## cr18129-colombiasi

## Source details

**Canonical URL:** [cr18129-colombiasi](https://www.imf.org/-/media/files/publications/cr/2018/cr18129-colombiasi.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2018/cr18129-colombiasi.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2018/cr18129-colombiasi.pdf.json)

---

### The Structure of Colombian Exports
- Real exchange rate and scope:
  - The real exchange rate has depreciated 30 percent from its peak in 2012.
  - Large real depreciation and moderate commodity-price outlook could enable substantial growth in nontraditional exports.
- Export composition and diversification:
  - Traditional products (oil, metals, raw agricultural products) account for around 60 percent of nominal exports.
  - Strong Revealed Comparative Advantage (RCA) in oil; weak RCA in technology-intensive products.
  - Export complexity (Hausmann method) in line with LA6 average and little changed over time.
  - Herfindahl concentration index: Colombia was the least diversified country in LA6 in 2013 after the increase in oil exports in 2007−13.
- Recent performance, elasticities, and quantified effects:
  - Since 2016, export increases have been moderate despite the large depreciation; possible causes include weak trading-partner demand, delayed depreciation effects, weak commodity prices, structural impediments.
  - Elasticity of exports to trading-partner growth: 2.2−2.6.
  - Elasticity of exports to the REER: -0.1.
  - Trading-partner growth in 2015−17 was about a percentage point below pre-2008 values → translates into a 2.2−2.6pp decline in export growth.
  - The 30 percent real depreciation would increase export growth by about 3pp.
  - Net effect: the two largely offset each other.
  - Model projection: average quarterly year-on-year export growth projected at around 3 percent in Colombia for 2017Q3−2018Q4 (historical reference: -1.7 percent growth rate observed in 2016Q1-2017Q2).
- Nontraditional exports and historical episodes:
  - Colombia lost nontraditional export market share in 2000−15; loss driven largely by very weak growth in Ecuador and Venezuela.
  - 61 large depreciations analyzed (17 in LA) across 53 commodity exporters in 1985−2015.
  - Median commodity exporter outcomes after a large depreciation:
    - Improves nontraditional market share by 9 percent in the three years following a large depreciation.
    - Cumulative gain of 4.8 percent in five years.
    - Commodities continue to dominate: share of commodities in total exports falls by just two percentage points for the median country.
  - Colombia scenario: if Colombia tracked the median episode, cumulative nontraditional export growth in 2015−20 would reach 30 percent; estimated current-account improvement: about 0.25pp (assumes imports do not increase).
- Structural impediments to export growth:
  - Tariffs:
    - Average tariffs have fallen to around 6.5 percent.
    - Tariff gap to OECD average narrowed to about 1.5 percentage points.
    - Nominal tariffs on agricultural products average 18 percent; tariffs as high as 49 and 70 percent for dairy and beef products.
  - Non-tariff costs and customs/infrastructure:
    - Nontariff costs account for three quarters of the cost of imports; Garcia and others (2016): cost of importing was 36 percent of a product’s price in 2012.
    - Number of nontariff barriers per product category grew from 4 in 2001 to 16 in 2014; for some categories (textiles, wood) 100 percent of imported value subject to measures.
    - Infrastructure and transportation gaps raise transport costs; exporter survey: 75 percent of respondents cite freight prices and bad roads as a top impediment.
    - Colombian customs relatively slow per World Bank data, especially for exports; authorities progressing with new port scanners and accelerated processing.
- Policy implications (exports):
  - Rapidly reduce tariff dispersion and cut proliferation of nontariff barriers.
  - Close infrastructure gaps (including 4G projects) and address inefficiencies in road transport (market structure, old truck fleet, barriers to entry).
  - Speed up customs procedures and disseminate information about trade agreements to accelerate nontraditional export growth.

### Labor Informality: Recent Reduction, Drivers, and Policy Responses
- Key statistics and trends:
  - Share of workers nationwide not contributing to social security dropped from 70 percent in Q1 2007 to 62.3 percent in Q3 2017.
  - Depending on the definition used, labor informality fell by 3.3 to 8.2 percentage points between 2007 and 2017.
  - Using the social security criterion, informality stood at 63 percent as of Q3 2017, relative to 70 percent in 2007.
  - Using the size criterion (urban concept), informality dropped below 50 percent for the first time in early 2014.
  - Firm informality estimates lie between 45−65 percent.
  - Estimates suggest between 20−40 percent of output is produced by the informal sector; Medina and Schneider (2018) find the informal sector accounts for 21−33 percent of GDP in recent years; Anif 2017 reports the shadow economy at around 40 percent of GDP (informal sector 33.5 percent + illegal sector 6.3 percent).
  - In 2016, IDB SIMS data: 67.3 percent of salaried workers were formal, but only 12.7 percent of independent workers were formal.
- Role of education and decomposition:
  - Expansion of higher education coverage: from 37 percent in 2010 to 52 percent of the age cohort in 2017.
  - Decomposition: increase in average level of education accounts for two-thirds of the fall in labor informality between Q1 2007 and Q2 2017.
  - Subnational finding: higher education index explains 44−49 percent of cross-sectional variation in informality across 23 largest cities and departments; regional differences in World Bank Doing Business explain 1−5 percent.
  - Formally by education group (2007→2017): no more than primary education formal rate increased from 12 to 15 percent; some secondary from 37 to 40 percent; tertiary from 71 to 73 percent. Share of workers with only primary education or less dropped from half to less than 40 percent; share with higher education increased by 7 percentage points to 23 percent.
- Policy measures taken and evidence of impact:
  - Major reforms: 2006 Entrepreneurship Law; 2010 Formalization and Job Creation Law; 2012 Tax Reform cut payroll taxes by 13.5 percentage points for workers earning below 10 minimum wages (reduced mandatory contributions from 29.5 percent to 16 percent of gross wage earnings for workers earning less than 10 minimum wages).
  - Impact evaluations: 2012 tax reform led to several hundred thousand additional formal jobs and reduced informality; evidence stronger for smaller firms and workers close to minimum wage.
- Structural impediments remaining:
  - Skill mismatches, especially technical.
  - Non-wage labor costs remain high (e.g., 4 percentage points payroll tax to finance Cajas de Compensacion Familiar).
  - Complex bureaucratic steps and high registro mercantil fees; registro mercantil firm-size threshold changed from 10 to 5 employees in 2010.
  - Minimum wage binding: at close to 90 percent of the median wage in 2015, Colombia had the highest ratio of minimum wage to median wage in the OECD sample; suggests regional differentiation.
  - Tax burden for small and micro firms: average total effective tax rate around 50 percent of profits and for small firms as high as 100 percent.
- Policy recommendations (informality):
  - Continue strengthening coverage and quality of education; expand higher education supply and access for low-income students.
  - Develop the national framework (marco nacional de cualificaciones) to align higher education with labor market needs.
  - Consider alternative financing for Cajas de Compensacion Familiar and review their services to avoid duplication.
  - Consider regionally differentiated minimum wage with adequate offsets.
  - Reduce registro mercantil registration fees for small firms; make contributions less regressive while preserving registro’s role.
  - Cut red tape via Ventanilla Unica Empresarial; harmonize regional practices.
  - Strengthen enforcement and outreach (IT systems, Red Nacional de Formalizacion Laboral, formalization fairs).
  - Sequence reforms: prioritize labor market reforms before product market reforms; combine reforms in packages.
- Measurement and definitions:
  - Informality measures vary: worker-based (not contributing to social security) and DANE size/occupational (firms ≤5 employees, unpaid family aids, self-employed with low education); DANE urban definition covers major 13 or 23 cities.
  - Firm informality measurement remains limited and sporadic.

### Model Simulations: STRESS Model Calibration and Reform Scenarios
- Model type and calibration:
  - Small open-economy dynamic general equilibrium model with unemployment from hiring costs and wage bargaining, endogenous firm entry (STRESS model).
  - Distinction between formal and informal sectors; calibration targets and steady-state differences reported exactly:
    - Steady-state differences (Formal Sector vs. Informal Sector): Output 3.8 vs. 1.5; TFP 1 vs. 1; Capital Share 0.34 vs. 0.34; Capital Stock 24.99 vs. 12.04; Employment 0.32 vs. 0.55; Capital per employee 78.09 vs. 21.89; Labor productivity 11.88 vs. 2.73; Retailers Exit Rate 0.125 vs. 0.35; Probability of Hiring 0.05 vs. 0.7; Probability of Firing 0.09 vs. 0.7; Tax rate on employees 0.04 vs. 0; Tax rate on employers 0.16 vs. 0; Worker bargaining power 0.78 vs. 0.2; Retailer Entry Costs 2.89 vs. 0.06.
    - Calibration targets (Model vs Data): Private Investment as share of GDP 0.24 vs. 0.23; Openness as share of GDP 0.40 vs. 0.39; Unemployment rate 13.30 vs. 13.12; Shadow economy as share of GDP 0.30 vs. 0.21 to 0.4; Informal employment as share of total employment 0.63 vs. 0.66; Ratio of formal to informal sector wages 2.87 vs. 2.78.
- Simulation steady-state impacts (changes reported exactly):
  - Reduction in Firm Entry Costs by 10 percent:
    - Labor Informality (percentage points change): -0.2
    - Formal Employment (percentage points change): 0.4
  - Reduction in Payroll tax from 16 to 14 percentage points:
    - Labor Informality (percentage points change): -0.3
    - Formal Employment (percentage points change): 0.1
  - Reduction in Formal/Informal Worker Bargaining Power Wedge by 10 percent:
    - Labor Informality (percentage points change): -1.2
    - Formal Employment (percentage points change): 3.5
- Interpretation:
  - Labor market reforms and cutting firm entry costs lower informality in the long run by reallocating employment toward the formal sector.
  - Combining and sequencing reforms (labor first) tends to be more favorable per literature.

### Fiscal Rule, Shock Response, and Medium-Term Fiscal Strategy
- Fiscal rule and shock response (2013–2017):
  - Fiscal cost of declining oil revenue: 3.1 percent of GDP.
  - Increasing interest rate bill (partly due to peso depreciation): 0.6 percent of GDP.
  - Combined shock: 3.7 percent of GDP.
  - Direct application of the structural balance formula allowed the deficit to widen by about 1.7 percent of GDP between 2013−16.
  - 2013−17 net impact: widening of 1.3 percent of GDP (about 35 percent of the oil revenue+interest bill combined shock).
  - Revenue measures (two tax reforms 2014 and 2016, plus tax administration gains): about 1.3 percent of GDP of non-oil revenue (absorbing about 35 percent of the shock).
  - Primary expenditure cuts required: about 0.5 percent of GDP; capital expenditure cut: 0.8 percent of GDP.
  - 2017 outturn benefited from a one-off telecom fine: 0.5 percent of GDP.
- Authorities’ baseline medium-term plan and composition:
  - Structural deficit declines from 1.9 percent of GDP in 2017 to 1 percent of GDP by 2022.
  - Headline deficit will have to decline by 2.5pp of GDP by 2022.
  - One half (1.2 pp of GDP) of the narrowing will come from higher oil revenues and lower interest expenses.
  - Non-oil tax revenue will decline over the medium-term due to expiration of the corporate surcharge CREE and one-off telecom fine in 2017 outweighing tax administration and formalization gains.
  - Primary expenditure cuts of about 1.7 percent of GDP will be required by 2022; fixed capital formation cuts by 1.1 percent of GDP by 2022.
  - Gross public debt will decline by about 9 pp of GDP between 2018 and 2023 to about 40 percent of GDP.
- Fiscal stance and short-term growth implications:
  - Front-loaded fiscal adjustment with most negative fiscal impulse by 2020.
  - Negative fiscal impulse from Central Government measured by change in non-oil primary structural deficit: 0.5 percent of GDP in each 2019 and 2020.
  - Combined CPS-level negative impulse: about 0.4 percent of GDP in 2019 and 0.9 percent of GDP in 2020.
  - For 2021 and 2022 fiscal stance implies a small positive impulse (0.1 percent of GPD) at the CPS level.
  - Colombia’s output gap expected to average about 1.3 percent of potential GDP during 2018−20.
  - Public investment at the CPS level will decline by about 1 percent of GDP between 2017 and 2020.
  - Fiscal multiplier of capital expenditure estimated between 0.15 and 0.46; expected decline in investment could subtract about 0.3 percentage points of GDP growth.
- Alternative adjustment scenario:
  - Keeping structural deficit constant during 2019 and 2020 at 1.9 percent of GDP creates space for 0.2−0.5 percent of GDP of additional room for primary expenditure to protect investment and social programs.
  - Gross public debt ratio would increase by about 0.8 pp of GDP by 2020 vis-à-vis the baseline but remain in a declining path.
- Public expenditure efficiency and priority reforms:
  - Expert Commission recommendations include unifying current expenditure and investment budgets on a program basis, evaluating budget inflexibilities, improving flexibility of oil-royalty system, and creating a Fiscal Council.
  - Social outcomes (2010→2017): income inequality (Gini) declined from 55.7 percent in 2009 to 50.8 percent in 2017; multidimensional poverty declined from 30.4 percent in 2010 to about 17 percent in 2017.
- Pension system diagnosis and reform options:
  - Low coverage: 24 percent of the elderly have a pension.
  - Regressive distribution: about 75 percent of pension subsidies received by the top two quintiles; only 12 percent received by the lowest two quintiles.
  - Most fiscal pension cost (3.7 percent of GDP) stems from special sub regimes.
  - Reform options include parametric changes (raise retirement age, link to life expectancy, increase years used for benefit calculation from 10 to 20 years, increase pension contribution to 18 percent, tax high pension benefits), using savings to strengthen non-contributory programs, structural options to phase out PAYG or create a small pillar 1, and homogenize special sub regimes.
- Sectoral recommendations:
  - Health: focus on prevention, better cost-benefit quantification, reduce drug costs, redirect payroll tax revenue to fund health services.
  - Education: improve access to quality early education, create oversight agency, ensure stable funding, strengthen local government funding and distribution linked to public university quality and efficacy.
  - Subsidies: improve data, focus on fewer better-monitored subsidy programs, strengthen evaluation.

*Source: IMF staff chapter excerpts (cr18129-colombiasi).*

### 1. The Structure of Colombian Exports __________________________________________________ 4

### 1. The Structure of Colombian Exports

### A. Introduction
- The real exchange rate has depreciated 30 percent from its peak in 2012.
- Purpose: assess how much boost to exports the depreciation could represent by:
  - exploring the response of exports to fundamentals;
  - examining behavior of non-traditional exports after large depreciations in commodity exporters;
  - identifying Colombia’s structural bottlenecks to international trade.
- Key takeaway: weak trading-partner growth has so far offset much of the depreciation’s boost; removing structural bottlenecks (infrastructure, non-tariff barriers) would support nontraditional export growth.

### B. The Structure of Colombian Exports
- Exports concentrated in oil and other commodities; limited diversification:
  - Traditional products (oil, metals, raw agricultural products) account for around 60 percent of nominal exports.
  - Strong Revealed Comparative Advantage (RCA) in oil; weak in technology-intensive products (Ding and Hadzi-Vaskov, 2017).
  - Export complexity (Hausmann method) in line with LA6 average and little changed over time.
  - Herfindahl concentration index: Colombia was the least diversified country in LA6 in 2013, after the increase in oil exports in 2007−13.
- Opportunity: the large real depreciation and moderate commodity-price outlook could enable substantial growth in nontraditional exports.

### C. Recent Export Performance and Outlook
- Since 2016, export increases have been moderate despite the large depreciation; possible causes: weak trading-partner demand, delayed depreciation effects, weak commodity prices, structural impediments.
- Model of export volume growth: function of commodity prices, REER, and trading-partner growth (quarterly data since 2001 for 23 countries); coefficients estimated on 2001Q1−2014Q2 and forecasts for 2014Q3−2017Q4.
- Model assessment:
  - Colombia’s recent export performance broadly in line with fundamentals but somewhat weaker than historical relationships.
  - Other countries: some overperformed significantly; other Latin American commodity exporters fared worse than Colombia.
- Elasticities (model-derived) for Colombia:
  - Elasticity of exports to trading-partner growth: 2.2−2.6.
  - Elasticity of exports to the REER: -0.1.
- Quantified offsetting effects:
  - Trading-partner growth in 2015−17 was about a percentage point below pre-2008 values → translates into a 2.2−2.6pp decline in export growth.
  - The 30 percent real depreciation would increase export growth by about 3pp.
  - Net effect: the two largely offset each other.
- Model projection:
  - Average quarterly year-on-year export growth projected at around 3 percent in Colombia for 2017Q3−2018Q4.
  - Historical reference: -1.7 percent growth rate observed in 2016Q1-2017Q2.

### D. A Deeper Look at Nontraditional Exports
- Nontraditional export market share dynamics:
  - Colombia lost nontraditional export market share in 2000−15; the loss was due largely to very weak growth in Ecuador and Venezuela.
  - Gap between total market share and market share excluding Ecuador and Venezuela is largest for manufactures.
  - Nontraditional trading-partner growth projected to improve but remain substantially below total trading-partner growth; analysis uses nominal export market shares (volume data not available).
- Price responsiveness:
  - April 2017 Western Hemisphere REO: export performance responds more to changing relative prices for noncommodity products, especially manufactures.
  - Lanau (2017b): a 10 percent depreciation increases value-added growth of the nontraditional sector by about 0.8pp, mostly via exports; effects significant on impact but take three years to materialize in full.
- Historical episodes of large depreciations (sample):
  - 61 large depreciations analyzed (17 in LA) across 53 commodity exporters in 1985−2015.
  - Large depreciation defined: REER depreciation of at least 20 percent in two years not fully reversed in the third year; episodes with a second large depreciation within five years excluded.
  - Median commodity exporter outcomes:
    - Improves nontraditional market share by 9 percent in the three years following a large depreciation.
    - Cumulative gain of 4.8 percent in five years (gains dissipate somewhat).
    - For the median country, the real depreciation is largely permanent.
    - Commodities continue to dominate: share of commodities in total exports falls by just two percentage points for the median country.
- Colombia scenario:
  - If Colombia tracked the median episode, cumulative nontraditional export growth in 2015−20 would reach 30 percent.
  - Estimated current-account improvement: about 0.25pp (assumes imports do not increase).
  - Potentially larger gains if Colombia performed closer to the top quartile in the sample.

### E. Structural Impediments to Export Growth
- Tariffs:
  - Average tariffs have fallen to around 6.5 percent.
  - Tariff gap to OECD average narrowed to about 1.5 percentage points.
  - Tariff dispersion high across sectors, especially agriculture:
    - Nominal tariffs on agricultural products average 18 percent.
    - Tariffs as high as 49 and 70 percent for dairy and beef products (Fedesarrollo, 2017).
- Non-tariff costs and other barriers:
  - Nontariff costs account for three quarters of the cost of imports; Garcia and others (2016): cost of importing was 36 percent of a product’s price in 2012.
  - Three main areas of nontariff costs:
    - Non-tariff barriers: number of nontariff barriers per product category grew from 4 in 2001 to 16 in 2014; for some categories (textiles, wood) 100 percent of imported value subject to measures; examples include inspections, permits, departmental stamps that may unduly restrict competition.
    - Infrastructure and transportation: significant infrastructure gap increases transport costs to/from borders; 4G infrastructure projects could close gap; inefficient service providers, high bargaining power of truck drivers, barriers to entry, old truck fleet, many one-truck companies, restrictions on truck ownership; exporter survey: 75 percent of respondents cite freight prices and bad roads as a top impediment.
    - Customs procedures: Colombian customs relatively slow per World Bank data, especially for exports; authorities making progress with new port scanners and accelerated processing.
- Figure references in source: tariffs (Figure 7) and time to import/export (Figure 8) illustrate these points.

### F. Conclusion
- Structural facts:
  - Exports heavily concentrated in commodities; comparative advantage in noncommodity products was weak in 2013−15; export diversification low partly due to the commodity-price boom.
- Recent performance:
  - Exports grew moderately recently but in line with historical relationships given fundamentals; weak growth reflects adverse external conditions (low oil prices, weak trading-partner growth, especially Ecuador and Venezuela), not clear underperformance.
  - Weak trading-partner growth largely offset the positive effects of the 30 percent real depreciation, particularly for nontraditional exports.
  - Significant lag exists in opening/reopening markets, implying a delayed pick-up in nontraditional export growth.
- Outlook and potential:
  - Staff models predict an acceleration in export growth given global growth assumptions.
  - Historical median commodity-exporter experience after large depreciations (4.8 percentage point increase in nontraditional market share over five years) would imply cumulative nominal export growth of 30 percent in Colombia in 2015−20.
- Policy implications:
  - Rapid correction of structural impediments (reduce tariff dispersion, cut proliferation of nontariff barriers, improve infrastructure, address inefficiencies in road transport, speed up customs procedures) and dissemination of information about trade agreements would speed and support medium-term export growth.

### Annex I. Technical Details
- Export metric definitions:
  - Revealed Comparative Advantage (RCA): share of export j in total exports of country i over share of product j exports worldwide in total worldwide exports.
  - Hausmann Complexity of Exports Index: captures diversity and ubiquity of country i’s exports (see Hausmann et al., 2013).
  - Herfindahl Concentration Index: squared sum of the share of export j in total exports of country i.
- Export performance model (specification summary):
  - Exports growth x_{i,t} modeled as function of lags of REER changes, commodity price changes, and export-weighted real GDP growth in partner countries; Colombia regressions include dummies for 2013Q2 and 2014Q2 to capture an erratic spike and trough.
  - Four different models estimated; final over/underperformance measure is a weighted average of the four (weights based on root mean square error); divide by count of forecast quarters to obtain weighted average export forecast deviation by country.

*Source: IMF staff chapter “1. The Structure of Colombian Exports” (April 16, 2018).*

### 1.      Colombia has achieved a significant reduction in labor informality over the past years.

### cr18129-colombiasi - 1.      Colombia has achieved a significant reduction in labor informality over the past years.

### Overview and key statistics
- The share of workers nationwide who are not contributing to social security dropped from 70 percent in Q1 2007 to 62.3 percent in Q3 2017.
- Depending on the definition used, labor informality fell by 3.3 to 8.2 percentage points between 2007 and 2017.
- Using the social security criterion, informality stood at 63 percent as of Q3 2017, relative to 70 percent in 2007.
- Using the size criterion (urban concept), informality dropped below 50 percent for the first time in early 2014.
- Firm informality estimates lie between 45−65 percent.
- Estimates suggest that between 20−40 percent of output is produced by the informal sector; Medina and Schneider (2018) find the informal sector accounts for 21−33 percent of GDP in recent years; Anif 2017 reports the shadow economy (informal sector 33.5 percent + illegal sector 6.3 percent) at around 40 percent of GDP.
- In 2016, according to IDB SIMS data, 67.3 percent of salaried workers were formal, but only 12.7 percent of independent workers were formal.
- Expansion of higher education coverage: from 37 percent in 2010 to 52 percent of the age cohort in 2017.
- Decomposition result: the increase in the average level of education accounts for two-thirds of the fall in labor informality between Q1 2007 and Q2 2017.
- Subnational exercise: a higher score on the higher education index explains between 44−49 percent of cross-sectional variation in informality across 23 largest cities and their departments; regional differences in the World Bank Doing Business index explain 1−5 percent of the variation.
- Between 2007 and 2017 formality rates by education group: no more than primary education increased from 12 to 15 percent; some type of secondary education from 37 percent to 40 percent; tertiary education from 71 to 73 percent. Share of workers with only primary education or less dropped from half to less than 40 percent; share with higher education increased by 7 percentage points to 23 percent.

### Policy measures and their impacts
- Major formalization initiatives cited:
  - 2006 Entrepreneurship Law.
  - 2010 Formalization and Job Creation Law.
  - 2012 Tax Reform which cut payroll taxes by 13.5 percentage points from close to 30 percent for workers earning below 10 minimum wages (reducing mandatory contributions from 29.5 percent to 16 percent of gross wage earnings for workers earning less than 10 minimum wages).
- Impact evidence:
  - Available impact evaluations suggest the 2012 tax reform led to several hundred thousand additional formal jobs and reduced labor informality.
  - Kugler et al. (2017) show the 2012 tax reform reduced informality, especially for workers in smaller firms, and those close to the minimum wage.
  - Morales and Medina (2016) also find a positive impact on formal-sector job creation.
- Other 2010 measures included: simplified bureaucratic procedures, reduced payroll taxes and social security contributions for small firms, micro-credit for individuals under 28, technical training, and financial support programs.

### Role of education and human capital
- Informality decreases sharply with education levels: 85 percent of workers with postgraduate education are formal, but only 9 percent of those without any education are formal (earlier paragraph); elsewhere: a worker with a postgraduate degree is nine times as likely to be formal than a worker without any education, and twice as likely as a worker with a high school degree, but no tertiary education.
- Two separate exercises demonstrate the paramount role of human capital:
  - Decomposition of the fall in informality shows changes in the size of education groups dominate over within-group changes; the increase in average education accounts for two-thirds of the fall in informality between Q1 2007 and Q2 2017.
  - Cross-sectional analysis across regions finds a higher education index explains 44−49 percent of variation in informality across 23 largest cities and departments.
- Policy implication: continued focus on providing high quality education is key for further gains in productivity and formality. Specific suggestions include strengthening coverage and quality of education, expanding higher education supply, and programs to support access for students from low income families, mindful of fiscal constraints.

### Structural impediments and reform recommendations
- Remaining impediments highlighted:
  - Skill mismatches, especially at the technical level.
  - Non-wage labor costs remain high even after the 2012 tax reform (e.g., 4 percent payroll tax to finance Cajas de Compensación Familiar).
  - Complex bureaucratic steps and firm entry costs (registro mercantil fees, red tape).
  - National minimum wage is much more binding in some regions than others, promoting informality where it is binding.
- Policy recommendations:
  - Consider alternative sources of financing for the Cajas de Compensación Familiar and review their services to avoid duplication with other government programs.
  - Consider a regionally differentiated minimum wage.
  - Reduce registration fees for the registro mercantil for small firms.
  - Cut red tape through the Ventanilla Unica Empresarial to improve firm formality.
  - Combine reforms in a package and sequence labor market reforms prior to product market reforms, as suggested by related literature.

### Model simulations and scenario findings
- Staff calibrated the SPR STRESS model to capture the trade-off: formal firms access larger markets while informal firms avoid regulations and taxes.
- Simulation results:
  - Permanently cutting entry costs in the formal sector leads to a steady state with lower informality and higher GDP.
  - Labor market reforms such as lower formal-sector workers’ bargaining power and reducing payroll taxes similarly lead to a steady state with lower informality and higher GDP.
- Literature suggests combining reforms and sequencing labor market reforms before product market reforms is often more beneficial.

### Measurement and definition issues
- Informality is multi-dimensional; can be studied from worker or firm perspective.
- Common worker-based measure in Colombia: workers who do not contribute to social security (used by Ministry of Labor and DNP).
- DANE official (size-based/occupational) definition: informal workers are (i) work in firms with 5 or fewer employees; (ii) are business owners of firms with 5 or less employees; (iii) are unpaid family aids and housekeepers; or (iv) are self-employed, with an educational level less than University or Technical education. DANE’s preferred definition is purely urban (major 13 cities or 23 cities with metropolitan areas). Firm size criterion changed from 10 to 5 employees starting in 2010.
- Firm informality is harder to measure given administrative and survey data limitations; evidence is sporadic and limited in geographic coverage.

*Prepared by Frederik Toscani (WHD) and Zsuzsa Munkacsi (SPR).*

### 25.      At the subnational level, access to

### cr18129-colombiasi - 25.      At the subnational level, access to

### Education and Informality: Empirical Findings
- Access to and quality of higher education is highly associated with labor formality at the subnational level.
- There exists large regional variation in informality in Colombia which can be exploited to gauge the determinants.
- Simple cross-sectional regressions suggest that education, and, in particular, access to and quality of higher education (which includes technical professional, technological and university education) has by far the largest explanatory power, outweighing even GDP per capita.
- The higher education index explains 44−49 percent of the cross-sectional variation.
- Regional differences in the World Bank Doing Business index explain 1−5 percent of the variation.
- The higher education index (‘Indice de Progreso de la Educacion Superior (IPES)) is compiled annually by the Ministry of Education and combines:
  - the share of 17−21-year-old enrolled in higher education,
  - the share of students with high scores in standardized exams,
  - the share of students who graduate from their course within 14 semesters.
- Expansion of higher education coverage from 37 percent in 2010 to 52 percent in 2017 is an important ongoing development.

### Taxes, Regulation, and Administrative Burdens
- Colombia scored 123th out of 137 countries in the World Economic Forum burden of government regulation index, indicating complex bureaucratic steps as a significant barrier for formal firms.
- Complexity arises from regional variation in regulations and complicated procedures for paying departmental and local taxes.
- Policy actions underway:
  - Implementation of the ventanilla unica empresarial to streamline and bundle necessary steps.
  - Recommendation for regional governments to harmonize practices in line with best performing regions.
- Firm entry costs:
  - Firms must register with the Chambers of Commerce through the registro mercantil and renew annually by paying a fee.
  - Costs of registration are higher than in peer countries and particularly costly for smaller firms, disincentivizing formality.
  - Suggested consideration: different and less regressive contribution systems while preserving the registro’s positive role.
- Tax burden for small and micro firms:
  - Gomez and Steiner (2015) find the average total effective tax rate is around 50 percent of profits and for small firms it can be as high as 100 percent.

### Labor Market Institutions and Costs
- Minimum wage:
  - At close to 90 percent of the median wage in 2015, Colombia had the highest ratio of minimum wage to median wage of full-time employees in the OECD sample.
  - The wage distribution is compressed just above the minimum wage; minimum wages are particularly binding in regions with lower labor productivity.
  - Reform suggestion: making the minimum wage vary by groups of regions.
- Non-wage labor costs:
  - According to Anif (2015), total non-wage labor costs, including pensions, health, holidays, parafiscales, and others, add up to 39-52 percent of the wage of a worker.
  - Of the 16 percentage points payroll tax, 4 percentage points are used to pay for Cajas de Compensacion Familiar which provide bundled benefits.
  - Recent reform suggestions propose finding alternative financing sources for the funds to further lower payroll taxes in the spirit of the 2012 tax reform.
- Social security distortions:
  - A part of formal workers’ health care contributions is used to finance the subsidized system, acting as a tax.
  - The subsidized system offers nearly the same services as the contributory one; a clearer link between contributions and benefits could make formality more attractive.
- Enforcement and outreach:
  - Authorities have increased available resources for enforcement; the Ministry of Labor is streamlining IT systems and agency interactions to improve enforcement of labor laws.
  - The Ministry of Labor is reaching out to rural areas through the ‘Red Nacional de Formalizacion Laboral’ and started formalization fairs in 2017 to facilitate immediate steps toward formalization.

### Modeling Reform Options: STRESS Model Calibration and Steady State
- Model type:
  - Small open-economy dynamic general equilibrium model with unemployment due to hiring costs and wage bargaining, and endogenous firm entry (STRESS model).
  - Distinction between formal and informal sectors in labor and goods markets.
- Key steady-state calibrated differences (Formal Sector vs. Informal Sector):
  - Output: 3.8 vs. 1.5
  - TFP: 1 vs. 1
  - Capital Share: 0.34 vs. 0.34
  - Capital Stock: 24.99 vs. 12.04
  - Employment: 0.32 vs. 0.55
  - Capital per employee: 78.09 vs. 21.89
  - Labor productivity: 11.88 vs. 2.73
  - Retailers Exit Rate: 0.125 vs. 0.35
  - Probability of Hiring: 0.05 vs. 0.7
  - Probability of Firing: 0.09 vs. 0.7
  - Tax rate on employees: 0.04 vs. 0
  - Tax rate on employers: 0.16 vs. 0
  - Worker bargaining power: 0.78 vs. 0.2
  - Retailer Entry Costs: 2.89 vs. 0.06
- Calibration targets (Model vs Data):
  - Private Investment as share of GDP: 0.24 vs. 0.23
  - Openness as share of GDP: 0.40 vs. 0.39
  - Unemployment rate: 13.30 vs. 13.12
  - Shadow economy as share of GDP: 0.30 vs. 0.21 to 0.4
  - Informal employment as share of total employment: 0.63 vs. 0.66
  - Ratio of formal to informal sector wages: 2.87 vs. 2.78

### Simulation Results: Steady-State Impact of Reforms
- Labor market reforms and cutting firm entry costs lower informality in the long-run by increasing formal sector employment.
- Steady-State Impact of Reforms (changes reported exactly):
  - Reduction in Firm Entry Costs by 10 percent:
    - Labor Informality (percentage points change): -0.2
    - Formal Employment (percentage points change): 0.4
  - Reducation in Payroll tax from 16 to 14 percentage points:
    - Labor Informality (percentage points change): -0.3
    - Formal Employment (percentage points change): 0.1
  - Reducation in Formal/Informal Worker Bargaining Power Wedge by 10 percent:
    - Labor Informality (percentage points change): -1.2
    - Formal Employment (percentage points change): 3.5
- Previous work suggests combining reforms in a package and sequencing labor market reforms prior to product market reforms is often favorable.

### Policy Recommendations and Implications
- Continue strengthening coverage and quality of education, building on expansion from 37 percent in 2010 to 52 percent in 2017.
- Expand programs to allow low-income students to access high-quality tertiary education and expand the supply of education.
- Develop and implement the national framework (marco nacional de cualificaciones) to ensure higher education degrees are pertinent to national needs and reduce skills mismatches.
- Streamline administrative procedures (e.g., ventanilla unica empresarial) and harmonize regional practices to reduce bureaucratic burdens.
- Consider less regressive registration fee structures for registro mercantil to reduce disincentives for small and micro firms to formalize.
- Consider regional differentiation of minimum wage to account for productivity differences across regions, accompanied by adequate offsets.
- Explore alternative financing for Cajas de Compensacion Familiar to lower payroll taxes while preserving worker benefits.
- Improve clarity between social security contributions and benefits to make formality more attractive.
- Strengthen enforcement and outreach (IT systems, Red Nacional de Formalizacion Laboral, formalization fairs) to increase costs of informality and protect vulnerable workers.
- Sequence and combine labor and product market reforms, prioritizing labor market reforms before product market reforms for favorable outcomes.

*International Monetary Fund — Colombia staff report (excerpt).*

### 1.      The introduction of the fiscal rule in 2011 further strengthened Colombia’s fiscal

### The introduction of the fiscal rule in 2011 further strengthened Colombia’s fiscal framework

### Fiscal rule, shock response, and fiscal adjustment since 2014
- The fiscal responsibility law of 2003 provides a qualitative anchor including publication of an annual medium-term fiscal framework compatible with debt sustainability for the broad public sector (NFPS).
- The fiscal rule law of 2011 added a quantitative anchor to the central government, focusing on a structural balance target (adjusted by GDP and oil prices) with some degree of policy countercyclicality.
- Colombia’s response to the 2014−15 large oil price shock was guided by the fiscal rule and combined revenue measures, expenditure restraint and some widening of the headline deficit.
- Shock and fiscal impact (2013–2017):
  - Fiscal cost of declining oil revenue: 3.1 percent of GDP.
  - Increasing interest rate bill (partly due to peso depreciation): 0.6 percent of GDP.
  - Combined shock: 3.7 percent of GDP.
  - Direct application of the structural balance formula allowed the deficit to widen by about 1.7 percent of GDP between 2013−16.
  - 2013−17 net impact: widening of 1.3 percent of GDP (about 35 percent of the oil revenue+interest bill combined shock).
  - Revenue measures (two tax reforms 2014 and 2016, plus tax administration gains): about 1.3 percent of GDP of non-oil revenue (absorbing about 35 percent of the shock).
  - Primary expenditure cuts required: about 0.5 percent of GDP; capital expenditure cut: 0.8 percent of GDP.
  - 2017 outturn benefited from a one-off telecom fine: 0.5 percent of GDP.

### Baseline medium-term adjustment outlook (authorities’ plan)
- Structural deficit path:
  - Structural deficit declines from 1.9 percent of GDP in 2017 to 1 percent of GDP by 2022.
  - Headline deficit will have to decline by 2.5pp of GDP by 2022.
- Composition of the headline deficit narrowing to 2022:
  - One half (1.2 pp of GDP) of the narrowing will come from higher oil revenues and lower interest expenses (partly from better-than-expected Ecopetrol profitability in 2017).
  - Non-oil tax revenue will decline over the medium-term due to expiration of the corporate surcharge CREE and one-off telecom fine in 2017 outweighing tax administration and formalization gains from the 2016 reform.
  - Primary expenditure cuts of about 1.7 percent of GDP will be required by 2022.
  - Fixed capital formation cuts by 1.1 percent of GDP by 2022.
- Debt outlook:
  - Gross public debt will decline by about 9 pp of GDP between 2018 and 2023 to about 40 percent of GDP (15 pp below the emerging market average of 55 percent of GDP).
- Fiscal stance and output gap:
  - Front-loaded fiscal adjustment with most negative fiscal impulse by 2020.
  - Negative fiscal impulse from Central Government measured by change in non-oil primary structural deficit: 0.5 percent of GDP in each 2019 and 2020.
  - Combined CPS-level negative impulse: about 0.4 percent of GDP in 2019 and 0.9 percent of GDP in 2020.
  - For 2021 and 2022 fiscal stance implies a small positive impulse (0.1 percent of GPD) at the CPS level.
  - Colombia’s output gap expected to average about 1.3 percent of potential GDP during 2018−20.
- Near-term growth impact:
  - Public investment at the CPS level will decline by about 1 percent of GDP between 2017 and 2020.
  - Fiscal multiplier of capital expenditure estimated between 0.15 and 0.46; expected decline in investment could subtract about 0.3 percentage points of GDP growth.

### Alternative fiscal adjustment scenario: shift consolidation to 2021/2022
- Scenario features:
  - Structural deficit kept constant during 2019 and 2020 at 1.9 percent of GDP.
  - Creates space for 0.2−0.5 percent of GDP of additional room for primary expenditure to protect key investment and social programs.
  - Gross public debt ratio would increase by about 0.8 pp of GDP by 2020 vis-à-vis the baseline but would remain in a declining path.

### Public expenditure efficiency and efficacy (Expert Commission findings)
- Context and priority:
  - Commission called to improve expenditure progressivity and efficiency; emphasized that additional revenue measures will be needed to protect growth-enhancing programs.
  - Recommendations particularly relevant amid medium-term expenditure restraint imposed by the fiscal rule and persistent social needs.
- Aggregate budget and institutional recommendations:
  - Unify current expenditure and investment budget on a program-based fashion.
  - Carry out detailed evaluation of all sources of budget inflexibility (e.g., legally-mandated expenses and transfers).
  - Improve flexibility of the oil-royalty system.
  - Create a Fiscal Council to cost government initiatives and assess overall fiscal policy.
- Social outcomes and subsidy distribution (2015 data):
  - Income inequality (Gini): declined from 55.7 percent in 2009 to 50.8 percent in 2017.
  - Poverty (multidimensional; national definition): declined from 30.4 percent in 2010 to about 17 percent in 2017.
  - Social subsidies distribution (percent of GDP and by quintile):
    - Education (includes job training): Total 2.97 percent of GDP; distribution by quintile: 25.7, 23.4, 21.4, 18.1, 11.4 (quintiles 1 to 5).
    - Pensions (includes Colombia Mayor): Total 2.31 percent of GDP; distribution by quintile: 14.3, 7.8, 13.7, 23.4, 50.8.
    - Health: Total 1.85 percent of GDP; distribution by quintile: 33.7, 23.6, 19.7, 15.1, 8.0.
    - Total social subsidies: 9.02 percent of GDP; distribution by quintile: 22.4, 19.9, 18.8, 18.8, 20.2.
- Pension system diagnosis:
  - Low coverage: 24 percent of the elderly have a pension.
  - Regressive distribution: about 75 percent of pension subsidies are received by the top two quintiles; only 12 percent received by the lowest two quintiles.
  - Most fiscal pension cost (3.7 percent of GDP) stems from special sub regimes (e.g., teachers, military, police).
  - Low-income programs are underfunded (benefit per person only represents about 5 percent of GDP per capita).
- Pension reform options:
  - Parametric changes: increase retirement age and link it to life expectancy; increase the number of years used to compute pension benefit (from 10 years to 20 years); increase pension contribution (18 percent); tax high pension benefits.
  - Use parametric savings to strengthen non-contributory programs (such as BEPS and Colombia Mayor).
  - Structural options: phase out PAYG system entirely or transform it into a small pillar 1 complementing the defined-contribution system.
  - Homogenize retirement age and benefits in special sub regimes with the general system.
  - Increase labor formality to expand pension coverage.
- Other sectoral recommendations:
  - Health: focus more on prevention; provide incentives for correct quantification of cost and benefits of medical services; reduce drug costs by addressing monopolistic structures; redirect revenue from a remaining payroll tax (cajas) to fund health services.
  - Education: improve access to quality early education; create clearer oversight agency; ensure stable funding not linked to a particular tax revenue; strengthen local government funding and improve distribution to reduce quality dispersion; link distribution to public universities to quality and efficacy.
  - Subsidies: improve data on fiscal resources used for subsidies; focus on fewer better-monitored subsidy programs; strengthen evaluation of largest programs.

### Subnational fiscal performance and implications
- Role and framework:
  - Subnational governments are an important part of Colombia’s fiscal framework; annual medium-term fiscal framework includes qualitative targets for NFPS which includes subnationals.
  - Since 1997, the traffic-light law imposes limits to subnational indebtedness; subnationals are main target of bi-annual oil-royalty system.
- Fiscal behavior and constraints:
  - Subnational headline balance has varied significantly with strong improvements in first year of local administrations (e.g., 2012, 2016) due to limited inter-administration coordination and need to build capacity and draw development plans.
  - Fiscal restraint at subnational level often reflected in lower public investment.
  - Multi-year public investment projects hindered by capacity constraints leading to accumulation of unused resources (financial asset accumulation).
- Subnational financial position (2011 vs 2017, in percent of GDP):
  - Gross Debt: 1.6 (2011) → 1.5 (2017)
    - Domestic: 1.2 → 1.2
      - Banks: 0.08 → 0.8
      - Bonds: 0.05 → 0.4
    - External: 0.03 → 0.3
  - Financial Assets: 2.5 → 4.1
    - Domestic: 2.2 → 2.8
    - External: 0.3 → 1.3
  - Net Debt: 0.9 → 2.6
- Growth implications:
  - Estimates by BanRep’s Fisco model (Rincon et al 2015) and IMF’s FSGM (Andrle et al., 2015) suggest public investment has a higher growth dividend than public consumption and can foster higher private investment.
  - BanRep’s Fisco model suggests private investment might decline after a surge of public consumption if the deficit/debt increase leads to future higher distortionary taxation.

*Prepared by Daniel Rodríguez-Delgado.*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2018/cr18129-colombiasi.pdf_