## 1. Estimating Kosovo’s Shadow Economy

## Source details

**Canonical URL:** [1. Estimating Kosovo’s Shadow Economy](https://www.imf.org/-/media/files/publications/wp/2019/wpiea2019278-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2019/wpiea2019278-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2019/wpiea2019278-print-pdf.pdf.json)

---

### I. Introduction
- Objective: examine drivers of shadow economies in Europe (focus on emerging economies) and recommend policies to increase formality.
- Key empirical preview:
  - Primary determinants in Europe: regulatory quality and tax administration, plus macro factors including productivity and trade openness.
  - Remittances are significantly negatively associated with informality, suggesting migration and shadow economy can be substitutes.
  - Determinants for the Eastern Europe group: regulatory quality, government effectiveness, and human capital.
- Methodology overview:
  - Two-stage analysis combining empirical and MIMIC approaches.
  - Use Schneider and Hassan (2016) estimates (157 countries, 1999–2013) for regression analysis.
  - Re-estimate shadow economies for 47 European countries for 1999–2016 using MIMIC, guided by regression results.

### II. Defining and Measuring the Shadow Economy
- Definition adopted: mostly legal productive activities deliberately hidden from official authorities that, if recorded, would contribute to GDP (excludes illegal/criminal activities and do-it-yourself, charitable or household activities).
- Measurement approaches:
  - Direct approaches: surveys, tax auditing, compliance methods.
  - Indirect approaches: (i) income–expenditure GDP discrepancies, (ii) GDP growth vs electricity consumption growth, (iii) money demand vs actual currency circulating.
  - Model approach: Multiple Indicator, Multiple Causes (MIMIC) model — latent variable with causes and indicators; yields share of GDP.
- Paper’s empirical strategy:
  - Use Schneider and Hassan (2016) estimates for regressions to identify Europe-relevant causes.
  - MIMIC chosen causal variables: productivity (GDP per worker), government effectiveness, tax revenues, trade volume, agriculture value-added.
  - MIMIC chosen indicator variables: GDP growth and labor force participation rate.

### III. Size, Evolution, and Costs of the Shadow Economy
- Cross-country range in Europe: less than 10 to over 40 percent of GDP.
- Group averages and notes:
  - Advanced Economies: around 10–20 percent of GDP on average.
  - Emerging Economies: around 30–35 percent of GDP on average.
  - Most CIS countries: above 40 percent of GDP in many cases.
- Time dynamics:
  - Many countries saw increases since early 2000s (e.g., Croatia, Cyprus, Greece, Serbia); others saw declines (e.g., Czech Republic, Macedonia).
  - Most countries: increases in 2008–2010 and then declines to around pre-crisis levels.
- Economic and social costs (enumerated):
  - Distortions in labor markets; forgone revenue from underreporting of wages and output.
  - Suboptimal provision of public goods; weaker public revenues affecting growth.
  - Lower provision of and access to financing; limits on productivity and innovation.
  - Distorted national accounts and surveillance complicating policy.

### IV. Determinants of the Shadow Economy (conceptual and empirical correlations)
- Conceptual grouping:
  - Exit factors (voluntary informality): burdensome regulation, complex taxation, administrative barriers, corruption, low monitoring/enforcement, low benefits of formal employment, low quality of public goods.
  - Exclusion factors (forced informality): lack of formal-sector opportunities, low productivity, low skills/human capital.
- Empirical correlations summarized:
  - Positive correlations with shadow economy: unemployment, corruption, agriculture.
  - Negative correlations with shadow economy: GDP per capita, credit to private sector, revenue burden, human development, regulation quality.
- Literature highlights:
  - Weak institutional quality and regulatory burden identified as robust causes.
  - Higher tax burden and weaker tax administration associated with larger shadow economy.
  - Trade openness negatively associated with shadow economy.
  - Productivity and human capital negatively associated with shadow economy.
  - Agriculture tends to be positively associated in developing-country literature but may differ in European samples.
- Migration/remittances:
  - Two channels: substitution (remittances reduce informality) and safety-net/capital (remittances can encourage informality); net sign is country-dependent.

### V. Empirical Results (regression findings)
- Samples:
  - Panel of 40 European countries over 2000–2013 using fixed effects.
  - Eastern Europe sub-sample using random effects over a shorter period.
- Benchmark FE specification: Shadow Economyi,t = αi + βXi,t + δt Time + ui,t
- Variables found statistically significant and negatively associated with shadow economy in full sample (Table 1 summaries and coefficients):
  - Productivity (GDP/worker): -.221**, -.266***, -.271*** (reported across columns)
  - Remittances: -.038***, -.056**, -.022***
  - Regulatory Quality: -.149***, -.148**
  - Trade Openness: -.276***, -.286***
  - Agriculture VA/GDP: -.125***, -.150***
  - Government Effectiveness: -.162**
  - Human Capital (Penn World Tables): -.805***, -.782***
  - Fiscal Freedom: -.001**
  - Minimum Wage: -.011
- Estimation details (from Table 1):
  - Estimation methods: FE, FE, RE, RE
  - R-square: 0.317, 0.334, 0.710, 0.776
  - Observations: 509, 289, 205, 205
  - Countries: 40, 28, 23, 23
  - Years: 2000-13, 2004-13, 2005-13, 2005-13
  - Significance notation: ***p<0.01, **p<0.05, *p<0.1
- Eastern Europe sub-sample result:
  - Institutional factors (regulatory quality, government effectiveness, human capital) are especially important drivers.

### MIMIC Re-estimation Approach and Implementation
- Methodology: MIMIC approach estimates unobserved shadow economy using causal and indicator variables (see Appendix II for model equations and normalization).
- Sample and period: 47 countries over 1999–2016.
- Causal variables used in MIMIC: productivity (GDP per worker), government effectiveness, tax revenues, trade volume as percent of GDP, agriculture value-added as percent of GDP.
- Indicator variables used: GDP growth, investment, labor force participation rate.
- Models: Two specifications (Model A and Model B) produce similar results; Model B slightly higher for most countries with an average difference of around one percent.
- Comment on MIMIC estimates: absolute values sensitive to sample/variables/calibration; relative rankings more robust.

### VI. Estimated Sizes and Comparative Results (selected figures)
- Aggregate and group averages for 2013 (Table 2):
  - Sample average: 30.1 (Hassan & Schneider 2016), Model A 28.5, Model B 30.1
  - Europe (excl CIS): 25.2, 24.8, 25.8
  - Advanced: 20.5, 20.7, 21.5
  - Emerging: 33.2, 34.1, 35.6
  - EU: 23.2, 23.4, 24.4
  - CIS: 51.6, 40.9, 44.2
- Comparison to previous literature:
  - Advanced economies average: 20.7 percent in 2013 (updated) vs. 20.5 percent in Hassan and Schneider (2016).
  - Emerging economies average: 34.1 percent in 2013 (updated) vs. 33.2 percent in Schneider and Hassan (2016).
- Country-level notes:
  - Updated estimates higher for some countries (Bulgaria, Croatia, Germany, Norway, Baltic countries) and lower for others (Albania, Greece, Spain) relative to previous literature.
- Kosovo-specific estimate:
  - Estimated shadow economy for Kosovo: 38.8 percent of GDP in 2016.
  - Methodological note: absence of a Kosovo base-year (2000) MIMIC estimate led to using an average of five Western Balkan countries (Albania, Bosnia and Herzegovina, Macedonia, Montenegro, and Serbia) from Schneider and Hassan (2016) as a proxy to build the Kosovo time series.

### VII. Policy Recommendations and Options (enumerated)
- Overarching message: a comprehensive, country-specific package of reforms targeting the most relevant determinants is needed; size of the shadow economy is strongly and inversely related to per capita income and improved institutions.
- Improving regulation and institutional quality:
  - Reduce regulatory and administrative barriers (examples: automatic licensing; “one-stop-shop” registration; reducing registration fees).
  - Increase transparency and engagement (e.g., mandatory public electronic auctions for public procurement; public identification of tax evaders; industry engagement and follow-up audits).
  - Improve governance and public administration quality, rule of law, and accountability (enforce competition rules; sound regulatory frameworks; redistributive fiscal policies; fiscal transparency; transparency of ownership structures of financial institutions).
  - Specific institutional measures: establish clear rules/procedures for recruiting/training civil servants; strengthen property rights (improve cadastres and property registration); reduce court backlogs and improve case management.
- Taxation-related policies:
  - Increase tax compliance via improved registration, audit, and collection; facilitate information exchange between government agencies; broaden tax base by eliminating distortionary exemptions.
  - Automate and computerize procedures to reduce contact between tax officials and taxpayers and reduce bureaucracy and corruption.
  - Promote electronic payments: increasing electronic payments by an average of 10 percent annually for at least four consecutive years can reduce the size of the shadow economy by up to 5 percent (Schneider and Kearney 2013); note limitations where both transaction parties benefit from non-reporting or where barter/cryptocurrencies are used.
- Labor market reforms and human capital development:
  - Address “exclusion” factors: promote private-sector job creation and skill formation to provide pathways out of informality.
  - Policy measures: increase hiring and firing flexibility where labor laws are overly restrictive; strengthen enforcement and monitoring (e.g., mandatory registration of new workers); tailor employment and training measures for vulnerable groups (e.g., youth); support returning migrants with training and skills recognition; make vocational education more relevant; improve efficiency of education spending through better project prioritization, screening, and monitoring.

### VIII. Conclusion
- A multi-pronged reform strategy tailored to country-specific determinants is required to reduce the shadow economy.
- Policy menu emphasized for emerging economies: reduce regulatory and administrative burdens, promote transparency, improve government effectiveness, improve tax compliance and automation, promote electronic payments, and implement labor market and human capital policies to facilitate transitions from informal to formal activity and foster inclusive growth.

*Source: wpiea2019278-print-pdf - IMF staff calculations and analysis in "1. Estimating Kosovo’s Shadow Economy" (excerpted sections I–V, MIMIC re-estimation, Appendix III).*

### 1. Estimating Kosovo’s Shadow Economy..........................................................19

### 1. Estimating Kosovo’s Shadow Economy

### Figures
- 1. Shadow Economy in Europe ........................................................................9
- 2. Shadow Economy Estimation: The MIMIC Model ..............................................17

### Tables
- 1. Summary of Empirical Results .....................................................................15
- 2. Summary of MIMIC Estimations ..................................................................18

### Appendices
- I. Empirical Results with Non-European CIS Countries ...........................................23
- II. MIMIC Model ........................................................................................24
- III. MIMIC Estimation Results ........................................................................25

*Source: wpiea2019278-print-pdf - 1. Estimating Kosovo’s Shadow Economy (pages and item listing as provided).*

### References................................................................................................27

### References................................................................................................27

### I. INTRODUCTION
- Informality declined across Europe in recent years but remains significant, especially in Eastern Europe.
- Average shadow economy sizes reported:
  - Advanced Economies: around 15–20 percent of GDP.
  - Emerging Economies: around 30-35 percent of GDP.
- Shadow economies persist for a variety of reasons: avoidance of taxes and pension or social security payments, avoidance of labor and product market regulations, serving as source of employment and income in absence of formal opportunities, or as a safety net during cyclical downturns (Loayza and Rigolini 2006, Medina et al. 2017, Schneider 2004).
- Costs associated with large shadow economies include:
  - Distortions in the labor market.
  - Forgone revenue from underreporting of wages and output.
  - Suboptimal provision of public goods.
  - Lower provision of and access to financing.
  - Limited scale of production impeding productivity and innovation.
- Paper objectives:
  - Examine drivers of shadow economies in Europe, with focus on emerging economies.
  - Recommend policies to increase formality.
- Key empirical findings preview:
  - Primary determinants in Europe: regulatory quality and tax administration, plus macro factors including productivity and trade openness.
  - Remittances are significantly negatively associated with informality, suggesting migration and shadow economy can be substitutes.
  - Determinants for the Eastern Europe group: regulatory quality, government effectiveness, and human capital.
- Methodology overview:
  - Re-estimate shadow economy sizes for European countries including recent years based on determinants identified in empirical analysis.

### II. DEFINING AND MEASURING THE SHADOW ECONOMY
- Definition adopted (Schneider and coauthors): mostly legal economic and productive activities deliberately hidden from official authorities that, if recorded, would contribute to GDP (excluding illegal/criminal activities and do-it-yourself, charitable or household activities). “Informality” used interchangeably with “shadow economy”.
- Measurement approaches:
  - Direct approaches: surveys, tax auditing, compliance methods — detailed but may lack representativeness and cross-country consistency.
  - Indirect approaches: (i) discrepancy between income and expenditure GDP measures, (ii) difference between GDP growth and electricity consumption growth, (iii) difference between estimated money demand and actual currency circulating — sensitive to assumptions (elasticity, velocity of money, base year).
  - Model approach: Multiple Indicator, Multiple Causes (MIMIC) model — latent variable representing shadow economy with causes and indicators; estimated system yields share of GDP. Shortcomings include sensitivity to data, specifications, sample, calibration, starting values (Breusch 2005).
- No single ideal method; methodology choice governed by data availability and research objectives; multiple methods can improve accuracy.
- This paper’s empirical strategy:
  - Two-stage analysis combining empirical and MIMIC approaches.
  - Use Schneider and Hassan (2016) estimates (157 countries, 1999–2013) for regression analysis to identify Europe-relevant factors.
  - Re-estimate shadow economies for 47 European countries for 1999–2016 using MIMIC, guided by regression results.
  - Chosen causal variables for MIMIC: productivity (GDP per worker), government effectiveness, tax revenues, trade volume, agriculture value-added.
  - Chosen indicator variables for MIMIC: GDP growth and labor force participation rate.
  - Note: Hassan and Schneider (2016) original input variables included government spending as percent of GDP, unemployment rate, self-employment rate, Economic and Business Freedom Indices, and M1/M2 and labor force participation as indicators.

### III. SIZE, EVOLUTION, AND COSTS OF THE SHADOW ECONOMY
- Cross-country range: share of shadow economy ranges from less than 10 to over 40 percent of GDP across European countries.
- Group averages and notes:
  - Advanced Economies: tends to be smaller at around 10–20 percent of GDP on average.
  - Emerging economies: on average around 30–35 percent of GDP.
  - Most CIS countries: above 40 percent of GDP in many cases.
- Time dynamics:
  - Average size in Europe broadly similar to mid-2000s, but heterogeneous across countries.
  - Many countries saw increases since early 2000s (e.g., Croatia, Cyprus, Greece, Serbia), others saw declines (e.g., Czech Republic, Macedonia).
  - Most countries: shadow economies increased in 2008–2010 and then declined to around pre-crisis levels.
- Relationship with development:
  - Shadow economy share is strongly negatively correlated with income per capita across samples and time periods.
  - In advanced economies shadow economy dominated by tax evasion and undeclared labor in registered firms (Schneider and Buehn 2012).
  - In developing economies higher share of informal workers reflecting lack of formal sector opportunities (Oviedo 2009).
- Economic and social implications (enumerated costs):
  - Public revenue and services: untaxed activity weakens public revenues → fewer/weaker public goods and services → can reduce growth and increase informality via weakened government effectiveness perceptions (Schneider 2004).
  - Innovation and productivity: informal operation limits firm growth, R&D, innovation, and hiring (Bobbio 2016) → skews resource allocation, reduces capital accumulation and potential output.
  - Labor markets: large shadow economy associated with high/persistent unemployment and low labor force participation (Schneider 2013); informal work tends to be lower-paying, less secure, weaker training (Williams 2015).
  - Financial access: banks avoid lending to unregistered firms and informal borrowers → hampers financial deepening (Gobbi and Zizza 2012).
  - Data and surveillance: distorts economic indicators (national accounts, employment, income, consumption) and complicates macro analysis and policy choices.

### IV. DETERMINANTS OF THE SHADOW ECONOMY
- Empirical correlations (summary):
  - Positive correlations with shadow economy: unemployment, corruption, agriculture.
  - Negative correlations with shadow economy: GDP per capita, credit to private sector, revenue burden, human development, regulation quality.
- Conceptual framing: determinants grouped as “exit” factors (voluntary informality) and “exclusion” factors (forced informality).
  - Exit factors (examples):
    - Burdensome and costly regulation, including high entry costs, trade barriers.
    - Complex/excessive taxation and poor tax administration.
    - Administrative barriers including excessive paperwork, corruption.
    - Low monitoring and enforcement.
    - Low benefits of formal employment/registration.
    - Low quality of public goods and services.
    - Individual preference for self-employment.
  - Exclusion factors (examples):
    - Burdensome and costly regulation, including high entry costs, trade barriers.
    - Lack of opportunities in formal sector for certain demographic/ethnic groups.
    - Low productivity.
    - Low skills and low human capital.
- Literature highlights (key drivers):
  - Weak institutional quality: excessive regulation, inefficient government institutions, weak rule of law, widespread corruption → larger shadow economy. Regulatory burden identified as most robust cause (Dabla-Norris, Gradstein, and Inchauste 2008).
  - Tax burden and tax administration: higher tax burden and/or lower monitoring/enforcement → stronger incentives for tax evasion and underreporting (Schneider and Williams 2013, Hassan and Schneider 2016).
  - Trade openness: negatively associated with shadow economy (Torgler and Schneider 2007) since trade is more transparent and easier to tax.
- Role of productivity, human capital, and agriculture:
  - Higher productivity (GDP per worker) associated with smaller informal sectors (Porta and Shleifer 2008).
  - Lower human capital correlated with larger shadow economies (Porta and Shleifer 2008, Dabla-Norris, Gradstein, and Inchauste 2008).
  - Larger agricultural sector tends to be positively associated with shadow economy in developing-country literature (Vuletin 2008, Schneider 2014).
- Migration and remittances:
  - Two opposing channels:
    - Substitution: migration/remittances can substitute for informal employment (negative relationship).
    - Safety-net/capital: remittances can encourage informality by providing capital or safety net for choosing informal work (positive relationship; example: Moldova, Ganta 2012).
  - Net sign depends on which effect dominates.

### V. EMPIRICAL RESULTS
- Objective: identify determinants of the shadow economy in Europe using Schneider and Hassan (2016) estimates as dependent variable and macro/institutional indicators as independent variables.
- Samples:
  - Panel of 40 European countries over 2000–2013 (full sample) using fixed effects.
  - Sub-sample of “Eastern European” countries using a random effects model over a shorter period (country set specified in text).
- Benchmark fixed-effects specification:
  - Shadow Economyi,t = αi + βXi,t + δt Time + ui,t
  - Shadow Economyi,t: size of shadow economy as share of GDP.
  - αi: country fixed effects; Xi,t: vector of macro and institutional variables; Timet: time fixed effects.
- Variables found statistically significant and negatively associated with size of shadow economy in full sample (see Table 1, regressions 1 and 2):
  - Regulatory quality (World Bank Regulatory Quality index): negative relationship with shadow economy.
  - Tax burden / Fiscal freedom (Heritage Foundation Fiscal Freedom index): negative (if weak) relationship with shadow economy.
  - Productivity (GDP per worker): negative relationship with shadow economy.
  - Trade openness (trade volume/GDP): negative relationship with shadow economy.
  - Remittances (share of GDP): negative relationship with shadow economy, suggesting migration and informality can be substitutes when controlling for country fixed effects.
  - Agriculture value-added per GDP: negative relationship with shadow economy in European sample (contrary to some literature focused on developing countries); possible explanation: formalized agriculture in advanced Europe and sample composition.
- Notes on estimation:
  - Imposing country fixed effects reduces explanatory power of variables with limited time variation.
  - Selection of indicators follows literature; results broadly robust to alternative indicators (World Bank’s Regulatory Quality, Heritage’s Fiscal Freedom, World Bank’s Government Effectiveness; alternatives listed in text).
- Eastern Europe sub-sample approach:
  - Benchmark random-effects specification: Shadow Economyi,t = αi + βXi,t + δt Time + vi + εi,t
  - Country set includes: Albania, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Macedonia, Moldova, Montenegro, Poland, Romania, Russia, Serbia, Slovakia, Slovenia, Turkey, and Ukraine, and also Greece and Cyprus (see text).
  - Empirical findings indicate institutional factors (regulatory quality, government effectiveness, human capital) more clearly drive shadow economies in Eastern Europe.

*Source: Excerpt from IMF working paper (References section and Sections I–V).*

### 2013. We find that the shadow economy is again negatively associated with productivity,

### wpiea2019278-print-pdf - 2013. We find that the shadow economy is again negatively associated with productivity,

### Empirical findings on drivers of the shadow economy
- Main negative associations (Europe sample, regressions referenced in text and Table 1):
  - Productivity (GDP/worker): -.221**, -.266***, -.271*** (reported across columns)
  - Remittances: -.038***, -.056**, -.022***
  - Regulatory Quality: -.149***, -.148**
- Additional important negative relationships identified:
  - Government Effectiveness: -.162** (World Bank Government Effectiveness Index used)
  - Human Capital: -.805***, -.782*** (Penn World Tables human capital index)
- Other coefficients and controls reported in Table 1:
  - Trade Openness: -.276***, -.286***
  - Agriculture VA/GDP: -.125***, -.150***
  - Minimum Wage: -.011
  - Fiscal Freedom: -.001**
- Estimation details from Table 1:
  - Estimation: FE, FE, RE, RE
  - R-square: 0.317, 0.334, 0.710, 0.776
  - Observations: 509, 289, 205, 205
  - Countries: 40, 28, 23, 23
  - Years: 2000-13, 2004-13, 2005-13, 2005-13
  - Significance notation: ***p<0.01, **p<0.05, *p<0.1
- Broader sample including non-European CIS countries (Armenia, Azerbaijan, Georgia, Kazakhstan, Kyrgyz Republic, and Tajikistan) confirms negative association with productivity and remittances and shows stronger importance of institutional factors: government effectiveness, human capital, rule of law, and, to a lesser extent, corruption and ease of paying taxes.

### MIMIC re-estimation approach and implementation
- Methodology: multiple indicator-multiple cause (MIMIC) approach used to estimate unobserved shadow economy (see Appendix II for model equations and normalization).
- Sample and period: 47 countries over 1999–2016 (country list provided in text).
- Causal variables used in MIMIC estimation: productivity (GDP per worker), government effectiveness, tax revenues, trade volume as percent of GDP, agriculture value-added as percent of GDP.
- Indicator variables used: GDP growth, investment, labor force participation rate.
- Models: Two specifications (Model A and Model B) produce similar results; Model B slightly higher for most countries with an average difference of around one percent.
- Comment on MIMIC estimates: subject to limitations and sensitivity to sample and variables; absolute values can be sensitive but relative ranking is more robust.

### Estimated sizes and comparative results (selected figures)
- Aggregate and group averages for 2013 (Table 2):
  - Sample average: 30.1 (Hassan & Schneider 2016), Model A 28.5, Model B 30.1
  - Europe (excl CIS): 25.2, 24.8, 25.8
  - Advanced: 20.5, 20.7, 21.5
  - Emerging: 33.2, 34.1, 35.6
  - EU: 23.2, 23.4, 24.4
  - CIS: 51.6, 40.9, 44.2
- Comparison to previous literature:
  - Average estimate for advanced economies: 20.7 percent in 2013 (updated) compared to 20.5 percent in Hassan and Schneider (2016).
  - Average estimate for emerging economies: 34.1 percent in 2013 (updated) compared to 33.2 percent in Schneider and Hassan (2016).
- Country-level note: updated estimates higher for some countries (Bulgaria, Croatia, Germany, Norway, Baltic countries) and lower for others (Albania, Greece, Spain) relative to previous literature.
- Kosovo specific estimate:
  - Estimated shadow economy for Kosovo: 38.8 percent of GDP in 2016.
  - Methodological note: absence of a Kosovo base-year (2000) MIMIC estimate led to using an average of five Western Balkan countries (Albania, Bosnia and Herzegovina, Macedonia, Montenegro, and Serbia) from Schneider and Hassan (2016) as a proxy to build the Kosovo time series.

### Policy recommendations and options
- Overarching message: a comprehensive, country-specific package of reforms targeting the most relevant determinants is needed; size of the shadow economy is strongly and inversely related to per capita income and improved institutions.
- Improving regulation and institutional quality:
  - Reduce regulatory and administrative barriers (examples: automatic licensing in Georgia; “one-stop-shop” registration in Estonia; reducing registration fees).
  - Increase transparency and engagement (e.g., mandatory public electronic auctions for public procurement; public identification of tax evaders; industry engagement and follow-up audits).
  - Improve governance and public administration quality, rule of law, and accountability (longer-term reforms include enforcing competition rules, sound regulatory frameworks for infrastructure and finance, redistributive fiscal policies and fiscal transparency, transparency of ownership structures of financial institutions).
  - Specific institutional measures: establish clear rules and procedures for recruiting/training civil servants; strengthen property rights (improve cadastres and property registration); reduce court backlogs and improve case management.
- Taxation-related policies:
  - Increase tax compliance via improved registration, audit, and collection; facilitate information exchange between government agencies; broaden tax base by eliminating distortionary exemptions.
  - Automate and computerize procedures to reduce contact between tax officials and taxpayers and reduce bureaucracy and corruption.
  - Promote electronic payments: evidence cited that increasing electronic payments by an average of 10 percent annually for at least four consecutive years can reduce the size of the shadow economy by up to 5 percent (Schneider and Kearney 2013); note limitations where both transaction parties benefit from non-reporting or where barter/cryptocurrencies are used.
- Labor market reforms and human capital development:
  - Encourage formalization by addressing “exclusion” factors: promote private-sector job creation and skill formation to provide pathways out of informality.
  - Policy measures: increase hiring and firing flexibility where labor laws are overly restrictive; strengthen enforcement and monitoring (e.g., mandatory registration of new workers); tailor employment and training measures for vulnerable groups (e.g., youth); support returning migrants with training and skills recognition; make vocational education more relevant; improve efficiency of education spending through better project prioritization, screening, and monitoring.

### Conclusion
- A multi-pronged reform strategy tailored to country-specific determinants is required to reduce the shadow economy.
- Policy menu for emerging economies emphasized in the report: reduce regulatory and administrative burdens, promote transparency, improve government effectiveness, improve tax compliance and automation, promote electronic payments, and implement labor market and human capital policies to facilitate transitions from informal to formal activity and foster inclusive growth.

*Source: IMF staff calculations and analysis as presented in the provided content unit.*

### Appendix III. MIMIC Estimation Results

### Appendix III. MIMIC Estimation Results

### Methodological caveats on benchmarking and calibration
- Application of the so-called calibration or benchmarking procedure, regardless which one is used, requires experimentation, and a comparison of the calibrated values in a wide academic debate.
- At this stage of research it is not clear which benchmarking method is the best or most reliable.
- See Dell’Anno and Schneider (2009) for a detailed discussion on different benchmarking procedures.
- Latest discussion and critique of the MIMIC procedure referenced: Breusch (2016), Feige (2016a,b), Schneider (2016) and Hashimzade and Heady (2016).

### Empirical estimates and scope
- Table. Shadow economy estimates*
- * Shadow economy estimates are based on Model A, excluding CIS countries.
- Sources: IMF staff calculations.

### References cited (selection from appendix)
- Dell’Anno and Schneider (2009) — detailed discussion on benchmarking procedures (referenced in text).
- Breusch, T. (2005). Estimating the underground economy using MIMIC models. Working Paper, National University of Australia, Canberra, Australia.
- Breusch (2016); Feige (2016a,b); Schneider (2016); Hashimzade and Heady (2016) — cited as part of recent critique/discussion of MIMIC.
- Additional cited literature relevant to shadow economy measurement, determinants, and policy includes:
  - Bobbio, E. (2016). Tax evasion, firm dynamics and growth (No. 357). Bank of Italy.
  - Dabla-Norris, E., Gradstein, M., & Inchauste, G. (2008). The determinants of informality. Journal of Development Economics, 85(1), 1–27.
  - Elek, P., Köllő, J., Reizer, B., & Szabó, P. A. (2012). Detecting Wage Under-Reporting Using a Double-Hurdle Model.
  - EY Report (2017) Reducing the Shadow Economy through Electronic Payments.
  - Hassan, M., & Schneider, F. (2016). Size and Development of the Shadow Economies of 157 Worldwide Countries: Updated and New Measures from 1999 to 2013. J Glob Econ, 4(218), 2.
  - Medina, L., A. Jonelis, and M. Cangul, 2017, The Informal Economy in Sub-Saharan Africa: Size and Determinants, International Monetary Fund, WP/17/156.
  - Vuletin, G. J. (2008). Measuring the informal economy in Latin America and the Caribbean (No. 8–102). International Monetary Fund.
  - Yasser, A., & Medina, L. Measuring the Informal Economy in the Caucasus and Central Asia (pp. 1–16). IMF Working Paper 13/37, May 2013, рр. 1–17.
- Full list of references continues in the appendix (selected titles shown above).

*Source: IMF staff calculations, Appendix III. MIMIC Estimation Results.*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2019/wpiea2019278-print-pdf.pdf_