Labor Regulation, Enforcement, and Employment: Lessons from China. Albert Park Hong Kong University of Science and Technology

Transcription

1 Labor Regulation, Enforcement, and Employment: Lessons from China Albert Park Hong Kong University of Science and Technology

2 Motivations Debates over optimal labor regulation Concerns about enforcement in emerging markets/developing countries Main points: Labor regulations do impact employment in China (despite no aggregate unemployment problem) differential regulation or differential enforcement of regulations create costly distortions

3 Labor Regulation in China Employment of migrant workers relatively unregulated for most of the reform period 2008 Labor Contract Law significantly increased worker protections for all workers, and increased penalties for noncompliance Impacts of this reform still unclear 2004 Minimum Wage Law required biannual adjustments of minimum wage for all workers

4 Internationally, China s Labor Regulations are Highly Protective

5 Outline Some background on China s Labor Contract Law Evidence from 2009 national firm survey Evidence from 2015 employer-employee survey in Guangdong Implications for India

6 China s Labor Contract Law Effective: January 1, 2008 Labor Contracts After 2 fixed-term contracts, or 10 years of employment, contract must be open-ended Limits on probationary period (1-3 months depending on contract length) Regulations on temporary work agencies (labor service companies) Severance conditions 30-day written notice Severance pay: one month s pay for each year of service (half month s pay if less than 6 months), double severance pay for unfair dismissal Enhanced penalties for failing to provide labor contracts, provide social insurance coverage

7 China Urban Labor Survey (CULS3), February 2010 In each of 6 cities, survey 700 local resident households and 600 migrant households Surveyed >15,000 adults, including 6000 migrants 3-stage PPS sampling of urban sub-districts, neighborhoods, and households Detailed enumeration of all dwellings in each neighborhood Surveys directed by CASS, working closely with city Statistical Bureaus

8 Informality by residence status (CULS, 2001, 2005, and 2010)

9 China National Firm Survey, November 2009 Surveyed firms in 8 provinces: 4 coastal provinces (Shandong, Jiangsu, Zhejiang, and Guangdong), one northeast province (Jilin), one central province (Hubei), one northwest province (Shaanxi), and one southwest province (Sichuan). Representative sample of >2000 manufacturing firms in 25 municipalities Focus on employment at 4 points in time: end-2007, mid- 2008, end-2008, mid-2009 Sampling frame: all firms who ever had credit relationship with any financial institution Key collaborators: People s Bank of China Research Department, Peking University, CASS

10 How strictly have labor regulations been enforced? (China firm survey 2009) Very strict Strict Not strict By period: Jan-Jun Jul-Dec Jan-Jun By size: Smallest nd quartile rd quartile Largest Firms report strict enforcement, with no weakening during the crisis. Smaller firms report less strict enforcement than larger firms.

11 Labor regulations and firm employment decisions Have labor regulations made it more difficult for Has new your firm to hire Law reduce and fire workers? hiring? (% (% yes) yes) Has new Law reduce firing? (% yes) Total

12 Did new Labor Law impact labor costs? Province significant increase some increase no increase Zhejiang Jiangsu Guangdong Shandong Jilin Hubei Shaanxi Sichuan Total

13 Empirical Research Questions What are determinants of Labor Law enforcement? How did city variation in enforcement affect the impacts of the Labor Law on employment changes?

14 Empirical specification: determinants of labor regulation enforcement strictness Firm static model: L ijc,t = α 1 X c, α 2 X ijc + α 3 Exporter ijc,2007 *S jc,t + λ t + ε ijc,t Dynamic model: L ijc,t = δ 1 L c, δ 2 X c, δ 3 X ijc + δ 4 Exporter ijc,2007 *S jc,t + λ t + u ijc,t Subscripts i=firm, j=sector, c=city, t=year X ijc includes firm time-invariant characteristics (including 2007 characteristics, e.g., size, exporter) X c,2007 are city characteristics before Labor Law implemented λ t are time period fixed effects Notes: interacted terms also included independently as regressors, constant term omitted

15 Determinants of enforcement strictness STATIC MODEL: Findings: enforcement stricter for: - state sector (not private/foreign) - Sichuan, Jiangsu, Shaanxi, Jilin (no strong pattern) - larger firms - most recent period DYNAMIC MODEL: Enforcement increased more for cities with more firms that: - had less strict initial enforcement - food, not raw mat. or capital - foreign firms - interior provinces - larger - most recent period 2007 enforcement Enforcement City enforcement 2007 city enforcement (0.040)** Sector-consumer products share (0.03) (0.00) Sector-raw materials share (0.03) (0.002)* Sector-capital & equipment share (0.03) (0.004)** Sector-other share (0.039)* (0.00) Ownership-private share (0.045)* (0.00) Ownership-Joint/Ltd/Other share (0.04) (0.00) Ownership-foreign share (0.049)** (0.005)* Province-Jiangsu (0.028)** (0.03) Guangdong (0.04) (0.04) Shandong (0.04) (0.04) Jilin (0.063)* (0.04) Hubei (0.07) (0.03) Shaanxi (0.046)** (0.025)** Sichuan (0.046)** (0.030)** Size quartile 2 share (0.023)** (0.00) Size quartile 3 share (0.024)** (0.00) Size quartile largest share (0.026)** (0.00) Firm age (0.00) 0.00 end (0.02) (0.007)* mid (0.02) (0.007)**

16 Theory: Labor Law and Employment Changes Consider employment (E t ) to be a simple ratio of a firm s optimal employment given perfect enforcement of the Labor Law (L t ) and the actual strictness of Labor Law enforcement (s t ): E t = L t /s t Here, 0<s t 1, so that lower s t leads to higher E t Then ΔlnE = lne 2 - lne 1 = ΔlnL Δlns New Labor Law reduces optimal employment given perfect enforcement (L 2 <L 1 ), this effect is constant across cities

17 Theory: Labor Law and Employment Changes Consider two possible impacts of Law on strictness of enforcement: Brings strictness of all cities closer to being perfect (s 2 =1), then ΔlnE = ΔlnL + lns 1 > initial strictness is associated with increases in employment No change in strictness, then ΔlnE = ΔlnL > strictness has no effect on employment change, but this is sensitive to assumptions about the complementarity between strictness and substance of the law

18 How has enforcement of labor regulations affected firm employment decisions? ΔE ijc,t = β 1 L c, β 2 X c, β 3 X ijc + β 4 S jc,t + β 5 Exporter ijc,2007 S jc,t L c, λ t + u ijc,t Key identification issue: endogeneity of enforcement Focus on pre-labor Law enforcement (predetermined) City-level measures of enforcement, less endogenous than firm-level measures but could be correlated with other city characteristics associated with employment growth Lc,2007 is city-level measurement, so should cluster standard errors at city level But small number of clusters (25), so use bootstrapped standard errors (Cameron, Gelbach, Miller, 2008)

19 Determinants of change in ln(production workers): Results using different enforcement measures With city controls 2007 city enforcement 0.073*** Ln(mean training cost per firm) 0.030*** Mean training days *** Notes: ***significant at 1% level, * significant at 10% level. Results are from separate regressions using each enforcement measure and same specification as in previous tables. Also add city control variables: Ln(population in 2007), ln(gdp per capita in 2007), secondary sector share in 2007, GDP growth rate in 2007, ln(wage in 2007), ln(budget exp p.c.)

20 City-level Regressions City enforcement Δlog(production workers) biannual city enforcement (0.046)** (0.025)** (0.104)* Observations R-squared Robust standard errors in parentheses * significant at 5%; ** significant at 1% One unit stricter enforcement in 2007 associated with a smaller increase in enforcement strictness One standard deviation difference in 2007 city enforcement leads to 3.2% difference in employment growth over 18 month period Positive employment effects robust to controlling for different 2007 city initial conditions

21 Market Reaction to Labor Law Labor law increased value of laborintensive firms with better pre-laws compliance (size, audit firms) Evidence of productivity gains from reallocation. Source: Ling Ji and Shang-jin Wei (2013)

22 China s Minimum Wage Laws 2004 reform required biannual adjustment, expanded coverage to all workers, substantially increase penalties for noncompliance Minimum wages have reduced employment of low-skilled workers, especially younger, female, and less educated workers (Fang and Lin, 2014)

23 2015 China Employer-Employee Survey (CEES) (partially supported by WB Network on Jobs and Development grant) Sample: 550 Manufacturing Firms in Guangdong 5000 Workers (10 per firm) Organizers: HKUST, Tsinghua, CASS, Wuhan U. Motivation: How are firms responding to rapidly rising labor costs?

24 Firm Size Distribution

25 Barriers to Firm Development To what extent have following factors been a barrier to your firm s development? (prohibitive or very significant factor, top responses) Labor cost Market demand Taxation Technical talent Worker skills Innovation Managerial talent

26 Manufacturing Wages (Annual) Nominal Real

27 Employee Social Insurance Coverage by City

28 Employee Social Insurance Coverage by Ownership

29 Employee Social Insurance Coverage by Sector

30 Determinants of pension coverage Young workers 13% less likely Local residents 10% more likely College/high school graduates 33%/21% more likely Workers with longer tenure and higher wages more likely Private firms least likely VARIABLES (4) Employer provides pension coverage Age30_below *** (0.039) Age30_ (0.036) male (0.028) local 0.103*** (0.040) middle 0.101* (0.053) high 0.214*** (0.050) college 0.325*** (0.034) tenure 0.024*** (0.004) lwage 0.211*** (0.047) private *** (0.020) foreign *** (0.018) lemployment *** Sector fixed effect City fixed effect (0.007) Observations 2,414 Y Y

31 Desired flexibility or an uneven playing field? Local variation in social insurance requirements (e.g., types, reference wage) Voluntary social insurance coverage? The value of benefits (portability, confidence) The costs of opting out (paternalism, financial sustainability of systems) Confusion from multiple systems (including social pensions) Selective enforcement? Rationale for temporary exemptions Costs of enforcement and the burden of regulation

32 Implications for India Differential labor regulations a major issue for India For example, over 1000 different minimum wages, often varying by industry, education, skill, etc. High cost of expanding firm size Differential enforcement a major issue in India and nearly all developing countries, including China Costly regulations harder to enforce, more distorting Consider impacts of enforcement as well as law provisions when designing policy Enforcement innovations?

33 Provides thought leadership on business and policy challenges in emerging economies 40+ Faculty Associates Founded in 2013 with support from EY POLICY BRIEFS WORKING PAPERS iems.ust.hk SEMINARS AND CONFERENCES