When Does Transparency Improve Public Goods Provision? Street-level Discretion, Targetability, and Local Public Service Delivery

Transcription

1 When Does Transparency Improve Public Goods Provision? Street-level Discretion, Targetability, and Local Public Service Delivery Monika Bauhr and Ruth Carlitz University of Gothenburg, Sweden Workshop on Systems of Public Service Delivery in Developing Countries 15th May / 12

2 Motivation Considerable variation within countries when it comes to accessing public goods and services 2 / 12

3 Motivation Considerable variation within countries when it comes to accessing public goods and services Patterns of access also vary among different aspects of service delivery (Kramon and Posner 2013; Batley and Mcloughlin 2015) 2 / 12

4 Motivation Considerable variation within countries when it comes to accessing public goods and services Patterns of access also vary among different aspects of service delivery (Kramon and Posner 2013; Batley and Mcloughlin 2015) What is the role of transparency in explaining such differences? 2 / 12

5 Transparency and Public Goods Provision Increasing attention to political institutional factors (not only economic) as potential determinants of variations in access to public services (i.e. Lieberman 2014) 3 / 12

6 Transparency and Public Goods Provision Increasing attention to political institutional factors (not only economic) as potential determinants of variations in access to public services (i.e. Lieberman 2014) Increased government transparency widely expected to improve the quality of public service delivery: derived from the logic of PA theory: reduces information asymmetries and allows for monitoring 3 / 12

7 Transparency and Public Goods Provision Increasing attention to political institutional factors (not only economic) as potential determinants of variations in access to public services (i.e. Lieberman 2014) Increased government transparency widely expected to improve the quality of public service delivery: derived from the logic of PA theory: reduces information asymmetries and allows for monitoring Empirical findings on effects of transparency on improved access to service delivery are surprisingly mixed (Joshi 2013; Bauhr and Grimes 2014; Chong et al 2015; Malesky et al 2012; Björkman and Svensson 2012; Keefer and Khemani 2012; Fox 2015) 3 / 12

8 Transparency and Street-Level Discretion Targetability of public services as a political tool 4 / 12

9 Transparency and Street-Level Discretion Targetability of public services as a political tool Physically excluding citizens from accessing public services can be very visible and costly (i.e. road blocks, preventing children from attending school) 4 / 12

10 Transparency and Street-Level Discretion Targetability of public services as a political tool Physically excluding citizens from accessing public services can be very visible and costly (i.e. road blocks, preventing children from attending school) Street level discretion allows local officials to more subtly target services to particular individuals within a community 4 / 12

11 Transparency and Street-Level Discretion Targetability of public services as a political tool Physically excluding citizens from accessing public services can be very visible and costly (i.e. road blocks, preventing children from attending school) Street level discretion allows local officials to more subtly target services to particular individuals within a community However, creates information asymmetries between principals (central government officials or citizens) and agents (local government officials) 4 / 12

12 Transparency and Street-Level Discretion Targetability of public services as a political tool Physically excluding citizens from accessing public services can be very visible and costly (i.e. road blocks, preventing children from attending school) Street level discretion allows local officials to more subtly target services to particular individuals within a community However, creates information asymmetries between principals (central government officials or citizens) and agents (local government officials) Transparency may redress these information asymmetries 4 / 12

13 Hypothesis Transparency has a stronger effect on the quality of service provision in sectors where street level discretion is high than in sectors where street level discretion is lower 5 / 12

14 Empirical Strategy and Data Data from the Vietnam provincial Governance and Public Administration Performance Index ( PAPI), / 12

15 Empirical Strategy and Data Data from the Vietnam provincial Governance and Public Administration Performance Index ( PAPI), I 14,000 randomly selected respondents in 207 districts, 414 communes and 828 villages 6 / 12

16 Empirical Strategy and Data Data from the Vietnam provincial Governance and Public Administration Performance Index ( PAPI), I 14,000 randomly selected respondents in 207 districts, 414 communes and 828 villages I Studying non-democratic setting means we can distinguish transparency from democracy 6 / 12

17 Empirical Strategy and Data Data from the Vietnam provincial Governance and Public Administration Performance Index ( PAPI), I 14,000 randomly selected respondents in 207 districts, 414 communes and 828 villages I Studying non-democratic setting means we can distinguish transparency from democracy Dependent Variables: Quality of education, quality of health services, quality of infrastructure (access to improved water and roads) 6 / 12

18 Empirical Strategy and Data Data from the Vietnam provincial Governance and Public Administration Performance Index ( PAPI), I 14,000 randomly selected respondents in 207 districts, 414 communes and 828 villages I Studying non-democratic setting means we can distinguish transparency from democracy Dependent Variables: Quality of education, quality of health services, quality of infrastructure (access to improved water and roads) Independent Variables: Budget and poverty list transparency 6 / 12

19 Empirical Strategy and Data Data from the Vietnam provincial Governance and Public Administration Performance Index ( PAPI), I 14,000 randomly selected respondents in 207 districts, 414 communes and 828 villages I Studying non-democratic setting means we can distinguish transparency from democracy Dependent Variables: Quality of education, quality of health services, quality of infrastructure (access to improved water and roads) Independent Variables: Budget and poverty list transparency Control Variables: Self-reported economic conditions, level of education, gridded population data, nightlights data (to proxy economic growth) 6 / 12

20 Empirical Strategy and Data Data from the Vietnam provincial Governance and Public Administration Performance Index ( PAPI), I 14,000 randomly selected respondents in 207 districts, 414 communes and 828 villages I Studying non-democratic setting means we can distinguish transparency from democracy Dependent Variables: Quality of education, quality of health services, quality of infrastructure (access to improved water and roads) Independent Variables: Budget and poverty list transparency Control Variables: Self-reported economic conditions, level of education, gridded population data, nightlights data (to proxy economic growth) Commune level fixed effects regressions (and pooled OLS regressions) 6 / 12

21 Vietnam Provinces 7 / 12

22 Vietnam Provinces and Communes 8 / 12

23 Vietnam Population (2015 WorldPop) 9 / 12

24 Vietnam Night Lights (2013) 10 / 12

25 Results Commune-level increases in transparency positively associated with commune-level increases in quality of education 11 / 12

26 Results Commune-level increases in transparency positively associated with commune-level increases in quality of education Also see positive association with commune-level increases in quality of health services 11 / 12

27 Results Commune-level increases in transparency positively associated with commune-level increases in quality of education Also see positive association with commune-level increases in quality of health services No association between changes in commune-level transparency and access to improved water or roads 11 / 12

28 Results Commune-level increases in transparency positively associated with commune-level increases in quality of education Also see positive association with commune-level increases in quality of health services No association between changes in commune-level transparency and access to improved water or roads Similar results in pooled, cross-sectional regressions 11 / 12

29 Results Commune-level increases in transparency positively associated with commune-level increases in quality of education Also see positive association with commune-level increases in quality of health services No association between changes in commune-level transparency and access to improved water or roads Similar results in pooled, cross-sectional regressions Suggestive evidence in favor of our hypothesis that effects of transparency are conditional on degree of street-level discretion 11 / 12

30 Next Steps Need better understanding of Vietnam context - actors involved in service provision, incentives, contingencies, and feedback loops 12 / 12

31 Next Steps Need better understanding of Vietnam context - actors involved in service provision, incentives, contingencies, and feedback loops Delve into mechanisms linking transparency with improved service delivery explore other dimensions of PAPI data 12 / 12

32 Next Steps Need better understanding of Vietnam context - actors involved in service provision, incentives, contingencies, and feedback loops Delve into mechanisms linking transparency with improved service delivery explore other dimensions of PAPI data I Participation at local levels 12 / 12

33 Next Steps Need better understanding of Vietnam context - actors involved in service provision, incentives, contingencies, and feedback loops Delve into mechanisms linking transparency with improved service delivery explore other dimensions of PAPI data I Participation at local levels I Vertical accountability towards citizens 12 / 12

34 Next Steps Need better understanding of Vietnam context - actors involved in service provision, incentives, contingencies, and feedback loops Delve into mechanisms linking transparency with improved service delivery explore other dimensions of PAPI data I Participation at local levels I Vertical accountability towards citizens I Control of corruption in the public sector 12 / 12

35 Next Steps Need better understanding of Vietnam context - actors involved in service provision, incentives, contingencies, and feedback loops Delve into mechanisms linking transparency with improved service delivery explore other dimensions of PAPI data I Participation at local levels I Vertical accountability towards citizens I Control of corruption in the public sector I Public administrative procedures 12 / 12

36 Next Steps Need better understanding of Vietnam context - actors involved in service provision, incentives, contingencies, and feedback loops Delve into mechanisms linking transparency with improved service delivery explore other dimensions of PAPI data I Participation at local levels I Vertical accountability towards citizens I Control of corruption in the public sector I Public administrative procedures Your suggestions?? 12 / 12