## cr18308-mexicoselectedissues

## Source details

**Canonical URL:** [cr18308-mexicoselectedissues](https://www.imf.org/-/media/files/publications/cr/2018/cr18308-mexicoselectedissues.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2018/cr18308-mexicoselectedissues.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2018/cr18308-mexicoselectedissues.pdf.json)

---

### Stylized facts — formality and wages
- Definition of informal workers (INEGI criteria): non-agricultural informal firms, self-employed agricultural workers, unpaid workers, non-salaried workers (at both formal and informal firms), and workers without access to social security health services in both formal and informal firms. None have access to IMSS benefits. All other workers are defined as formal.
- Formally employed share: formality increased from 42 percent in 2005 to 44 percent by the end of 2017.
- Composition of informality across firm types:
  - Around 22 percent of workers work at formal firms but are informal (not salaried or without full benefits).
- Formalization by sector (2017 data):
  - Agriculture: 9 percent formal
  - Construction: 20 percent formal
  - Other services: 48 percent formal
  - Manufacturing: 64 percent formal
- Share of informal workers employed at formal firms by sector:
  - Construction: 8 percent of informal workers work at formal firms
  - Other services: 33 percent
  - Manufacturing: 42 percent
- Wage distributions and inequality:
  - Variance of log wages: .74 in the formal sector and .65 in the informal sector.
  - Formal jobs are better paid by 49 percent on average (unadjusted).

### Labor market implications of informality — measured gaps, returns, and life-cycle effects
- Measured wage differentials and decomposition:
  - Raw average formal pay premium: 49 percent.
  - After controlling for education and demographics, residual gap reduces to 24 percent; thus differences in observable characteristics account for 51 percent of the overall wage gap.
  - After controlling for individual fixed effects (worker fixed effects), estimated wage premium falls to 6 percent; observable and unobservable characteristics (for switchers) account for 88 percent of the overall wage gap.
- Returns to education:
  - Returns to education (relative to no education) are lower for informal workers than for formal ones, particularly for completed high school and professional degrees.
- Wage growth over the life cycle:
  - Conditional wage growth profiles show significantly lower returns to experience for informal workers than for formal workers.
  - Formal workers peak wages around age 50; peaks occur before the mid-40s for informal workers.
  - Informal workers at formal firms (non-salaried without health benefits) exhibit lower wage growth than informal workers at informal firms (self-employed).
- Implications:
  - Informality can depress human capital accumulation (schooling incentives and on-the-job training), implying potential long-term productivity costs.

### Regulatory drivers of informality
- Differential treatment and compliance costs:
  - Formal salaried workers require enrollment in IMSS and employer contributions proportional to wages with a regressive fixed cost component.
  - Non-compliance fines: in the range of 20–350 daily minimum wages per non-registered worker.
  - Formal salaried workers face state payroll taxes not applied to non-salaried informal workers.
  - Federal income taxes are withheld for salaried workers; non-salaried workers file directly (leading to greater tax evasion among the latter).
- Hiring/firing costs:
  - Formal firms hiring salaried workers can be sued for unfair dismissals, creating contingent liabilities; the 2017 labor reform limited firing costs and aimed to facilitate dispute resolution processes, but implementing regulation is yet to be passed.
- Broader distortions and research findings:
  - Non-contributory benefits, limited value of contributory benefits, size-specific tax regimes (Repeco), and enforcement policies induce labor and capital misallocation toward the informal sector.
  - Formalization frictions lead to significant aggregate TFP losses in Mexico (summary conclusion from Levy (2018) cited).

### Policy trends affecting formalization
- Two main recent policy trends that likely widened the incentive gap:
  - Significant increases in the collection of payroll taxes applicable only to salaried workers. (Levy (2018) estimates an increase of 1.9 percent of GDP in the collection of income and payroll taxes from formal workers.)
  - Expansion in coverage and size of non-contributory programs (non-contributory health and pension programs) which lowers the relative benefits of contributory programs.
- Proposed minimum wage increases by the incoming administration are likely to contribute to the incentive gap.

### Minimum wage policy and informality — empirical associations
- Historical pattern:
  - Real minimum wages collapsed in the 1990s, remained relatively stable in the 2000s, and started rising significantly in the past three years (period covered), when minimum wages went from 37 to 40 percent of median wages.
  - According to comparable OECD data, the ratio of minimum to median wages in Mexico remains below OECD peers and other Latin American economies.
- Empirical specification used to assess municipal-level associations:
  - Percentile regressions and formalization regressions estimated:
    - Percentile_mit = γ Min_Wage_mt + β X_mt + ψ_t + θ_m + ε_mt
    - Formality_mt = γ Min_Wage_mt + β X_mt + ψ_t + θ_m + ε_mt
  - Min_Wage_mt is the ratio of minimum to median wages in municipality m at time t; X_mt includes municipal demographics (mean age and education levels); ψ_t are time fixed effects; θ_m are municipality fixed effects. Robustness includes municipality-specific linear time trends. Dependent variables: 10th, 25th, 50th, 75th, and 90th percentiles and share of formal workers.
- Findings on distributional effects:
  - Past minimum-to-median wage increases were associated with larger wage increases among the lowest percentiles of the distribution of both formal and informal workers, consistent with spillover effects from minimum wages.
- Associations with informality:
  - Past increases in minimum to median wage ratios were associated with increasing informality.
  - Coefficients on minimum-to-median wages using formality as the dependent variable show significant negative coefficients with and without a vector of controls.
  - Pattern: informality increased the most in municipalities that experienced the greatest increases in minimum-to-median wages.
  - Effects appear to be driven by both movements from formal to informal firms and by movement from formal to informal contractual arrangements within formal firms.
  - Caveats:
    - Documented patterns reflect associations that do not prove a causal relationship.
    - Past minimum wage changes over 2005–17 have been relatively smooth and may not be informative of non-linear effects from larger abrupt policy changes.
- Overall implication:
  - Minimum wage increases, while potentially inequality-reducing, risk increasing informality.

### Conclusions and policy recommendations — labor market and minimum wage
- Main findings on informality and labor market dynamics:
  - Informality tends to select workers with lower earnings potential and limits their development.
  - Informality is more prevalent among younger and less educated workers.
  - Informality appears to lead workers towards a path of limited earnings and perhaps limited skill growth potential.
  - Firm-data evidence documents lower output levels and growth in informal firms, highlighting a channel linking labor market duality and aggregate productivity.
- Long-term reform directions:
  - Reduce dependence on payroll taxes that are biased towards formal salaried workers while transitioning towards a social insurance system that provides good-quality services for all, irrespective of their salaried/non-salaried status.
  - Ease firing and hiring restrictions of salaried workers while increasing protections to the unemployed through a more universal unemployment insurance scheme.
  - Such profound long-term transformations should be implemented only after careful review of policy alternatives guided by experiences in other countries and detailed impact analysis.
- Short-term reform guidance:
  - Build towards a system where the non-exclusive targets of boosting social protection and removing distortionary restrictions are achieved.
  - Policy proposals, such as hikes in the minimum wage, should be gradual, viewed in the context of other distortionary policies, and carefully weigh equity benefits against the potential displacement of labor towards unproductive informality.

### Firm life cycle analysis — methodology
- Objective:
  - Understand why firms in Mexico do not invest more and thus remain relatively small and unproductive.
- Data and measurement:
  - Use several waves of the Mexican Economic Census; firm size measured as number of employees.
  - Define age groups in five-year intervals: 0–4 years, 5–9, 10–14, 15–19, 20–24.
  - Construct cohorts born in census waves 1993, 1998, 2003, and 2008 and follow cohort evolution across waves where observable.
  - Normalize firm sizes to one at birth to aggregate across states and sectors.
- Improvements over prior approaches:
  - Follow the same cohort of firms over time (unlike cross-sectional age-group averages).
  - Include the full universe of non-agricultural firms in Mexico, including small and informal ones.
  - Distinguish life cycles by 6-digit industry and state, allowing regressions with state, industry, and wave fixed effects.
  - Address attrition bias by regressing life-cycle levels on cohort survival shares and using the regression error term as a corrected life cycle.

### Firm life cycle analysis — descriptive findings
- Aggregate and median dynamics:
  - Mexican firms no longer grow after some 10-15 years of age.
  - The median sector in Mexico sees firms less than double in size before stagnating after 10-15 years of age.
  - After correcting for potential attrition bias, the picture is somewhat more pronounced but still shows stagnation.
- Comparison with other countries/sectors:
  - Hsieh and Klenow (2014) estimate that manufacturing firms in Mexico stagnate after reaching around two times their initial size around age 20.
  - US firms, by contrast, continue growing throughout their lifetimes to more than seven times their initial size.
- Distributional features:
  - Only a small share of firms experiences significant growth in higher ages; even the 90th percentile remains far below the US results.
- Geographic and sectoral patterns:
  - Firms with stronger life cycles tend to be clustered along the US border and in industries related to the North American supply chain (e.g., transportation, food, basic metals).
  - Life cycles in manufacturing are weakest in Mexico’s less developed South and stronger closer to the border with the US.
- NAFTA cohort effects (illustrative examples):
  - In the transportation sector, initial cohorts appear to have experienced more growth:
    - Firms in the 1993 cohort in the motor manufacturing sector grew in size by a factor of 44 by the time they were 5–10 years of age.
    - The 1998 and 2008 cohorts experienced 10 and 1.3-fold growth, respectively.
  - These patterns may reflect a level effect from NAFTA for early cohorts or larger initial investments for later cohorts.

### Econometric specification and key regression results
- Regression specification:
  - Q_{i,s,t,c} = α_i + α_s + α_w + ∑_{AG=2}^{5} (β_{AG} D_{AG} + γ_{AG} D_{AG} X_{i,s,t}) + X_{i,s,t} + ε_{i,s,t,c}
  - Q_{i,s,t,c} is the growth the average firm experienced between birth and wave t in industry i and state s in cohort c. Fixed effects for 6-digit industry i, state s and wave t included. Age group dummies D_{AG} capture life-cycle dynamics; fundamentals X_{i,s,t} computed from Mexican Economic Census and SIMBAD municipal-level indicators.
- Sample size and model fit:
  - Regressions make use of up to 56,800 observations.
  - R-squared by regression: 0.071 (Reg 1); 0.220 (Reg 2); 0.239 (Reg 3); 0.253 (Reg 4); 0.264 (Reg 5); 0.253 (Reg 6); 0.250 (Reg 7).
- Life-cycle coefficients (selected):
  - Dummy 5-9 years: 0.634*** [0.030] (Reg 1); 0.478*** [0.007] (Reg 2).
  - Dummy 10-14 years: 0.969*** [0.040] (Reg 1); 0.667*** [0.009] (Reg 2).
  - Dummy 15-19 years: 1.177*** [0.057] (Reg 1); 0.777*** [0.012] (Reg 2).
  - Dummy 20-24 years: 1.373*** [0.146] (Reg 1); 0.810*** [0.020] (Reg 2).
  - Log of initial size: -0.248*** [0.011] (Reg 3); -0.282*** [0.011] (Reg 4); -0.405*** [0.017] (Reg 5); -0.387*** [0.017] (Reg 6); -0.379*** [0.018] (Reg 7).
  - Share of initial cohort surviving: -0.041*** [0.008] (Reg 3); -0.048*** [0.009] (Reg 4); -0.040*** [0.010] (Reg 5); -0.038*** [0.010] (Reg 6); -0.030*** [0.010] (Reg 7).
  - Observations by regression: 56,804 (Reg 1); 55,660 (Regs 2–3); 53,117 (Reg 4); 29,390 (Regs 5–6); 25,248 (Reg 7).

### Determinants of weaker or stronger firm life cycles — empirical associations
- Informality:
  - Informality measured as the share of firms that pays neither social security contributions nor VAT.
  - Informality coefficients (informality, share): -1.379*** [0.062] (Reg 4); -1.048*** [0.098] (Reg 5); -1.242*** [0.098] (Reg 6); -1.325*** [0.109] (Reg 7).
  - Predicted life cycles: informal firms hardly grow after age 4 when controlling for other variables; formal firms grow to about 4.5 times their size by age 20–25 (based on regression predictions with interaction terms).
  - Interpretation: informal firms less likely to invest/grow (to avoid tax registration or due to lack of bank credit) and can depress investment by formal firms via unfair cost advantages.
- Market concentration:
  - Herfindahl index (based on 20 largest firms in sector-state) negatively associated with life-cycle growth.
  - Coefficients for Market concentration, Herfindahl: -0.203*** [0.052] (Reg 4); -0.377*** [0.118] (Reg 5); -0.610*** [0.119] (Reg 6); -1.047*** [0.201] (Reg 7).
  - Firms invest and grow more in less concentrated markets; firms in very concentrated industries (99th percentile) have notably weaker life cycles than in very diversified industries (1st percentile).
- Distance from large city:
  - Distance from large city (100 km) coefficients: -0.964*** [0.182] (Reg 4); -0.939*** [0.297] (Reg 5); -0.873*** [0.300] (Reg 6); -0.494 [0.347] (Reg 7; not significant).
  - Findings: more remotely located firms experience less growth than firms closer to customers, competitors, and suppliers.
- Human capital and services:
  - Illiteracy, share: -0.018*** [0.003] (Reg 4); 0.006 [0.012] (Reg 5); 0.064*** [0.012] (Reg 6); 0.102*** [0.016] (Reg 7).
  - Financial and internet access:
    - Share of firms with bank account: 0.458*** [0.019] (Reg 6).
    - Share of firms with bank credit: 0.204*** [0.021] (Reg 6).
    - Share of firms with internet access: 0.129*** [0.022] (Reg 6).
  - Interpretation: firms with better access to financial and internet services exhibit stronger life cycles.

### Firm life cycle conclusions and policy recommendations
- Main empirical conclusions:
  - Firm life cycle growth in Mexico is significantly lower than in more advanced economies such as the US.
  - The average Mexican firm appears to no longer grow after some 15 years of age and stagnates at less than double its initial size.
  - Several regions and sectors deviate from this pattern; some industries in Mexico’s North (close to US border) exhibit stronger life cycles.
- Identified distortions associated with weaker firm investment and growth:
  - High prevalence of informality — informal firms do not grow much over their life cycles while formal firms grow solidly.
  - High market concentration — associated with weaker growth, likely via undue market power.
  - Geographic remoteness from population centers — firms farther from large cities grow less.
  - Limited access to financial and telecommunication (internet) services — associated with weaker growth.
- Policy recommendations:
  - Continue and deepen structural reforms begun with the Pacto por Mexico (2012).
  - Fight informality to improve firm growth prospects and level the competitive playing field.
  - Strengthen competition policies to reduce undue use of market power and improve incentives to invest.
  - Expand access to financial services and telecommunication (internet) services for firms.
  - Implement targeted infrastructure investments to better connect remote regions to major markets.

### Coverage and targeting of social transfers and programs
- Individuals in the bottom income quintile receive less than 30 percent of total social transfers, a smaller share than in other Latin American and OECD countries.
- The farmland subsidies program Proagro (formally Procampo):
  - Benefits a greater share of households in the bottom income quintile than in any other quintile.
  - The average amount received by households in the top quintile is only marginally lower than the average transfer amount to households in the bottom quintile.
  - The average Proagro transfer to the top quintile is higher than in the middle three quintiles.
- Non-monetary medical transfers and healthcare spending:
  - Non-monetary government transfers covering medical expenses benefit the same proportion of households in all five income quintiles.
  - Transfer amounts increase with income, resulting in richer households getting a higher share of their health expenditures covered by the government.
- Government scholarships and education transfers:
  - Government scholarships for secondary education accrue more frequently to households in the top half of the income distribution, while scholarships for primary and tertiary education are more evenly distributed.
  - Scholarship amounts are much higher for students enrolled in tertiary education institutions, which are more likely to belong to top income deciles.
  - Government scholarships thus increase inequalities.
- Contribution of programs to inequality reduction:
  - Prospera and old-age social assistance programs account for 74 percent of the reduction in the Gini coefficient achieved by government transfers.

*Source: MEXICO — International Monetary Fund (selected excerpts).*

### References ____________________________________________________________________________ 11

### FORMALITY AND EQUITY—LABOR MARKET CHALLENGES IN MEXICO

### A. Stylized Facts
- Definition of informal workers (INEGI criteria): non-agricultural informal firms, self-employed agricultural workers, unpaid workers, non-salaried workers (at both formal and informal firms), and workers without access to social security health services in both formal and informal firms. None have access to IMSS benefits. All other workers are defined as formal.
- Formally employed share: formality increased from 42 percent in 2005 to 44 percent by the end of 2017.
- Composition of informality across firm types:
  - Around 22 percent of workers work at formal firms but are informal (not salaried or without full benefits).
- Formalization by sector (2017 data):
  - Agriculture: 9 percent formal
  - Construction: 20 percent formal
  - Other services: 48 percent formal
  - Manufacturing: 64 percent formal
- Share of informal workers employed at formal firms by sector:
  - Construction: 8 percent of informal workers work at formal firms
  - Other services: 33 percent
  - Manufacturing: 42 percent
- Wage distributions and inequality:
  - Variance of log wages: .74 in the formal sector and .65 in the informal sector.
  - Formal jobs are better paid by 49 percent on average (unadjusted).

### B. Labor Market Implications of Informality
- Measured wage differentials and decomposition:
  - Raw average formal pay premium: 49 percent.
  - After controlling for education and demographics, residual gap reduces to 24 percent; thus differences in observable characteristics account for 51 percent of the overall wage gap.
  - After controlling for individual fixed effects (worker fixed effects), estimated wage premium falls to 6 percent; observable and unobservable characteristics (for switchers) account for 88 percent of the overall wage gap.
- Returns to education:
  - Returns to education (relative to no education) are lower for informal workers than for formal ones, particularly for completed high school and professional degrees.
- Wage growth over the life cycle:
  - Conditional wage growth profiles show significantly lower returns to experience for informal workers than for formal workers.
  - Formal workers peak wages around age 50; peaks occur before the mid-40s for informal workers.
  - Informal workers at formal firms (non-salaried without health benefits) exhibit lower wage growth than informal workers at informal firms (self-employed).
- Implications:
  - Informality can depress human capital accumulation (schooling incentives and on-the-job training), implying potential long-term productivity costs.

### C. Regulatory Drivers of Informality
- Differential treatment and compliance costs:
  - Formal salaried workers require enrollment in IMSS and employer contributions proportional to wages with a regressive fixed cost component.
  - Non-compliance fines: in the range of 20–350 daily minimum wages per non-registered worker.
  - Formal salaried workers face state payroll taxes not applied to non-salaried informal workers.
  - Federal income taxes are withheld for salaried workers; non-salaried workers file directly (leading to greater tax evasion among the latter).
- Hiring/firing costs:
  - Formal firms hiring salaried workers can be sued for unfair dismissals, creating contingent liabilities; the 2017 labor reform limited firing costs and aimed to facilitate dispute resolution processes, but implementing regulation is yet to be passed.
- Broader distortions and research findings:
  - Non-contributory benefits, limited value of contributory benefits, size-specific tax regimes (Repeco), and enforcement policies induce labor and capital misallocation toward the informal sector.
  - Formalization frictions lead to significant aggregate TFP losses in Mexico (summary conclusion from Levy (2018) cited).

### D. Policy Trends Affecting Formalization
- Two main recent policy trends that likely widened the incentive gap:
  - Significant increases in the collection of payroll taxes applicable only to salaried workers. (Levy (2018) estimates an increase of 1.9 percent of GDP in the collection of income and payroll taxes from formal workers.)
  - Expansion in coverage and size of non-contributory programs (non-contributory health and pension programs) which lowers the relative benefits of contributory programs.
- Proposed minimum wage increases by the incoming administration are likely to contribute to the incentive gap.

### E. Minimum Wage Policy and Informality
- Historical pattern:
  - Real minimum wages collapsed in the 1990s, remained relatively stable in the 2000s, and started rising significantly in the past three years (period covered), when minimum wages went from 37 to 40 percent of median wages.
  - According to comparable OECD data, the ratio of minimum to median wages in Mexico remains below OECD peers and other Latin American economies.
- Empirical specification used to assess municipal-level associations:
  - Percentile regressions and formalization regressions estimated:
    - Percentile_mit = γ Min_Wage_mt + β X_mt + ψ_t + θ_m + ε_mt
    - Formality_mt = γ Min_Wage_mt + β X_mt + ψ_t + θ_m + ε_mt
  - Where Min_Wage_mt is the ratio of minimum to median wages in municipality m at time t; X_mt includes municipal demographics (mean age and education levels); ψ_t are time fixed effects; θ_m are municipality fixed effects. Robustness includes municipality-specific linear time trends. Dependent variables: 10th, 25th, 50th, 75th, and 90th percentiles and share of formal workers.
- Findings on distributional effects:
  - Past minimum-to-median wage increases were associated with larger wage increases among the lowest percentiles of the distribution of both formal and informal workers, consistent with spillover effects from minimum wages documented in the structural labor literature.

*Prepared by Jorge Alvarez (WHD); October 19, 2018.*

### 19.      However, past increases in minimum to median wage ratios were also associated with

### Mexico: Selected Issues — Minimum Wages, Informality, and Firm Life Cycles

### Minimum wage increases and informality
- Past increases in minimum to median wage ratios were associated with increasing informality.
- Coefficients on minimum-to-median wages using formality as the dependent variable show significant negative coefficients with and without a vector of controls.
- The results document a pattern where informality increased the most in municipalities that experienced the greatest increases in minimum-to-median wages.
- These effects appear to be driven by both movements from formal to informal firms and by movement from formal to informal contractual arrangements within formal firms.
- Caveats:
  - The documented patterns reflect associations that do not prove a causal relationship between minimum wages and both distributional and formality outcomes.
  - Past minimum wage changes over the period of study (2005–17) have been relatively smooth and may not be informative of non-linear effects that might occur from larger abrupt policy changes.
- Overall implication:
  - Minimum wage increases, while potentially inequality-reducing, risk increasing informality.

### Conclusions and policy recommendations
- Main findings on informality and labor market dynamics:
  - Informality tends to select workers with lower earnings potential and limits their development.
  - Informality is more prevalent among younger and less educated workers.
  - Informality appears to lead workers towards a path of limited earnings and perhaps limited skill growth potential.
  - Firm-data evidence documents lower output levels and growth in informal firms, highlighting a channel linking labor market duality and aggregate productivity.
- Long-term reform directions:
  - Reduce dependence on payroll taxes that are biased towards formal salaried workers while transitioning towards a social insurance system that provides good-quality services for all, irrespective of their salaried/non-salaried status.
  - Ease firing and hiring restrictions of salaried workers while increasing protections to the unemployed through a more universal unemployment insurance scheme.
  - Such profound long-term transformations should be implemented only after careful review of policy alternatives guided by experiences in other countries and detailed impact analysis.
- Short-term reform guidance:
  - Build towards a system where the non-exclusive targets of boosting social protection and removing distortionary restrictions are achieved.
  - Policy proposals, such as hikes in the minimum wage, should be gradual, viewed in the context of other distortionary policies, and carefully weigh equity benefits against the potential displacement of labor towards unproductive informality.

### Firm life cycle analysis — methodology
- Objective:
  - Understand why firms in Mexico do not invest more and thus remain relatively small and unproductive.
- Data and measurement:
  - Use several waves of the Mexican Economic Census; firm size measured as number of employees.
  - Define age groups in five-year intervals: 0–4 years, 5–9, 10–14, 15–19, 20–24.
  - Construct cohorts born in census waves 1993, 1998, 2003, and 2008 and follow cohort evolution across waves where observable.
  - Normalize firm sizes to one at birth to aggregate across states and sectors.
- Improvements over prior approaches:
  - Follow the same cohort of firms over time (unlike cross-sectional age-group averages).
  - Include the full universe of non-agricultural firms in Mexico, including small and informal ones.
  - Distinguish life cycles by 6-digit industry and state, allowing regressions with state, industry, and wave fixed effects.
  - Address attrition bias by regressing life-cycle levels on cohort survival shares and using the regression error term as a corrected life cycle.

### Firm life cycle analysis — descriptive findings
- Aggregate and median dynamics:
  - Mexican firms no longer grow after some 10-15 years of age.
  - The median sector in Mexico sees firms less than double in size before stagnating after 10-15 years of age.
  - After correcting for potential attrition bias, the picture is somewhat more pronounced but still shows stagnation.
- Comparison with other countries/sectors:
  - Hsieh and Klenow (2014) estimate that manufacturing firms in Mexico stagnate after reaching around two times their initial size around age 20.
  - US firms, by contrast, continue growing throughout their lifetimes to more than seven times their initial size.
- Distributional features:
  - Only a small share of firms experiences significant growth in higher ages; even the 90th percentile remains far below the US results.
- Geographic and sectoral patterns:
  - Firms with stronger life cycles tend to be clustered along the US border and in industries related to the North American supply chain (e.g., transportation, food, basic metals).
  - Life cycles in manufacturing are weakest in Mexico’s less developed South and stronger closer to the border with the US.
- NAFTA cohort effects (illustrative examples):
  - In the transportation sector, initial cohorts appear to have experienced more growth:
    - Firms in the 1993 cohort in the motor manufacturing sector grew in size by a factor of 44 by the time they were 5–10 years of age.
    - The 1998 and 2008 cohorts experienced 10 and 1.3-fold growth, respectively.
  - These patterns may reflect a level effect from NAFTA for early cohorts or larger initial investments for later cohorts.
  - The basic metals sector shows heterogeneity: life cycles weakened over time in some industries while they strengthened in others, suggesting sector-specific dynamics.

*Source: MEXICO — International Monetary Fund (selected excerpts).*

### 15.      In this section, we aim to better understand the determinants of firm growth over the

### 15. In this section, we aim to better understand the determinants of firm growth over the life cycle in Mexico.

### Econometric specification and data
- Regression specification:
  - Q_{i,s,t,c} = α_i + α_s + α_w + ∑_{AG=2}^{5} (β_{AG} D_{AG} + γ_{AG} D_{AG} X_{i,s,t}) + X_{i,s,t} + ε_{i,s,t,c}
  - Q_{i,s,t,c} is the growth the average firm experienced between birth and wave t in industry i and state s in cohort c.
  - Fixed effects included for each 6-digit industry i, state s and wave t.
  - Age group dummies D_{AG} capture average life-cycle dynamics for each industry-state pair.
  - Fundamentals X_{i,s,t} computed from Mexican Economic Census and SIMBAD municipal-level indicators.
  - Interactions D_{AG} X_{i,s,t} permit computing average life cycles conditional on fundamentals.
- Sample size and observations:
  - Regressions make use of up to 56,800 observations.
  - All regressions include industry, state, and wave fixed effects (omitted in tables).

### Key regression results and life-cycle patterns
- Baseline (Regression 1) with life cycles normalized to 1 in initial age group:
  - Coefficients indicate average industry sees firms grow to some 2.4 (1+1.4) times initial size after 20–24 years once fixed effects are controlled for.
- Outlier-adjusted results (Regression 2; largest and smallest 1 percent realizations of dependent variable dropped):
  - Life-cycle estimates closer to initial descriptive findings: firms grow to about 1.5–1.6 times initial size.
  - R-squared rises from 0.07 to 0.22 when outliers are dropped.
- Controls for initial size and attrition (Regression 3):
  - Log of initial size coefficient negative and significant, indicating weaker life-cycle growth when firms are large from the outset.
  - Share of initial cohort surviving coefficient negative and significant, indicating higher observed growth in sectors where a smaller share of the initial population survives (consistent with selection of more productive survivors).
- Selected coefficient estimates (from Table 2; robust standard errors in brackets):
  - Dummy 5-9 years: 0.634*** [0.030] (Reg 1); 0.478*** [0.007] (Reg 2); values across Regressions 3–7 vary but remain positive and significant.
  - Dummy 10-14 years: 0.969*** [0.040] (Reg 1); 0.667*** [0.009] (Reg 2).
  - Dummy 15-19 years: 1.177*** [0.057] (Reg 1); 0.777*** [0.012] (Reg 2).
  - Dummy 20-24 years: 1.373*** [0.146] (Reg 1); 0.810*** [0.020] (Reg 2).
  - Log of initial size: -0.248*** [0.011] (Reg 3); -0.282*** [0.011] (Reg 4); -0.405*** [0.017] (Reg 5); -0.387*** [0.017] (Reg 6); -0.379*** [0.018] (Reg 7).
  - Share of initial cohort surviving: -0.041*** [0.008] (Reg 3); -0.048*** [0.009] (Reg 4); -0.040*** [0.010] (Reg 5); -0.038*** [0.010] (Reg 6); -0.030*** [0.010] (Reg 7).
  - Observations by regression: 56,804 (Reg 1); 55,660 (Regs 2–3); 53,117 (Reg 4); 29,390 (Regs 5–6); 25,248 (Reg 7).
  - R-squared by regression: 0.071 (Reg 1); 0.220 (Reg 2); 0.239 (Reg 3); 0.253 (Reg 4); 0.264 (Reg 5); 0.253 (Reg 6); 0.250 (Reg 7).

### Determinants of weaker or stronger firm life cycles
- Informality:
  - Informality measured as the share of firms that pays neither social security contributions nor VAT.
  - Informality coefficients (informality, share): -1.379*** [0.062] (Reg 4); -1.048*** [0.098] (Reg 5); -1.242*** [0.098] (Reg 6); -1.325*** [0.109] (Reg 7).
  - Predicted life cycles: informal firms hardly grow after age 4 when controlling for other variables; formal firms grow to about 4.5 times their size by age 20–25 (based on regression predictions with interaction terms).
  - Interpretation: informal firms less likely to invest/grow (to avoid tax registration or due to lack of bank credit) and can depress investment by formal firms via unfair cost advantages.
- Market concentration:
  - Herfindahl index (based on 20 largest firms in sector-state) negatively associated with life-cycle growth.
  - Coefficients for Market concentration, Herfindahl: -0.203*** [0.052] (Reg 4); -0.377*** [0.118] (Reg 5); -0.610*** [0.119] (Reg 6); -1.047*** [0.201] (Reg 7).
  - Firms invest and grow more in less concentrated markets; firms in very concentrated industries (99th percentile) have notably weaker life cycles than in very diversified industries (1st percentile) (Figure 6 predictions).
- Distance from large city:
  - Distance from large city (100 km) coefficients: -0.964*** [0.182] (Reg 4); -0.939*** [0.297] (Reg 5); -0.873*** [0.300] (Reg 6); -0.494 [0.347] (Reg 7; not significant).
  - Findings: more remotely located firms experience less growth than firms closer to customers, competitors, and suppliers.
- Human capital and services:
  - Illiteracy, share: -0.018*** [0.003] (Reg 4); 0.006 [0.012] (Reg 5); 0.064*** [0.012] (Reg 6); 0.102*** [0.016] (Reg 7).
  - Financial and internet access (included in later regressions):
    - Share of firms with bank account: 0.458*** [0.019] (Reg 6).
    - Share of firms with bank credit: 0.204*** [0.021] (Reg 6).
    - Share of firms with internet access: 0.129*** [0.022] (Reg 6).
  - Interpretation: firms with better access to financial and internet services exhibit stronger life cycles.

### Conclusions and policy recommendations
- Main empirical conclusions:
  - Firm life cycle growth in Mexico is significantly lower than in more advanced economies such as the US.
  - The average Mexican firm appears to no longer grow after some 15 years of age and stagnates at less than double its initial size.
  - Several regions and sectors deviate from this pattern; some industries in Mexico’s North (close to US border) exhibit stronger life cycles.
- Identified distortions associated with weaker firm investment and growth:
  - High prevalence of informality — informal firms do not grow much over their life cycles while formal firms grow solidly.
  - High market concentration — associated with weaker growth, likely via undue market power.
  - Geographic remoteness from population centers — firms farther from large cities grow less.
  - Limited access to financial and telecommunication (internet) services — associated with weaker growth.
- Policy recommendations:
  - Continue and deepen structural reforms begun with the Pacto por Mexico (2012).
  - Fight informality to improve firm growth prospects and level the competitive playing field.
  - Strengthen competition policies to reduce undue use of market power and improve incentives to invest.
  - Expand access to financial services and telecommunication (internet) services for firms.
  - Implement targeted infrastructure investments to better connect remote regions to major markets.

*International Monetary Fund: Mexico — Selected Issues (section on determinants of firm growth over the life cycle).*

### 8.      Other social benefits are less well targeted. Individuals in the bottom income quintile

### 8. Other social benefits are less well targeted.

### Coverage and targeting of social transfers
- Individuals in the bottom income quintile receive less than 30 percent of total social transfers, a smaller share than in other Latin American and OECD countries (Figure 7, top left panel).
- The farmland subsidies program Proagro (formally Procampo) benefits a greater share of households in the bottom income quintile than in any other quintile, but:
  - the average amount received by households in the top quintile is only marginally lower than the average transfer amount to households in the bottom quintile, and
  - the average Proagro transfer to the top quintile is higher than in the middle three quintiles.

### Non-monetary medical transfers and healthcare spending
- Non-monetary government transfers covering medical expenses benefit the same proportion of households in all five income quintiles.
- Transfer amounts increase with income, following the growth of healthcare expenditures with income, resulting in richer households getting a higher share of their health expenditures covered by the government (Figure 7, middle right panel).

### Government scholarships and education transfers
- Government scholarships for secondary education accrue more frequently to households in the top half of the income distribution, while scholarships for primary and tertiary education are more evenly distributed (Figure 7, bottom left panel).
- Scholarship amounts are much higher for students enrolled in tertiary education institutions, which are more likely to belong to top income deciles.
- Government scholarships thus increase inequalities.

### Contribution of programs to inequality reduction
- A comparison of the effect of various government transfer programs on inequality confirms the predominant role played by Prospera and old-age social assistance programs.
- Those programs account for 74 percent of the reduction in the Gini coefficient achieved by government transfers (Figure 8).
- Figure 8 displays the Gini coefficient for total household income and counterfactuals (excluding Prospera and old-age social assistance; excluding all government transfers) over 2008–2016, with the plotted values spanning the 0.42 to 0.52 range on the vertical axis.

*Source: cr18308-mexicoselectedissues; Sources: INEGI; IMF staff calculations; World Bank ASPIRE database.*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2018/cr18308-mexicoselectedissues.pdf_