## wpiea2021051-print-pdf

## Source details

**Canonical URL:** [wpiea2021051-print-pdf](https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021051-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2021/english/wpiea2021051-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2021/english/wpiea2021051-print-pdf.pdf.json)

---

### Key findings
- Resource misallocation in India’s formal manufacturing sector is sizable and shows significant heterogeneity across Indian states.
- Eliminating misallocation to achieve US efficiency could result in manufacturing productivity gains of about 40-60 percent in India.
- Econometric evidence suggests that reforms to increase labor market flexibility are associated with higher TFP and lower misallocation, especially in states where informality is relatively high.
- India’s labor laws have remained among the more restrictive laws in the world and strict labor laws may relate to large informality in the Indian economy.

### Methodology and measures
- Uses micro-level balance sheet data to:
  - calculate aggregate total factor productivity (TFP); and
  - calculate the dispersion of firms’ revenue productivity (TFPR) as a standard measure of resource misallocation (Hsieh and Klenow (2009)).
- Measures and definitions:
  - Revenue productivity (TFPR): TFPR_jiss = P_jiss A_jiss.
  - TFPR dispersion: standard deviation of revenue productivity (in logs) from industry means; variance of log(TFPR_jis) used as main misallocation measure.
  - Skewness measure: TFPR̄_js,p50 = TFPR_js,p50 / TFPR̄_js; median-mean TFPR ratio indicates direction of skew (median = mean → no misallocation).
  - TFP gap: TFP_gap_j s = TFP_j s* / TFP_j s − 1 and state-level aggregator TFP_gap_j = (Π TFP_gap_j s ^ρ_j s).
- Key equation forms:
  - Production: Y_jiss = A_jiss K_jiss^{α_s} L_jiss^{1−α_s}.
  - Profit maximization with distortions τ^K, τ^L, τ^Y (Equation (1) as in source).
  - Sector aggregate Y_js via CES aggregator (Equation (2)); aggregate TFP expression as in source (Equation (3)).
- Assumptions:
  - TFPR and A jointly log-normal (Hsieh and Klenow (2009)); details in Appendix I.
  - Factor shares set equal to corresponding U.S. manufacturing industry at 3-digit level; labor shares scaled by 3/2 to reflect nonwage compensation.

### Empirical scope, data, and stylized facts
- Scope: formal manufacturing sector across 20 Indian states, 41 sectors, and four years (fiscal years 2003/04, 2006/07, 2008/09, and 2010/11).
- Data source: Annual Survey of Industries (ASI) compiled by CSO; covers factories under the Factories Act, 1948 (thresholds >10 workers with power or >20 without power); census of registered units with 100+ employees and a 1/5 sample of remaining registered firms.
- Sample construction and cleaning:
  - No firm-level panel due to changing random samples.
  - Replace negative values of value-added, capital, and labor compensation with missing.
  - Trim 1 percent tails of firm productivity (log TFPR) by year; trim 1 percent tails of firm relative distortions and relative productivity by 3-digit industry and year.
  - Exclude state-sector-year observations aggregated from fewer than 10 firms (keeps 46 percent of the sample).
- Stylized facts and summary statistics (selected exact figures):
  - Aggregate manufacturing TFP (in logs) across states: ranges from 3.2 in Bihar to 5.1 in Maharashtra.
  - Sector-state level variance of revenue productivity ranges between 0.9 and 2.4.
  - Twelve out of 20 states have median TFPR below state-industry mean TFPR; five states have a ratio exceeding 1.
  - Table 1 excerpts:
    - MAHARASHTRA: # Sector-Year Obs. 197, Avg # Firms per sector and year 72, Agg. TFP 5.12, Avg. Variance TFPR 0.71, Avg. Median TFPR 1.00
    - TAMIL NADU: 189, 87, 4.77, 0.67, 0.95
    - GUJARAT: 174, 67, 4.84, 0.69, 0.94
    - Average (across listed states): # Sector-Year Obs. 108, Avg # Firms per sector and year 39, Agg. TFP 4.42, Avg. Variance TFPR 0.70, Avg. Median TFPR 0.98
  - Table 3: Dispersion of Firm-Level TFPR (in logs) in India by year:
    - 2003: Median -0.20, Variance 0.79, P75/P25 0.93, P90/P10 1.86, # Firms 50,600, Share of firms with non-missing TFPR 0.68
    - 2006: Median -0.32, Variance 0.79, P75/P25 0.96, P90/P10 1.86, # Firms 59,819, Share 0.56
    - 2008: Median -0.16, Variance 0.79, P75/P25 0.95, P90/P10 1.83, # Firms 50,261, Share 0.59
    - 2010: Median -0.11, Variance 0.70, P75/P25 0.89, P90/P10 1.71, # Firms 48,120, Share 0.72
  - Cross-country comparison: variance of TFPR and the TFP gap are respectively 1.5 and 3 times higher in India in 2003 than in the U.S. in 1997.

### State characteristics and reform measures
- Employment Protection Legislation (EPL) index:
  - Index from Dougherty (2008), rescaled zero to one; computed in 2007 and time-invariant in analysis.
  - Higher EPL indicates more reforms toward labor market flexibility.
  - Examples: Andhra Pradesh and Gujarat EPL = 0.96; West Bengal and Chhattisgarh EPL = 0.5.
- Other state controls:
  - CMIE state database: logs of credit per capita, road density (kilometer per thousand square kilometers), unregistered manufacturing NSDP share of state NSDP (informality proxy).
  - Product Market Regulation (PMR) index rescaled zero to three; higher PMR → more supportive of competition.
  - World Bank Enterprise Surveys (2014): share of firms citing access to electricity or inadequately educated workforce as obstacles.

### Empirical framework and hypotheses on informality
- Baseline econometric specification: sector-state-year outcomes (TFP, variance TFPR, median-mean TFPR) regressed on EPL, informality (unregistered NSDP share), controls Z, industry-year fixed effects; state-clustered standard errors.
- Key hypothesis: easing labor market regulations will produce larger formal sector improvements where informality is high due to (i) productive formal firms constrained by informal competitors, (ii) informality endogenous to regulations, and (iii) potential extensive-margin formalization of firms.
- Interaction approach: include EPL × informality to capture differential effects by informality level; analysis intended as suggestive evidence, not causal identification of a single channel.

### Main econometric results (exact coefficients and significance)
- Table 5: Baseline (Observations 2,017)
  - Outcome (1) Aggregate TFP:
    - Employ. Protec. Legislation: 0.039
    - Share of Unregistered NSDP: -1.470***
    - Access to Electricity as an obstacle: -0.555***
    - Road density: 0.073***
    - # Firms: 0.510***
    - R-squared 0.467
  - Outcome (2) Variance TFPR:
    - Employ. Protec. Legislation: -0.164
    - Share of Unregistered NSDP: -0.207***
    - # Firms: -0.018*
    - R-squared 0.467
  - Outcome (3) Median TFPR:
    - Employ. Protec. Legislation: -0.275**
    - Share of Unregistered NSDP: 0.129
    - Access to Electricity: -0.086*
    - Road density: -0.009
    - # Firms: 0.021
    - R-squared 0.191
- Table 6: Baseline with Additional Controls (Observations 2,017)
  - Employ. Protec. Legislation: -0.120, -0.147, -0.264**
  - Share of Unregistered NSDP: -1.467***, -0.224***, 0.118
  - Product Market Regulation: -0.005, -0.045, -0.060*
  - Road density: 0.090***, 0.012, -0.007
  - # Firms: 0.480***, -0.017, 0.025*
  - R-squared 0.470 (cols 1 and 2), 0.196 (col 3)
- Table 7: EPL × Informality interaction (Observations 2,017)
  - Employ. Protec. Legislation: -0.516, -0.274***, -0.058
  - EPL*Unreg. NSDP (Median): 1.413*, 0.248*, -0.376*
  - Unregistered NSDP (Median): -1.413**, -0.230*, 0.332**
  - Road density: 0.111***, 0.017**, -0.013
  - # Firms: 0.550***, -0.012, 0.013
  - R-squared 0.446, 0.463, 0.194
  - Interpretation: easing EPL in high-informality states has a large, significant positive effect on TFP; dispersion of TFPR increases (likely extensive-margin effects) while median-mean TFPR falls (reducing negative skew).
- Table 8: Aggregate labor productivity decomposition (Observations 2,017)
  - Employ. Protec. Legislation: -0.579*, -0.587***, 0.008
  - EPL*Unreg. NSDP (Median): 1.584***, 1.071**, 0.513**
  - Unregistered NSDP (Median): -1.431***, -1.001***, -0.430**
  - # Firms: -0.024, -0.060*, 0.036*
  - R-squared 0.260, 0.380, 0.175
  - Interpretation: interaction term strongly positive for covariance term → easing EPL associated with improved allocative efficiency and higher average labor productivity in high-informality states.
- Table 9: Restricting sample to 2008 and 2010 (Observations 944)
  - Employ. Protec. Legislation: -0.627*, -0.378***, -0.167
  - EPL*Unreg. NSDP (Median): 1.241**, 0.226, -0.189
  - Unregistered NSDP (Median): -1.328**, -0.219*, 0.171
  - Road density: 0.101***, 0.008, -0.019***
  - # Firms: 0.520***, -0.005, 0.030*
  - R-squared 0.452, 0.522, 0.167
- Robustness:
  - Results qualitatively similar when using reform indexes computed in 2006/2007 and restricting firm-level data to 2008 and 2010.

### Scenario and calibration
- Calibrated scenario: advancing labor market reforms of an Indian state to the best performer (index = 1) using coefficients from Table 7 suggests significant potential aggregate TFP increases.
- States identified as likely to benefit most in absolute TFP gains in this scenario: West Bengal, Maharashtra and Jharkhand (Figure 8 referenced in source).

### Box 1 — Labor Market Regulations and Reforms in India (selected points)
- Regulatory overview:
  - Chapter V-B of the Industrial Disputes Act requires government approval for layoffs, retrenchments, and closures where it applies (factories with 100 or more workers).
  - Prior to 2017, around 250 labor laws at center and states; central government consolidated 44 central labor laws into four codes (Code on Wages passed August 2019; remaining three passed September 2020). Chapter V-B of the IDA has not been altered in these reforms.
- State-level reforms and COVID-19 relaxations:
  - Examples of state reforms: Rajasthan, Madhya Pradesh, and Haryana (2014) modified IDA to allow automatic retrenchment for factories with less than 300 workers; Gujarat and Maharashtra introduced variations allowing automatic retrenchment with specified severance.
  - COVID-19 temporary relaxations included exemptions for new investors (e.g., Madhya Pradesh 1,000 days; Gujarat 1,200 days) and emergency ordinances by several states.
- Stylized relationships:
  - Lower EPL (more reform) suggestively correlated with prevalence of smaller-than-optimal firms (median-mean TFPR < 1).
  - States with high informality (share of unregistered NSDP in manufacturing > 60 percent, highest quartile) show patterns consistent with many sub-optimally small formal firms.

### Policy conclusions and recommendations
- Conclusions:
  - Misallocation in India is sizable and heterogeneous across states; differences in labor market reform progress explain part of this heterogeneity.
  - States with more progress on labor market reforms tend to have lower misallocation (TFPR dispersion and skewness); effects particularly marked in states with large informal sectors.
  - Informality negatively affects aggregate TFP, reduces dispersion of TFPR via extensive margin effects, and increases negative skewness of TFPR distribution.
- Policy recommendations:
  - Modernize labor regulations to improve labor market flexibility, increase formal employment, and enhance capital allocation.
  - Reform the Industrial Disputes Act of 1947 and restrictive clauses under the Factories Act of 1948 to allow firms to expand and reach economies of scale.
  - Rebalance away from stringent Employment Protection Legislation toward broad-based unemployment insurance to foster efficiency and equity.
  - Complement labor market flexibility with strengthened safety nets to support inclusive growth.

*Source: wpiea2021051-print-pdf - References ________________________________________________________________23*

### References ________________________________________________________________23

### References

### Key findings
- Resource misallocation in India’s formal manufacturing sector is sizable and shows significant heterogeneity across Indian states.
- Eliminating misallocation to achieve US efficiency could result in manufacturing productivity gains of about 40-60 percent in India.
- Econometric evidence suggests that reforms to increase labor market flexibility are associated with higher TFP and lower misallocation, especially in states where informality is relatively high.
- India’s labor laws have remained among the more restrictive laws in the world and strict labor laws may relate to large informality in the Indian economy.

### Methodology and measures
- The paper uses micro-level balance sheet data to:
  - calculate aggregate total factor productivity (TFP); and
  - calculate the dispersion of firms’ revenue productivity (TFPR) as a standard measure of resource misallocation outlined in Hsieh and Klenow (2009).
- The skewness of the distribution of revenue productivity is calculated to assess the prevalence of firms within a state-sector that are too small or too large relative to a distribution with no misallocation.
- TFPR dispersion is measured as the standard deviation of revenue productivity (in logs) from industry means; the TFP gap is derived by equalizing TFPR within industries.

### Empirical scope and context
- Focus is on labor market rigidities in the formal manufacturing sector across Indian states.
- Labor laws in India are concurrently legislated by both center and state governments, producing complex, numerous, and rigid labor legislations.
- Some state-level reforms and limited federal-level reforms have been implemented, motivating state-level analysis of reform effectiveness on resource misallocation, productivity, and growth.

### Policy implications and interpretation
- Addressing distortionary policies (barriers to entry/exit, limits on firm capacity, tax and subsidy policies, and labor market rigidities) can reduce misallocation and raise aggregate productivity.
- State-level labor market reforms that increase labor market flexibility can improve aggregate TFP and reduce misallocation, with stronger effects in states exhibiting higher informality.

*Source: wpiea2021051-print-pdf - References ________________________________________________________________23*

### Section VII concludes.

### wpiea2021051-print-pdf - Section VII concludes.

### Literature reviews
- Enhanced appreciation of resource misallocation across firms and sectors to explain low aggregate productivity (Jones, 2016).
- Hsieh and Klenow (2009): firm-level price distortions create resource misallocation that explains productivity gaps between the U.S., China, and India.
- Low productivity growth can result from:
  - slow adoption of frontier technologies and best practices; and
  - inefficiencies in allocating productive resources caused in part by institutional features and government policies.
- India-specific findings:
  - State-specific labor and product market regulations, financial development, and infrastructure investment drive heterogeneity in firms’ output performance (Besley and Burgess (2004); Gupta et al (2008); Kapoor (2015)).
  - Firms in labor-intensive industries located in states with flexible labor markets have higher productivity than those in states with more stringent labor laws (Dougherty et al (2011)).
  - State-level pro-employer deregulation is associated with gains in total factor productivity growth (Dougherty et al (2014)).
  - Distortionary policies including firm-size tax distortions and strict labor laws, shortage of capital, and limited access to intermediate inputs contribute to misallocation in Indian manufacturing (Chatterjee (2011)).
- Literature on informality:
  - Chatterjee (2011): misallocation in the formal manufacturing sector tends to be larger compared to the informal sector; results sensitive to methodology.
  - Ulyssea (2018): in a Melitz (2003) type framework for Brazil, lower informality can be, but is not necessarily, associated with higher output, TFP growth, or welfare.
  - Alvarez and Ruane (2019): removing formal sector regulatory labor distortions in Mexico increases formal employment share but has no significant impact on misallocation or aggregate productivity.
  - Misch and Saborowski (2018): higher informality in Mexico associated with higher resource misallocation.
- Paper’s main contribution:
  - Uncover the sector-state pattern of resource misallocation in India.
  - Highlight the role of labor market rigidities given states’ role in setting labor regulations.
  - Propose a new measure of sector-level misallocation based on the Hsieh and Klenow (2009) framework.

### Empirical framework
- Follows Hsieh and Klenow (2009) theoretical approach: continuum of sectors s = 1,...,S_j in each state j, with M_jss firms per sector.
- Firms produce differentiated goods under monopolistic competition with Cobb–Douglas technology:
  - Y_jiss = A_jiss K_jiss^{α_s} L_jiss^{1−α_s}
- Firms maximize profits:
  - π_jiss = max_{K,L} P_jiss Y_jiss (1−τ^Y_jiss) − R_jiss (1+τ^K_jiss) K_jiss − w_jiss (1+τ^L_jiss) L_jiss
  - Distortions τ^K, τ^L, τ^Y can be positive (taxes) or negative (subventions). Positive distortions on capital and labor increase unit cost; positive distortions on final output reduce firm revenue.
- CES demand with elasticity of substitution θ>1 implies price is a fixed markup over marginal cost.

### Measuring misallocation
- Define revenue productivity (TFPR): TFPR_jiss = P_jiss A_jiss
- Expression for TFPR preserves factor prices, factor shares, distortions, and (1−τ^Y):
  - TFPR_jiss = (θ/(1−θ)) (R_jiss^{α_j} w_jiss^{1−α_j})^{α_s,...} · (1+τ^K_jiss)^{α_s} (1+τ^L_jiss)^{1−α_s} (1−τ^Y_jiss)
  - (Equation (1) as in source)
- Sector aggregate:
  - Sector output Y_js is CES aggregate of firm outputs: Y_js = (∑_{i=1}^{M_js} Y_jis^{(θ−1)/θ})^{θ/(θ−1)} (Equation (2))
  - Aggregate TFP TFP_js = (∑_i A_jis · TFPR_jis^{...})^{θ/(θ−1)} (Equation (3); full expression per source)
- Main measures of misallocation:
  - Dispersion of firm TFPR: variance of log(TFPR_jis). Assumes TFPR and true productivity A_jis jointly log-normal (Hsieh and Klenow (2009)); details in Appendix I.
  - Skewness measure: deviation of the median firm’s TFPR from the industry mean:
    - Industry mean TFPR: TFPR̄_js = [expression in source]
    - Skewness: TFPR̄_js,p50 = TFPR_js,p50 / TFPR̄_js
    - If TFPR̄_js,p50 = 1 (or log ratio = 0): median TFPR equals industry mean → marginal products equalized → no misallocation.
    - If "Median TFPR" > 1 (log ratio > 0): distribution has negative skewness; median firm TFPR above optimal level → median firm faces distortions reducing its size relative to optimal (e.g., size-dependent taxes).
    - If "Median TFPR" < 1 (log ratio < 0): distribution has positive skewness; median firm is sub-optimally too large → median firm faces distortions increasing its size relative to optimal (e.g., subsidies).
- Constructed “TFP gap” measures distance between “efficient” and “observed” output (details in Appendix I).

### Misallocation and informality
- Hypothesis: easing labor market regulations will produce different formal sector outcomes depending on informality level.
  - Where informality is high, easing regulations expected to produce more sizable improvements in formal sector outcomes.
- Mechanisms and intuition:
  - Productive formal firms may be unable to reach optimal size where informality is high due to competition from less productive informal firms that evade costly regulations and taxes.
  - Informality may be endogenous to labor market regulations; easing regulations where they bind most yields larger effects.
  - High informality may mark a larger mass of firms that would enter the formal sector if entry costs were lower; lowering entry costs (including easing labor regulations) could let such firms grow and reduce misallocation.
- Empirical approach: look for suggestive evidence via interaction of employment legislation and informality; do not test a single channel.

### Data and stylized facts
- Firm-level data: Annual Survey of Industries (ASI) compiled by CSO.
  - Covers factories under the Factories Act, 1948: firms employing more than 10 workers using power and those employing more than 20 workers without using power.
  - Survey is a census of registered manufacturing units with 100+ employees and a random sample of one-fifth of remaining registered firms.
  - Survey years used: fiscal years 2003/04, 2006/07, 2008/09, and 2010/11.
  - No firm-level panel due to changing random samples over time.
- Variables of interest: capital stock, labor compensation, gross output, intermediate inputs, value-added.
- Industry classification:
  - ASI uses NIC-1998 in 2003/04, NIC-2004 in 2006/07, NIC-2008 in 2008/09 and 2010/11.
  - Converted NIC-1998 and NIC-2004 into NIC-2008; sectors defined at 3-digit level.
- Identification strategy for distortions:
  - Set factor shares in Cobb–Douglas production function equal to corresponding U.S. manufacturing industry (3-digit) as in Hsieh and Klenow (2009) and Chatterjee (2011).
  - Industry capital and labor shares from NBER productivity database (1958–2011); average factor shares computed and labor shares scaled by 3/2 to reflect nonwage compensation.
- Data cleaning:
  - Replace negative values of value-added, capital, and labor compensation with missing.
  - Trim 1 percent tails of firm productivity (log TFPR) by year, then 1 percent tails of firm relative distortions (log(TFPR_jis / TFPR̄_js)) and firm relative productivity (log(A_jis / Ā_js)) by 3-digit industry and year.
- Scope:
  - 20 Indian states, 41 sectors, and four years; states have on average 64 sector-year observations.
  - Analysis focuses on testing whether state-level labor market reforms measured as of 2007 associate with lower misallocation; coefficients estimated from variation across states within sector and year.

### State and sector characteristics in India
- Employment Protection Legislation (EPL) index:
  - Index of changes in EPL as in Dougherty (2008), rescaled from zero to one; computed in 2007 and time-invariant in this analysis.
  - Captures percentage of areas with labor reforms in the 1990s and 2000s, including legal changes to Industrial Dispute Act, Factories Act, Shop Act, Contract Labour Act, and procedural changes (registration, return filing, union representation).
  - Higher EPL associated with more reforms towards labor market flexibility.
  - Examples: Andhra Pradesh and Gujarat have EPL = 0.96; West Bengal and Chhattisgarh have EPL = 0.5.
  - Note: many states implemented labor reforms after 2007 not captured by this index.
- State-level controls and proxies:
  - CMIE state database: logs of credit per capita (measure of financial development / capital availability), road density (kilometer per thousand square kilometers, proxy for infrastructure), unregistered manufacturing NSDP share of state NSDP (proxy for informality).
    - Example: Delhi’s credit per capita about 1.5 times that of Bihar (the lowest).
    - CMIE data for unregistered manufacturing NSDP are missing for some states such as Uttar Pradesh.
  - Product Market Regulation (PMR) index from Conway and Herd (2009), re-scaled to range zero to three; higher PMR means regulatory environment more supportive of competition.
    - Examples: Goa and Haryana have most competitive environment; West Bengal and Gujarat have the least.
  - World Bank Enterprise Surveys (2014): share of firms reporting access to electricity or inadequately educated workforce as an obstacle by state (captures human capital and infrastructure at a point in time).

*Source: wpiea2021051-print-pdf - Section VII concludes.*

### Box 1. Labor Market Regulations and Reforms in India

### Box 1. Labor Market Regulations and Reforms in India

### Overview of regulation and recent legislative changes
- India’s labor market regulations are described as relatively strict, numerous, and outdated, including at the sub-national level.
- Chapter V-B of the Industrial Disputes Act (IDA) requires government approval for layoffs, retrenchments, and closures where this law applies on all factories with 100 or more workers.
- Prior to 2017, labor laws at both the center and states in India numbered to around 250 laws governing minimum wages, industrial dispute resolution, hiring and firing, and closure conditions.
- Since 2017 the central government has attempted to amalgamate 44 central labor laws into four codes on (i) industrial relations, (ii) wages, (iii) social security and welfare, and (iv) safety and working conditions.
  - The Code on Wages was passed by the Parliament in August 2019; the remaining three codes were passed in September 2020.
  - Some draft-code elements would create more flexibility; others could backtrack (example given: increasing firing costs from ½ to 1½ months of salary). Chapter V-B of the IDA has not been altered in these reforms.
  - The new social security code could constitute a major reform but depends on the design of new social security schemes by the central government.

### State-level reforms and COVID-19 emergency relaxations
- State reforms prior to 2017:
  - Rajasthan, Madhya Pradesh, and Haryana in 2014 modified their Industrial Disputes Act to allow automatic retrenchment for a factory with less than 300 workers.
  - Gujarat allowed automatic retrenchment of workers in any factory in the Specialized Economic Zones, Special Investment Regions, and National Investment and Manufacturing Zones, provided employers give severance payment of 60 days of wages for every year of employment.
  - Maharashtra in 2017 allowed automatic retrenchment for up to 300 workers.
- COVID-19 period temporary relaxations to stimulate employment and attract investment:
  - Madhya Pradesh provided exemption for new investors from existing labor laws for 1,000 days from the announcement in May 2020.
  - Gujarat granted labor-law exemptions for 1,200 days from the start of production of any new venture in the state.
  - Assam, Haryana, Odisha, Punjab, Rajasthan and Uttarakhand announced emergency ordinances providing similar incentives.

### Stylized facts: misallocation, firm size, and informality
- Cross-country and within-India magnitudes:
  - Resource misallocation in India appears large compared to the United States: the variance of TFPR and the TFP gap are respectively 1.5 and 3 times higher in India in 2003 than in the U.S. in 1997.
- Heterogeneity across states:
  - Aggregate manufacturing TFP (in logs) in a state ranges from 3.2 in Bihar to 5.1 in Maharashtra.
  - The average variance of revenue productivity across states varies in a narrow range; sector-state level variation ranges between 0.9 and 2.4.
  - Twelve out of 20 states in the sample have median TFPR below state-industry mean TFPR, indicating prevalence of larger-than-optimal firms; five states have a ratio exceeding 1 (presence of smaller-than-optimal firms).
- Relationships with labor regulation and informality:
  - Lower levels of labor market reforms (lower Employment Protection Legislation, EPL) are suggestively correlated with the prevalence of smaller-than-optimal firms; as labor market regulations become less restrictive, the median-mean TFPR ratio declines below 1 (Figure 5 in source).
  - Examples of EPL indices cited:
    - Bihar, Jharkhand, West Bengal and Kerala: EPL = 0.5 (states with median TFPR above 1 and suggestively lower EPL).
    - Gujarat and Andhra Pradesh: EPL = 0.96 (states with median-mean ratio around 1).
    - Karnataka and Orissa comparison: both EPL = 0.79, but Karnataka has median ≈ mean TFPR and relatively low informality; Orissa has median > mean and relatively high informality.
  - The median-mean ratio of TFPR and the share of unregistered net state domestic product (NSDP) in manufacturing are positively correlated: high informality appears associated with too many sub-optimally small firms (negative skew of TFPR).
  - Informality can operate via:
    - Intensive margin: outsourcing labor-intensive activities to the informal sector increases capital-to-labor ratio and reduces TFPR (raising misallocation).
    - Extensive margin: small and unproductive firms avoiding the formal sector can decrease measured misallocation in formal sector.
  - States with high informality: defined as share of unregistered NSDP in manufacturing in highest quartile (share higher than 60 percent). Examples across years provided: Jammu and Kashmir (2003), Chandigarh (2003), Delhi (2006, 2008, 2010), Bihar (2003, 2006, 2008, 2010), Nagaland (2003, 2006, 2008), Manipur (2003, 2006, 2008, 2010), Tripura (2006, 2008).

### Econometric framework and baseline expectations
- Main empirical specification (baseline OLS) models sector-state-year TFPR or TFP outcomes against:
  - EPL index (time invariant by construction),
  - Informality (ratio of unregistered NSDP in manufacturing over total NSDP by state and year),
  - Controls Z (state characteristics: total number of firms in logs, road density in logs, percentage of firms for which electricity is an obstacle),
  - Industry-year pair fixed effects and state-clustered standard errors.
- Main dependent variables: aggregate TFP, variance of TFPR (misallocation measure), and median-mean ratio of TFPR (skewness indicator).
- Expectations:
  - Rigid labor laws disincentivize firm expansion; states with more flexible labor markets expected to have (i) higher TFP, (ii) lower dispersion of TFPR, and (iii) less negative skew (lower median TFPR) — implying β1 expected negative.
- Sample construction note: exclude state-sector-year observations aggregated from fewer than 10 firms; this exclusion keeps 46 percent of the sample.

### Key econometric results
- Baseline regression (Table 5 summary):
  - Implementing labor market reforms (higher EPL) significantly shrinks negative distortions on firm-level prices as measured by the skewness (median-mean ratio) of TFPR (column 3).
  - Labor market reforms would likely reduce dispersion of TFPR and increase productivity, but these effects are not statistically significant in the dispersion regression (column 2).
  - Informality is associated with lower aggregate TFP in the formal sector (column 1) and associated with lower dispersion of firm-level TFPR (column 2).
  - Better infrastructure (higher road density, lower electricity constraints) is significantly correlated with greater aggregate TFP.
- Interaction of EPL and informality (Table 7 summary):
  - Easing EPL in states with high informality has a large, significant effect on TFP levels (column 1).
  - Dispersion of TFPR increases (column 2), likely due to extensive margin effects, but overall net efficiency gains occur.
  - Median-mean ratio of TFPR falls (column 3) — reducing negative skew and/or increasing positive skew — implying negative price distortions shrink in high-informality states when labor markets become more flexible.
  - Infrastructure controls remain robust in their positive association with TFP.
- Olley and Pakes decomposition (Table 8 summary):
  - Aggregate labor productivity decomposed into the simple average of firm-level productivity and the covariance between firm employment and labor productivity.
  - The interaction between EPL and informality has a strong positive relationship with the covariance term: easing labor market regulations is associated with improved allocative efficiency, increased average labor productivity, and positive contribution to aggregate labor productivity.
- Robustness and timing:
  - Using reform indexes computed in 2006/2007 and restricting firm-level data to 2008 and 2010 yields qualitatively similar results: interaction between EPL and unregistered NSDP share remains positive for TFP and negative for median-mean TFPR (Table 9).
- Scenario/calibration of potential TFP gains:
  - Calibrating the impact of advancing labor market reforms of an Indian state to the best performer (index = 1) using estimated coefficients from Table 7 suggests significant potential aggregate TFP increases.
  - West Bengal, Maharashtra and Jharkhand are identified as likely to benefit most in absolute TFP gains in this scenario (Figure 8).

### Policy conclusions and recommendations
- Main conclusions:
  - Misallocation in India is sizable and heterogeneous across states; part of this heterogeneity is explained by differences in progress on labor market reforms.
  - States that made more progress on labor market reforms tend to have lower degrees of misallocation (TFPR dispersion and skewness); effects are particularly marked in states with large informal sectors.
  - Informality has a complex relationship with productivity: it negatively affects TFP, reduces dispersion of TFPR (extensive margin), and increases negative skewness of TFPR distribution (more sub-optimally small firms in the formal sector).
- Policy recommendations:
  - Modernize labor regulations to improve labor market flexibility, increase formal employment, and enhance capital allocation.
  - Reforms to the Industrial Disputes Act of 1947 and restrictive clauses under the Factories Act of 1948 are identified as key to enhance labor market flexibility and allow firms to expand and reach economies of scale.
  - Rebalance away from stringent Employment Protection Legislation toward broad-based unemployment insurance to foster efficiency (higher and more stable employment and productivity) and equity (adequate protection to workers).
  - Complement labor market flexibility with strengthened safety nets to support inclusive growth.

*Source: Box 1. Labor Market Regulations and Reforms in India (excerpted from the provided IMF working paper content).*

### References

### References (wpiea2021051-print-pdf - References)

### Key literature cited
- Alvarez, J. and C. Ruane (2019), “Informality and Aggregate Productivity: The Case of Mexico,” IMF working paper, WP/19/257.
- Besley, T. and R. Burgess (2004), “Can Labor Regulation Hinder Economic Performance? Evidence from India,” The Quarterly Journal of Economics, 119(1):91–134.
- Blis M., P. Klenow and C. Ruane (2020), “Misallocation or Mismeasurement?” NBER Working Paper, No. 26711.
- Chatterjee, U. (2011), “Resource Misallocation and Efficiency in Developing Countries,” Ph.D. Dissertation, the University of California at Berkeley.
- Conway, P. and R. Herd (2009), “How competitive is product market regulation in India? An international and cross-state comparison,” OECD Journal: Economic Studies, 1–25.
- Dougherty, S. (2008), “Labour Regulation and Employment Dynamics at the State Level in India,” OECD Economics Department Working Paper, No.264, Paris.
- Dougherty, S., V. Frisancho and K. Krishna (2011), “Employment Protection Legislation and Plant-Level Productivity in India,” NBER Working Paper, No.17693.
- Dougherty, S., V. Frisancho and K. Krishna (2014), “State-level Labor Reform and Firm-level Productivity in India,” India Policy Forum, 10 (1) 1–56.
- Duranton, G., E. Ghani, A. G. Goswami and W. Kerr (2015), “The Misallocation of Land and Other Factors of Production in India,” World Bank Working Paper, No. 7221, Washington D.C.
- Duval, R.A. and P. Loungani (2019), “Designing Labor Market Institutions in Emerging and Developing Economies: Evidence and Policy Options,” IMF Working Paper, SDN/19/04.
- Gopinath, G., Ş. Kalemli-Özcan, L. Karabarbounis and C. Villegas-Sanchez (2017) "Capital Allocation and Productivity in South Europe," The Quarterly Journal of Economics, 132 (4) 1915-1967.
- Government of India (2019), “Nourishing Dwarfs to become Giants: Reorienting policies for MSME Growth,” The Economic Survey 2018-19, Department of Economic Affairs, Ministry of Finance, New Delhi.
- Gupta, P., R. Hasan and U. Kumar (2008), “Big Reforms but Small Payoffs: Explaining the Weak Record of Growth in Indian Manufacturing," India Policy Forum, Global Economy and Development Program, The Brookings Institution, 5(1): 59-123.
- Jones, C. I. (2016), “The Facts of Economic Growth,” Handbook of Macroeconomics, 2:3-69.
- Haltiwanger, J., R. Kulick and C. Syverson (2018), “Misallocation Measures: The Distortion That Ate the Residual,” NBER Working Papers, No. 24199.
- Hsieh C. T. and P. J. Klenow (2009), “Misallocation and Manufacturing TFP in China and India,” The Quarterly Journal of Economics, 1403–1448.
- Kapoor, R. (2015), “Creating Jobs in India’s Organised Manufacturing Sector,” The Indian Journal of Labour Economics, 349–375.
- Leon-Ledesma, M. A. and D. Christopoulos (2016), “Misallocation, Access to Finance, and Public Credit: Firm-Level Evidence,” Asian Development Review, 33(2): 119-143.
- Melitz, M. (2003), “The Impact of Trade on Intra-Industry Reallocations and Aggregate Industry Productivity,” Econometrica,71(6): 1695–1725.
- Melitz, M. and S, J. Redding (2014), “Missing Gains from Trade?” American Economic Review, 104:5 317–21.
- Midrigan, V. and D. Y. Xu (2014), "Finance and Misallocation: Evidence from Plant-Level Data," American Economic Review, 104 (2): 422-58.
- Misch, F. and C. Saborowski (2018), “Resource Misallocation and Productivity: Evidence from Mexico,” IMF Working Paper WP, No. 18/112, Washington.
- Nishida, M., A. Petrin, M. Rotemberg and T. White (2016), “Measuring Cross-Country Differences in Misallocation,” US Census Bureau Center for Economic Studies Paper, No. CES-WP-16-50.
- Olley, G., and A. Pakes (1996), “The Dynamics of Productivity in the Telecommunications Equipment Industry,” Econometrica, 64(6), 1263-1297.
- Restuccia, D. and R. Rogerson (2017), “The causes and costs of misallocation,” Journal of Economic Perspectives, 31(3): 151–174.
- Rotemberg, M. and T. White (2017), “Measuring Cross-Country Differences in Misallocation,” mimeo.
- Ulyssea, G. (2018), “Firms, Informality and Development: Theory and evidence from Brazil,” American Economic Review, 108 (8): 2015-47.

### Appendix I — Alternative Measures of Misallocation
- Variance of TFPR
  - Aggregate industry TFP in state j and sector s is expressed using the CES aggregator and TFPR as in equation (1) and (2).
  - Under joint lognormality of TFPR_R and A, aggregate TFP decomposition (Hsieh and Klenow (2009)) is:
    - log TFP_j s = 1/(σ−1) log(...) − σ^2/2 Var(log TFPR_R_j s)
  - The first term captures productivity gains due to technical efficiency.
  - The second term captures the negative effect of firm-level distortions on aggregate TFP; the variance of TFPR is the main measure of misallocation in the paper.
- TFP gap
  - The TFP gap is the distance between “efficient” and “observed” output; sector-level ratio:
    - TFP_j s / TFP_j s* = (...)^(1/(σ−1))
  - TFP gap defined as TFP_gap_j s = TFP_j s* / TFP_j s − 1
  - For state-level TFP gap the paper uses a Cobb-Douglas aggregator:
    - TFP_gap_j = (Π TFP_gap_j s ^ρ_j s)
  - For India-wide TFP gap the entire India is treated as one state.

### Data summary and descriptive statistics (selected exact figures)
- Table 1: Summary Statistics by State—ASI database (selected rows)
  - MAHARASHTRA: # Sector-Year Obs. 197, Avg # Firms per sector and year 72, Agg. TFP 5.12, Avg. Variance TFPR 0.71, Avg. Median TFPR 1.00
  - TAMIL NADU: 189, 87, 4.77, 0.67, 0.95
  - GUJARAT: 174, 67, 4.84, 0.69, 0.94
  - Average (across listed states): # Sector-Year Obs. 108, Avg # Firms per sector and year 39, Agg. TFP 4.42, Avg. Variance TFPR 0.70, Avg. Median TFPR 0.98
- Table 2: Summary Statistics—State Characteristics (selected rows; exact values)
  - GUJARAT: EPL 0.96, Unreg. NSDP 0.26, NSDP 9.71, Credit per capita 0.68, PMR -0.73, Road density 0.04, Education obstacle 0.00
  - MAHARASHTRA: EPL 0.75, Unreg. NSDP 0.33, NSDP 10.69, Credit per capita 1.71, PMR 2.70, Road density 0.07, Education obstacle 0.03
  - Average (across listed states): EPL 0.74, Unreg. NSDP 0.37, NSDP 9.42, Credit per capita 1.30, PMR 0.84, Road density 0.10, Education obstacle 0.24
- Table 3: Dispersion of Firm-Level TFPR (in logs) in India by year
  - 2003: Median -0.20, Variance 0.79, P75/P25 0.93, P90/P10 1.86, # Firms 50,600, Share of firms with non-missing TFPR 0.68
  - 2006: Median -0.32, Variance 0.79, P75/P25 0.96, P90/P10 1.86, # Firms 59,819, Share 0.56
  - 2008: Median -0.16, Variance 0.79, P75/P25 0.95, P90/P10 1.83, # Firms 50,261, Share 0.59
  - 2010: Median -0.11, Variance 0.70, P75/P25 0.89, P90/P10 1.71, # Firms 48,120, Share 0.72

### Empirical results — main regression findings (exact coefficients and significance)
- Table 5: Aggregate TFP, Misallocation and Labor Market Reforms – Baseline
  - Outcome variables: (1) Aggregate TFP; (2) Variance TFPR; (3) Median TFPR
  - Employ. Protec. Legislation: 0.039 (column 1), -0.164 (column 2), -0.275** (column 3)
  - Share of Unregistered NSDP: -1.470***, -0.207***, 0.129
  - Access to Electricity as an obstacle: -0.555***, -0.045, -0.086*
  - Road density: 0.073***, 0.012, -0.009
  - # Firms: 0.510***, -0.018*, 0.021
  - Observations 2,017; R-squared 0.467 (cols 1 and 2), 0.191 (col 3)
- Table 6: Baseline with Additional Control Variables
  - Employ. Protec. Legislation: -0.120, -0.147, -0.264**
  - Share of Unregistered NSDP: -1.467***, -0.224***, 0.118
  - Product Market Regulation: -0.005, -0.045, -0.060*
  - Inadequately educated workforce as an obstacle: -0.608*, 0.023, -0.050
  - Road density: 0.090***, 0.012, -0.007
  - # Firms: 0.480***, -0.017, 0.025*
  - Observations 2,017; R-squared 0.470 (cols 1 and 2), 0.196 (col 3)
- Table 7: Aggregate TFP, Misallocation, Labor Market Reforms and Informality
  - Employ. Protec. Legislation: -0.516, -0.274***, -0.058
  - EPL*Unreg. NSDP (Median): 1.413*, 0.248*, -0.376*
  - Unregistered NSDP (Median): -1.413**, -0.230*, 0.332**
  - Road density: 0.111***, 0.017**, -0.013
  - # Firms: 0.550***, -0.012, 0.013
  - Observations 2,017; R-squared 0.446, 0.463, 0.194
- Table 8: Aggregate Labor Productivity and the Covariance Between Labor and Labor Productivity
  - Employ. Protec. Legislation: -0.579*, -0.587***, 0.008
  - EPL*Unreg. NSDP (Median): 1.584***, 1.071**, 0.513**
  - Unregistered NSDP (Median): -1.431***, -1.001***, -0.430**
  - # Firms: -0.024, -0.060*, 0.036*
  - Observations 2,017; R-squared 0.260, 0.380, 0.175
- Table 9: Restricting Sample to Years 2008 and 2010
  - Employ. Protec. Legislation: -0.627*, -0.378***, -0.167
  - EPL*Unreg. NSDP (Median): 1.241**, 0.226, -0.189
  - Unregistered NSDP (Median): -1.328**, -0.219*, 0.171
  - Road density: 0.101***, 0.008, -0.019***
  - # Firms: 0.520***, -0.005, 0.030*
  - Observations 944; R-squared 0.452, 0.522, 0.167

*Source: Authors’ calculations (wpiea2021051-print-pdf - References).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021051-print-pdf.pdf_