## _wp12231 - References

## Source details

**Canonical URL:** [_wp12231 - References](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12231.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12231.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12231.pdf.json)

---

### I. Introduction — purpose and main contribution
- Objective: systematically document the human capital input into public administrations around the world by building a novel dataset on education attainment in public administrations.
- Scope of dataset: period 1981–2011 for 178 countries, using information extracted from CVs for over 131,877 mid to senior level officials (mainly central banks and ministries of economy and finance).
- Main empirical finding:
  - Little heterogeneity across regions when considering a non quality-adjusted measure of education attainment in public administrations.
  - Important cross-regional heterogeneity emerges once education is quality-adjusted by a country academic ranking.
- Additional contributions:
  - Documentation of patterns by gender and seniority across regions.
  - Exploratory evidence that:
    - Salary incentives matter for attracting highly educated staff.
    - Higher education attainment in public administrations is positively associated with government effectiveness (tax revenue mobilization, limiting corruption, public finance management, and private market support).

### II. Data — sources, sample, representativity
- Source and structure:
  - IMF Institute for Capacity Development’s Participant and Applicant Tracking System (PATS) repository of CVs.
  - Tabulated fields include country of residence, agency, age, gender, position and detailed educational background.
- Sample size and coverage:
  - 131,877 CVs of civil servants who applied to IMF training activities during 1981–2011 from 178 countries.
  - For expositional purposes, last ten years use: 41,019 unique applicants. Data is available at five year intervals during 1981–2011.
  - On average about 135 observations per country over the last ten years.
- Representativity considerations and diagnostics:
  - Age distribution skewed right; median is around thirty years. Age groups 20 to 60 are represented. Regional exception: South Asia and the Middle East where applicants are somewhat older.
  - Male to female ratio in sample is roughly 2 to 1; females are underrepresented.
  - Two potential selection biases discussed:
    - IMF selection of participants: mitigated because the sample comprises all applicants, not only selected participants.
    - Domestic authority nomination: explored by comparing headquarter activities (spontaneous applications) to non-headquarter activities (mostly nominated). Age distributions of the two subsamples are virtually identical, suggesting little bias from nomination requirement.

### III. Measures of education attainment in public administrations
- Two measures constructed:
  1. Non quality-adjusted measure (years of tertiary education)
     - Encoding: high school = 0; PhD = 9 (irrespective of where awarded).
     - Aggregation: simple average of numerical values by country and region.
     - Key empirical observation: ratio of officials with a higher education degree is on average close to 1 for all regions (Figure 7).
     - Substantial divergence from general population education attainment: e.g., Figure 8 shows large ratios for regions such as Sub-Saharan Africa and Latin America.
     - Result: little differences in average years of tertiary education across regions (Figure 9).
  2. Quality-adjusted measure (weighted by country academic ranking)
     - Ranking source: a Shanghai University League Table based country-wide academic ranking computed by Universitas; ranking of 48 countries used.
     - Ranking notes: the ranking complements the Shanghai score with other measures of academic performance and also uses unemployment rate for individuals holding a higher education degree (see Appendix).
     - Construction:
       - Weighted average of years of tertiary education where weight = country j academic score for the country where individual k studied.
       - The product (years × academic score) is divided by maxγ (effectively 9) to normalize to 1, then multiplied by a weight between 0 and 1 so final measure lies between 0 and 1.
       - Minimum = 0 (all officials high school only); maximum = 1 (all officials US PhD).
     - Ranking example: United States ranks first, the United Kingdom second, Canada third and Indonesia forty eighth.
     - Empirical implications:
       - Adjusted measure is more heterogeneous across regions than the non-adjusted measure (Figures 11 and 12).
       - East Asia ranks second behind advanced countries in adjusted measure, followed by Central and Eastern Europe; Central Asia ranks last in adjusted measure.
       - Within-region variability matters: e.g., Sub-Saharan Africa exhibits extremely high variability within region compared to its average (Figure 13).

- Seniority and gender patterns (using adjusted measure)
  - Seniority:
    - Managers defined as pay grade from division chief upwards.
    - Managers are generally more educated than non-managers except in Middle East and North Africa (managers less educated) and Central and Eastern Europe (managers and non-managers similar) (Figure 14).
    - “Missing middle”: education attainment does not increase monotonically with seniority (Figure 15).
  - Gender:
    - Female education attainment is roughly on par with male officials, but female officials are far fewer (sample male:female ratio roughly 2:1).
    - Central Asia and Central and Eastern Europe are exceptions with near gender parity in employment.
    - Women are underrepresented among managers (Figure 17) and even more underrepresented among senior managers in all regions (Figure 18).
    - Implication: hired women are as educated as men, but fewer are hired and promoted to senior management.

- Agency comparison:
  - Central bank staff is not necessarily more educated than staff at ministries of finance and economies except in advanced countries (Figure 19).

### IV. Applications — incentives and government effectiveness (exploratory)
- Incentives (public pay gaps)
  - Cross-correlation results (Figure 20):
    - No statistically significant association between education attainment in public administrations and wage difference between public administration and manufacturing.
    - Positive and statistically significant association between education attainment in public administrations and wage difference between public sector and the financial sector.
  - Interpretation: better public sector pay relative to wages in finance is associated with higher education levels in public administrations, consistent with financial sector being the main outside option for civil servants in central banks and ministries.

- Government effectiveness
  - Tax revenue mobilization:
    - Positive and statistically and economically strong cross-correlation between education attainment and tax revenue over GDP (Figure 21).
    - Association remains positive and significant after controlling for GDP per capita and education in the general population (Figure 22).
  - Corruption:
    - Negative and statistically significant association between education attainment in public administration and an indicator of corruption based on PRS Group 2012 data (Figure 23).
    - Relationship remains negative when controlling for GDP per capita and education in the general population (Figure 24).
    - Suggestive interpretation: more educated public workforce may foster accountability and limit corruption; causality not established.
  - Public financial management and bureaucratic quality:
    - Positive and statistically significant association between education attainment and a composite public financial management indicator (World Bank 2012) (Figure 25).
    - Positive and statistically significant association between education attainment and bureaucratic quality indicator (PRS Group 2012) (Figure 26).
    - Findings robust to controls for GDP per capita and population education level.
  - Domestic financial sector standards:
    - Positive and statistically and economically significant association between education attainment in public administration and a composite index of domestic financial sector standards (regulation, supervision, competition...) (Figure 27).
    - Interpretation: a more educated public workforce may help raise domestic financial sector standards; resonates with concerns that regulatory bodies lag private firms in talent attraction and fall behind in keeping up with financial innovation.

### V. Conclusions and suggested future research directions
- Core conclusion: significant differences exist between education attainment in the general population and education attainment in public administrations; therefore, population-level education statistics are imperfect proxies for public administration human capital when studying state capacity.
- Dataset value: enables more direct study of the importance of human capital in the public sector and opens new research avenues.
- Key avenues for further work suggested:
  - Establish causality in the associations between education attainment in public administrations and government effectiveness indicators.
  - Document organizational structure and incentive systems in public administrations and how they interact with human capital at all seniority levels.
  - Build complementary datasets for lower-level officials (e.g., tax collectors) to assess the interplay between human capital across hierarchies and public administration performance.

### Appendix: Country Academic Ranking — Variables (Source: Universitas (2012))
- O1: Total number of journal articles produced by higher education institutions
- O2: Total articles produced per head of population
- O3: Average impact of articles
- O4: Weighted Shanghai ranking scores for universities per head of population
- O5: Shanghai scores for best three universities
- O6: Tertiary enrolment rates
- O7: Percentage of population over 24 with a tertiary qualification
- O8: Number of researchers in the nation per head of population
- O9: Unemployment rate of the tertiary educated compared with school leavers
- Source: Universitas (2012) http://www.universitas21.com/article/collaborations/details/105/measure-4-output

### Figures and Notes — key captions and data sources
- Figure 1. Education Attainment in the General Population around the World — average years of schooling from Barro and Lee (2010).
- Figure 2. Example of Official’s CV — template from the IMF Institute for Capacity Development Participants and Applicants Database (2012).
- Figure 3. Overview of the Distribution of Age in Sample — applicants to IMF training activities during 1991-2011.
- Figure 4. Age Distribution by Region — applicants to IMF training activities during 1991-2011 for different regions.
- Figure 5. Gender Distribution by Region — applicants to IMF training activities during 1991-2011 for different regions.
- Figure 6. Age Distribution by Type of Course — HQ vs Non-HQ applicants during 1991-2011.
- Figure 7. Fraction of Officials with Tertiary Education by Region — applicants during 1991-2011 for eight regions.
- Figure 8. Education Attainments: Public Administration vs. General Population — ratio of fraction with higher education degree (applicants) over same fraction for general population; Barro and Lee (2012).
- Figure 9. Average Years of Tertiary Education in Public Administrations — applicants during 1991-2011 for different regions.
- Figure 10. Fraction of officials with a degree from a country ranked in the top 48 — applicants during 1991-2011; Universitas data.
- Figure 11. Normalized weighted years of tertiary education by region — quality adjusted measure normalized between 0-1.
- Figure 12. Adjusted vs. non adjusted education attainment in public administration — both measures for different regions.
- Figure 13. Ratio of public administrations’ education attainment over associated variance — quality adjusted measure divided by standard deviation within region.
- Figure 14. Education attainment by rank and region — non managers (0) vs managers (1).
- Figure 15. Education attainment by seniority and region — seniority levels 2 to 5 defined in legend.
- Figure 16. Gender differences in education attainment in public administrations — adjusted measure by gender and fraction in sample.
- Figure 17. Fraction of Managers who are male — manager gender fractions.
- Figure 18. Fraction of senior managers who are male — senior manager gender fractions.
- Figure 19. Education Attainment by Agency and Region — ministries of finance and economics (1) vs central banks (2).
- Figure 20. Education attainment and Relative Public Sector Pay — cross-correlations with public/manufacturing and public/financial sector pay; civil servants’ pay data from Clements et al. (2010).
- Figure 21. Education attainment and tax collection — cross-correlations with tax revenue over GDP; tax data from Baunsgaard and Keen (2010).
- Figure 22. Education attainment and tax revenues controlling GDP per education and education — controls: GDP per capita and Barro and Lee (2012).
- Figure 23. Education attainment and corruption — corruption indicator from PRS Group (2012).
- Figure 24. Education Attainment and Corruption controlling for GDP per capita and Barro Lee — GDP per capita from Heston and Summers (2009); Barro and Lee (2012) for population education.
- Figure 25. Education Attainment and Public Sector Management — public sector management data from World Bank Development Indicators (2011).
- Figure 26. Education attainment and bureaucratic quality — bureaucratic quality data from PRS Group (2012).
- Figure 27. Education and Domestic Financial Sector Standards — financial liberalization data from Ostry (2009).
- Table 1. Country Level Academic Ranking — Universitas (2012) ranking and sub-indices Q1-Q9.

### References (selected list from source)
- Andrews, Matt, Lant Pritchett and Michael Woolcok, 2012, “Escaping Capability Traps through Problem-Driven Iterative Adaptation (PDIA),” Working Paper no. 299, Center for Global Development, Washington DC.
- Arezki A., A. Dupuy and A. Gelb, 2012, “Resource Windfalls, Optimal Public Investment and Redistribution: The Role of Total Factor Productivity and Administrative Capacity,” IMF Working Paper no. 12/200, International Monetary Fund: Washington DC.
- Barro Robert J. & Lee Jong-Wha, 2010, "A New Data Set of Educational Attainment in the World, 1950–2010," NBER Working Papers 15902, National Bureau of Economic Research, Cambridge.
- Baunsgaard, Thomas & Michael Keen, 2010, "Tax revenue and (or?) trade liberalization," Journal of Public Economics, Elsevier, October, Vol. 94(9-10), pp. 563–577.
- Botero, Juan, Alejandro Ponce & Andrei Shleifer, 2012, "Education and the Quality of Government," NBER Working Papers 18119, National Bureau of Economic Research, Cambridge.
- Clements B., Gupta S., Karpowicz I., and S. Tareq, “Evaluating Government Employment and Compensation. Technical Notes and Manuals,” Fiscal Affairs Department, International Monetary Fund.
- Josse Delfgaauw & Robert Dur, 2008, "Incentives and Workers' Motivation in the Public Sector," Economic Journal, Royal Economic Society, Vol. 118(525), pp. 171–191, 01.
- Ostry, Jonathan D., Alessandro Prati & Antonio Spilimbergo, 2009, “Structural Reforms and Economic Performance in Advanced and Developing Countries,” IMF Occasional Paper.
- Rama, Martin, 1999, "Public Sector Downsizing: An Introduction," World Bank Economic Review, Oxford University Press, January, Vol. 13(1), pp. 1–22.

*Source: _wp12231 - References .............................................................................................................*

### References .............................................................................................................

### _wp12231 - References .............................................................................................................

### I. Introduction — purpose and main contribution
- Objective: systematically document the human capital input into public administrations around the world by building a novel dataset on education attainment in public administrations.
- Scope of dataset: period 1981–2011 for 178 countries, using information extracted from CVs for over 131,877 mid to senior level officials (mainly central banks and ministries of economy and finance).
- Main empirical finding: little heterogeneity across regions when considering a non quality-adjusted measure of education attainment in public administrations; important cross-regional heterogeneity emerges once education is quality-adjusted by a country academic ranking.
- Additional contributions:
  - Documentation of patterns by gender and seniority across regions.
  - Exploratory evidence that (i) salary incentives matter for attracting highly educated staff and (ii) higher education attainment in public administrations is positively associated with government effectiveness (tax revenue mobilization, limiting corruption, public finance management, and private market support).

### II. Data — sources, sample, representativity
- Source and structure:
  - IMF Institute for Capacity Development’s Participant and Applicant Tracking System (PATS) repository of CVs.
  - Tabulated fields include country of residence, agency, age, gender, position and detailed educational background.
- Sample size and coverage:
  - 131,877 CVs of civil servants who applied to IMF training activities during 1981–2011 from 178 countries.
  - For expositional purposes, last ten years use: 41,019 unique applicants. Data is available at five year intervals during 1981–2011.
  - On average about 135 observations per country over the last ten years.
- Representativity considerations and diagnostics:
  - Age distribution skewed right; median is around thirty years. Age groups 20 to 60 are represented. Regional exception: South Asia and the Middle East where applicants are somewhat older.
  - Male to female ratio in sample is roughly 2 to 1; females are underrepresented.
  - Two potential selection biases discussed:
    - IMF selection of participants: mitigated because the sample comprises all applicants, not only selected participants.
    - Domestic authority nomination: explored by comparing headquarter activities (spontaneous applications) to non-headquarter activities (mostly nominated). Age distributions of the two subsamples are virtually identical, suggesting little bias from nomination requirement.

### III. Measures of education attainment in public administrations
- Two measures constructed:
  1. Non quality-adjusted measure (years of tertiary education)
     - Encoding: high school = 0; PhD = 9 (irrespective of where awarded).
     - Aggregation: simple average of numerical values by country and region.
     - Key empirical observation: ratio of officials with a higher education degree is on average close to 1 for all regions (Figure 7). Substantial divergence from general population education attainment: e.g., Figure 8 shows large ratios for regions such as Sub-Saharan Africa and Latin America.
     - Result: little differences in average years of tertiary education across regions (Figure 9).
  2. Quality-adjusted measure (weighted by country academic ranking)
     - Ranking source: a Shanghai University League Table based country-wide academic ranking computed by Universitas; ranking of 48 countries used.
     - Ranking notes: the ranking complements the Shanghai score with other measures of academic performance and also uses unemployment rate for individuals holding a higher education degree (see Appendix).
     - Construction: weighted average of years of tertiary education where weight = country j academic score for the country where individual k studied. The product (years × academic score) is divided by maxγ (effectively 9) to normalize to 1, then multiplied by a weight between 0 and 1 so final measure lies between 0 and 1. Minimum = 0 (all officials high school only); maximum = 1 (all officials US PhD).
     - Ranking example: United States ranks first, the United Kingdom second, Canada third and Indonesia forty eighth.
     - Empirical implications:
       - Adjusted measure is more heterogeneous across regions than the non-adjusted measure (Figures 11 and 12).
       - East Asia ranks second behind advanced countries in adjusted measure, followed by Central and Eastern Europe; Central Asia ranks last in adjusted measure.
       - Within-region variability matters: e.g., Sub-Saharan Africa exhibits extremely high variability within region compared to its average (Figure 13).

- Seniority and gender patterns (using adjusted measure)
  - Seniority:
    - Managers defined as pay grade from division chief upwards.
    - Managers are generally more educated than non-managers except in Middle East and North Africa (managers less educated) and Central and Eastern Europe (managers and non-managers similar) (Figure 14).
    - “Missing middle”: education attainment does not increase monotonically with seniority (Figure 15).
  - Gender:
    - Female education attainment is roughly on par with male officials, but female officials are far fewer (sample male:female ratio roughly 2:1).
    - Central Asia and Central and Eastern Europe are exceptions with near gender parity in employment.
    - Women are underrepresented among managers (Figure 17) and even more underrepresented among senior managers in all regions (Figure 18).
    - Implication: hired women are as educated as men, but fewer are hired and promoted to senior management.

- Agency comparison:
  - Central bank staff is not necessarily more educated than staff at ministries of finance and economies except in advanced countries (Figure 19).

### IV. Applications — incentives and government effectiveness (exploratory)
- About incentives (public pay gaps)
  - Cross-correlation results (Figure 20):
    - No statistically significant association between education attainment in public administrations and wage difference between public administration and manufacturing.
    - Positive and statistically significant association between education attainment in public administrations and wage difference between public sector and the financial sector.
  - Interpretation: better public sector pay relative to wages in finance is associated with higher education levels in public administrations, consistent with financial sector being the main outside option for civil servants in central banks and ministries.

- About government effectiveness
  - Tax revenue mobilization:
    - Positive and statistically and economically strong cross-correlation between education attainment and tax revenue over GDP (Figure 21).
    - Association remains positive and significant after controlling for GDP per capita and education in the general population (Figure 22).
  - Corruption:
    - Negative and statistically significant association between education attainment in public administration and an indicator of corruption based on PRS Group 2012 data (Figure 23).
    - Relationship remains negative when controlling for GDP per capita and education in the general population (Figure 24).
    - Suggestive interpretation: more educated public workforce may foster accountability and limit corruption; causality not established.
  - Public financial management and bureaucratic quality:
    - Positive and statistically significant association between education attainment and a composite public financial management indicator (World Bank 2012) (Figure 25).
    - Positive and statistically significant association between education attainment and bureaucratic quality indicator (PRS Group 2012) (Figure 26).
    - Findings robust to controls for GDP per capita and population education level.
  - Domestic financial sector standards:
    - Positive and statistically and economically significant association between education attainment in public administration and a composite index of domestic financial sector standards (regulation, supervision, competition...) (Figure 27).
    - Interpretation: a more educated public workforce may help raise domestic financial sector standards; resonates with concerns that regulatory bodies lag private firms in talent attraction and fall behind in keeping up with financial innovation.

### V. Conclusions and suggested future research directions
- Core conclusion: significant differences exist between education attainment in the general population and education attainment in public administrations; therefore, population-level education statistics are imperfect proxies for public administration human capital when studying state capacity.
- Dataset value: enables more direct study of the importance of human capital in the public sector and opens new research avenues.
- Key avenues for further work suggested in the text:
  - Establish causality in the associations between education attainment in public administrations and government effectiveness indicators.
  - Document organizational structure and incentive systems in public administrations and how they interact with human capital at all seniority levels.
  - Build complementary datasets for lower-level officials (e.g., tax collectors) to assess the interplay between human capital across hierarchies and public administration performance.

*Source: _wp12231 - References .............................................................................................................*

### REFERENCES

### _wp12231 - REFERENCES

### References
- Andrews, Matt, Lant Pritchett and Michael Woolcok, 2012, “Escaping Capability Traps through Problem-Driven Iterative Adaptation (PDIA),” Working Paper no. 299, Center for Global Development, Washington DC.
- Arezki A., A. Dupuy and A. Gelb, 2012, “Resource Windfalls, Optimal Public Investment and Redistribution: The Role of Total Factor Productivity and Administrative Capacity,” IMF Working Paper no. 12/200, International Monetary Fund: Washington DC.
- Barro Robert J. & Lee Jong-Wha, 2010, "A New Data Set of Educational Attainment in the World, 1950–2010," NBER Working Papers 15902, National Bureau of Economic Research, Cambridge.
- Baunsgaard, Thomas & Michael Keen, 2010, "Tax revenue and (or?) trade liberalization," Journal of Public Economics, Elsevier, October, Vol. 94(9-10), pp. 563–577.
- Besley, Timothy, 2006, “Principled Agents? The Political Economy of Good Government,” The Lindahl Lectures, Oxford University Press.
- Besley, Timothy and Torsten Persson, 2009, “The Origins of State Capacity: Property Rights, Taxation and Politics,” American Economic Review, Vol. 99(4), pp. 1218–44.
- Besley, Timothy, Jose G. Montalvo & Marta Reynal‐Querol, 2011, "Do Educated Leaders Matter?," Economic Journal, Royal Economic Society, Vol. 121(554), pp. F205–08.
- Botero, Juan, Alejandro Ponce & Andrei Shleifer, 2012, "Education and the Quality of Government," NBER Working Papers 18119, National Bureau of Economic Research, Cambridge.
- Chong, Alberto, Rafael La Porta, Florencio Lopez-de-Silanes & Andrei Shleifer, 2012, "Letter Grading Government Efficiency," NBER Working Papers 18268, National Bureau of Economic Research, Cambridge.
- Clements B., Gupta S., Karpowicz I., and S. Tareq, “Evaluating Government Employment and Compensation. Technical Notes and Manuals,” Fiscal Affairs Department, International Monetary Fund.
- Heston, A., R. Summers and B. Aten, 2009, “Penn World Table Version 6.3,” Center for International Comparisons of Production, Income and Prices at the University of Pennsylvania, August.
- Josse Delfgaauw & Robert Dur, 2008, "Incentives and Workers' Motivation in the Public Sector," Economic Journal, Royal Economic Society, Vol. 118(525), pp. 171–191, 01.
- Jones, B. and B. Olken, 2005, "Do leaders matter? National Leadership and Growth Since World War II," Quarterly Journal of Economics, pp. 835–864.
- Ostry, Jonathan D., Alessandro Prati & Antonio Spilimbergo, 2009, “Structural Reforms and Economic Performance in Advanced and Developing Countries,” IMF Occasional Paper.
- Putnam, R., 1993, “Making Democracy Work: Civil Traditions in Modern Italy,” Princeton, NJ: Princeton University Press.
- Rama, Martin, 1999, "Public Sector Downsizing: An Introduction," World Bank Economic Review, Oxford University Press, January, Vol. 13(1), pp. 1–22.
- Alvaro Forteza & Martin Rama, 2006, "Labor Market 'Rigidity' and the Success of Economic Reforms Across More Than 100 Countries," Journal of Policy Reform, Taylor and Francis Journals, Vol. 9(1), pp. 75–105.

### Appendix: Country Academic Ranking — Variables (Source: Universitas (2012))
- O1: Total number of journal articles produced by higher education institutions
- O2: Total articles produced per head of population
- O3: Average impact of articles
- O4: Weighted Shanghai ranking scores for universities per head of population
- O5: Shanghai scores for best three universities
- O6: Tertiary enrolment rates
- O7: Percentage of population over 24 with a tertiary qualification
- O8: Number of researchers in the nation per head of population
- O9: Unemployment rate of the tertiary educated compared with school leavers
- Source: Universitas (2012) http://www.universitas21.com/article/collaborations/details/105/measure-4-output

### Figures and Notes — Captions, Data Sources, and Key Measurement Definitions
- Figure 1. Education Attainment in the General Population around the World
  - Note: The figure presents the average years of schooling from Barro and Lee (2010).
- Figure 2. Example of Official’s CV
  - Note: The figure presents a standard template from the IMF Institute for Capacity Development Participants and Applicants Database (2012).
- Figure 3. Overview of the Distribution of Age in Sample
  - Note: The figure presents the age distribution of applicants to International Monetary Fund training activities during the period 1991-2011.
- Figure 4. Age Distribution by Region
  - Note: The figure presents the age distribution of applicants to International Monetary Fund training activities during the period 1991-2011 for height different regions.
- Figure 5. Gender Distribution by Region
  - Note: The figure presents the gender distribution of applicants to International Monetary Fund training activities during the period 1991-2011 for height different regions.
- Figure 6. Age Distribution by Type of Course
  - Note: The figure presents the age distribution of applicants to International Monetary Fund training activities during the period 1991-2011 for two different groups namely applicants who applied to non IMF headquarters (Non-HQ) courses and IMF headquarters (HQ) training activities. Note that training activities at HQ are most exclusively by nomination that is domestic authorities have to formally request their employees in order for the latter to be considered in the associated training activities. In contrast, Non-HQ course are mostly by applications that is civil servants can freely apply to the associated training activities.
- Figure 7. Fraction of Officials with Tertiary Education by Region
  - Note: The figure presents the fraction of applicants to International Monetary Fund training activities who hold a tertiary education degree during the period 1991-2011 for eight different regions.
- Figure 8. Education Attainments: Public Administration vs. General Population
  - Note: The figure presents the ratio of fraction of applicants to International Monetary Fund training activities during the period 1991-2011 who have a higher education degree over the same fraction for the general population for height different regions. The data on education attainment for the general population are from Barro and Lee (2012).
- Figure 9. Average Years of Tertiary Education in Public Administrations
  - Note: The figure presents the average years of tertiary education of applicants to International Monetary Fund training activities during the period 1991-2011 for height different regions.
- Figure 10. Fraction of officials with a degree from a country ranked in the top 48
  - Note: The figure presents the fraction of applicants to International Monetary Fund training activities during the period 1991-2011 who have obtained a degree from a country ranked in the first 48 countries for height different regions. The data is from Universitas website listed below: http://www.universitas21.com/article/collaborations/details/105/measure-4-output
- Figure 11. Normalized weighted years of tertiary education by region
  - Note: The figure presents our quality adjusted measure of education attainment of applicants to International Monetary Fund training activities during the period 1991-2011 for height different regions. The quality adjusted measure of education attainment is computed as a weighted average of the number of years of educations using a country academic ranking as weights from Universitas (2012). The measure is normalized between 0-1. The measure takes a minimum value of 0 if all officials have a high school degree only and takes a maximum value of 1 if all officials have a US PhD.
- Figure 12. Adjusted vs. non adjusted education attainment in public administration
  - Note: The figure presents our quality adjusted measure (blue bars) and non quality adjusted measure (blue bars) of education attainments of applicants to International Monetary Fund training activities during the period 1991-2011 for height different regions.
- Figure 13. Ratio of public administrations’ education attainment in public administrations over associated variance
  - Note: The figure presents our quality adjusted measure of education attainment divided by the variability (proxied by standard deviation) in education attainment within region of applicants to International Monetary Fund training activities during the period 1991-2011 for height different regions. This new measure put in perspective the quality of education attainment with the level of variability within region.
- Figure 14. Education attainment by rank and region
  - Note: The figure presents our quality adjusted measure of education attainment of applicants to International Monetary Fund training activities during the period 1991-2011 for height different regions. For each region and each panel, the left hand side bar numbered 0 stands for non managers (e.g. analyst, economist, senior economist...) and the right hand side bar numbered 1 stands for managers.
- Figure 15. Education attainment by seniority and region
  - Note: The figure presents our quality adjusted measure of education attainments of applicants to International Monetary Fund training activities during the period 1991-2011 for height different regions. For each region, we present the level of education attainment for several levels of seniority going from 2 to 5. A higher number stands for higher level of seniority.
  - Legend: 2: Economist; Official; Statistician; Researcher; Analyst. 3: Advisor; Consultant; Expert; Specialist. 4: Head; Director; Chief; Manager. 5: Minister; Governor
- Figure 16. Gender differences in education attainment in public administrations
  - Note: The figure presents our quality adjusted measure of education attainments of applicants to International Monetary Fund training activities during the period 1991-2011 for height different regions. For each region, the figure presents the level of education attainments for female (F) and male (M). For each gender, the left hand side bar is level of education attainment and the right-hand side presents the fraction of the associated gender in our sample of applicants.
- Figure 17. Fraction of Managers who are male
  - Note: The figure presents the fraction our manager by gender among our sample of applicants to International Monetary Fund training activities during the period 1991-2011 for height different regions. For each region, the figure presents the fraction of female (left hand side) and male (right hand side) manager.
- Figure 18. Fraction of senior managers who are male
  - Note: The figure presents the fraction our senior manager by gender among our sample of applicants to International Monetary Fund training activities during the period 1991-2011 for height different regions. For each region, the figure presents the fraction of female (left hand side) and male (right hand side) senior manager.
- Figure 19. Education Attainment by Agency and Region
  - Note: The figure presents our quality adjusted measure of education attainments of applicants to International Monetary Fund training activities during the period 1991-2011 for height different regions. For each region, we present the level of education attainment for two different agencies. 1 stands for the ministry of finance and economics and 2 stands for central banks.
- Figure 20. Education attainment and Relative Public Sector Pay
  - Note: The figure presents the cross-correlations between our measure of adjusted education attainment in public administrations (X-axis in both panels) and ratio of public sector pay over manufacturing (left hand side panel Y-axis)/financial sector (right hand side panel Y-axis) or height different regions. The data on civil servants’ pay is from Clements et al. (2010).
- Figure 21. Education attainment and tax collection
  - Note: The figure presents the cross-correlations between our measure of adjusted education attainment in public administrations (X-axis in both panels) and tax revenue over GDP. The data on tax revenue is from Baunsgaard and Keen (2010).
- Figure 22. Education attainment and tax revenues controlling GDP per education and education
  - Note: The figure presents the cross-correlations between our measure of adjusted education attainment in public administrations (X-axis) and tax revenue over GDP (Y-axis) controlling for both GDP per capita and education attainment in the general population. The data on tax revenue is from Baunsgaard and Keen (2010). The data on education attainment in the general population are from Barro and Lee (2012).
- Figure 23. Education attainment and corruption
  - Note: The figure presents the cross-correlations between our measure of adjusted education attainment in public administrations (X-axis in both panels) and an indicator of corruption. The indicator of corruption based on PRS Group 2012 data.
- Figure 24. Education Attainment and Corruption controlling for GDP per capita and Barro Lee
  - Note: The figure presents the cross-correlations between our measure of adjusted education attainment in public administrations (X-axis) and corruption (Y-axis) controlling for both GDP per capita and education attainment in the general population. The data on GDP per capita is from Heston and Summers (2009). Education attainment data in the general population is from Barro and Lee (2012).
- Figure 25. Education Attainment and Public Sector Management
  - Note: The figure presents the cross-correlations between our measure of adjusted education attainment in public administrations (X-axis) and public sector management (Y-axis). The data on public sector management is from World Bank Development Indicators (2011).
- Figure 26. Education attainment and bureaucratic quality
  - Note: The figure presents the cross-correlations between our measure of adjusted education attainment in public administrations (X-axis) and bureaucratic quality (Y-axis). The data on bureaucratic quality is from PRS Group (2012).
- Figure 27. Education and Domestic Financial Sector Standards
  - Note: The figure presents the cross-correlations between our measure of adjusted education attainment in public administrations (X-axis) and financial sector liberalization (Y-axis). The data on financial liberalization is from Ostry (2009).
- Table 1. Country Level Academic Ranking
  - Note: The figure presents the Universitas (2012) ranking of countries in terms of the performance of the higher education system. The variables Q1-Q9 are sub-indices of the overall score. The data and further details on the sub-indices and indices are available on website listed below: http://www.universitas21.com/article/collaborations/details/105/measure-4-output

*Source: _wp12231 - REFERENCES (IMF PDF)._

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12231.pdf_