attitudes and thinking of corruption in police • Independent tested variables: • Size of family • Family well-being • Ln_wage • Wage increase for 2011-2010 • Ln_wage difference ideal-real • Controlls: • Gender • Tenure in police/Tenure in police squared • Higher education • Past working experience • Number of years living in the city • Rank-position (3 dummies) • Supervision (4 dummies) • Departments (4 dummies) • Country • Dependent variables: • Bribery acceptance by majority of your department. • Negative attitude towards additional (informal) work of policemen. • Saying that the price for informal help is defined. • Proxy for bribery approval.

19. Marginal effects from probit regression for thinking of bribery acceptance by majority of the department

20. Marginal effects from probit regression for negative attitude towards additional (informal) work of policemen

21. Marginal effects from probit regression for saying that the price for informal help is defined

22. Marginal effects from probit regression for bribery declaration approval (proxy)

23. 6. Conclusions on wage determinants and wage differences Themain determinants for higher wages in police are: • gender differences, • tenure (+), tenure2 (-), • Return to Education only in Bulgaria and positive effect of working hours only for Latvia. • Return to high rank in all countries, • Return to supervision only in Bulgaria and Latvia, • Regional difference in Russia and less in Kazahstan • Department differences in Bulgaria and Latvia Policemen are better paid in Bulgaria in comparison to Russia and less paid in Latvia and Kazakhstan Almost no real changes in wages in Russia, growth in wages in Kazakhstan , decrease in real wages in Latvia and Bulgaria

24. 6. Conclusions on corruption approval and probability of informal payment Opportunities for additional work havedecreased during the last 2 years for: 33.5% in Latvia 35% in Russia 50% in Kazakhstan 70% in Bulgaria. • The poor family budget positively affects corruption approval from policemen • The hypothesis about the direct effect of wage size proved to be true only for Russia • The hypothesis about wage increase within the two years of reforms was not confirmed for all countries. • The hypothesis of positive effect of real-ideal wage difference was shown to be true only for Russia. What about other countries?

25. 6. Conclusions next step for research ??? Values, culture, social norms???

26. Thank you for your attention

27. Additional slide of questions • 24. Какова Ваша заработная плата в полиции: сколько примерно денег вы получали на руки (включая премии, доплаты и надбавки, которые Вам платят в полиции) в среднем в месяц за 2011 год? • 28. Как Вы считаете, сколько примерно должен получать человек на вашей должности? • 30. Как относятся к дополнительным заработкам сотрудники Вашего подразделения? • 38. Насколько приемлемой большинство Ваших коллег в подразделении сочтет следующую деятельность (позицию)? Полицейский получил взятку • 39. Предположим, Вы узнаете, что один из Ваших коллег сообщил в службу внутренней безопасности о коррупции в вашем подразделении, как Вы к этому отнесетесь?