Can Labour Regulation Hinder Economic Performance? Evidence from India Timothy Besley and Robin Burgess PRESENTED BY: TS’OLO S. LEHATA & EDMUND OFEI

Outline  Introduction  Methodology and Results  Robustness Tests  Extensions

 Conclusions  Remarks

Introduction  Aim: To study the role of labour market regulation in

explaining manufacturing performance in Indian states between 1958 and 1992  Contribution: Sheds light on the long standing

debate on the role of the state in promoting or hindering economic development  It adds to a growing body of sub-national evidence of

labour regulation effects on economic performance.

Manufacturing in India  Consists of 2 sub-sectors: Registered (9% of output)

& unregistered (5%) of output  Saw an overall growth of 3% from 1960 to 1992  Differences across states: West Bengal moved from

1st to 7th while Andhra Pradesh grew by 7%

Labour Regulation in India  India has a central industrial regulation  Industries (Regulation and Development) Act of 1951  No formal amendments to this act at the state level

 Common set of industrial policies except in the area of industrial relations (Industrial Disputes Act of 1947)

for registered firms

Data  Data on labour regulations comes from 113

amendments of Industrial Disputes Act of 1947

 pro-worker 1, pro-employer - 1 or neutral 0  This Act applies to formal manufacturing sector,

included in Annual Survey of Industries, ASI

 ASI provides data on output, employment, wages,

investment and productivity at state level

Amendments Coding  A sample pro-employer reform from Andhra Pradesh in 1987

 “If in the opinion of the state government it is necessary or

expedient for securing the public safety or the maintenance of public order or services or supplies essential to the life of the community or for maintaining employment or industrial peace in the industrial establishment it may issue an order which (i) requires employers and workers to observe the terms and conditions of an order and (ii) prohibits strikes and lockouts in connection with any industrial dispute.” GETS CODED MINUS ONE IN THE DATA

 A sample pro-worker reform from West Bengal in 1980

 “The limit of 45 days for workers receiving 50% of their

wages upon being laid off (if they worked for more than a year) is removed.” GETS CODED ONE IN THE DATA

Methodology  classifies states as either “treatment” or “control”

states.  Control States do not experience any amendment activity in a pro-worker or pro-employer direction over the 1958-1992 period  They develop an econometric analysis to see whether regulation explains manufacturing patterns over time  Uses both time and cross section variation of labour regulation (common regulation provides best testing ground)

Labour Regulation as a Proxy for IR

Theoretical Considerations  We can reasonably suppose that all firms operate in

a common set of factor markets whose prices they treat as parametric.

 Assume firms produce a common manufactured

good

 Then labour regulation can affect productivity via:

-Relative Price Effect and -Expropriation Effect

Basic Estimation Yst = αs + βt + μrst−1 + ξXst + εst Where: Yst= logged outcome variable in state s at time t. rst= A Regulatory Measure ( Lagged one period to capture the GAP between enactment and implementation.) Xst = Other Exogenous Variables. αs = State Fixed Effects ( To capture state specific factors) βt = Year fixed Effects

Basic Estimation  Panel Data Regression  Pooled OLS

 Impact of Labour regulation on measures of output is

investigated.

 Measures of output: State output, Agricultural , Non-

Agricultural, Construction, Total manufacturing , Registered manufacturing, Unregistered manufacturing outputs per capita.

Results: Basic Estimation

 Effect on output is higher for unregistered

manufacturing firms than for registered manufacturing firms.  This suggests that labour regulation deter formal

registration and encourages firms to remain in the informal sector.

Robustness Test

 Economic Control Variables, Political Complexions,

State time trends, and regressions without West Bengal are introduced.  Other performance measures different from output

are used to ascertain the validity and reliability of the output results.

Robustness Test  Other measures of Performance:  Employment of Registered Manufacturing  Daily employment in registered manufacturing  Earnings per worker in registered manufacturing  Fixed capital formation  Number of factories per capita  Value added per employee

Robustness Test: Conclusion

 Labour Regulation is correlated with Poor Economic

Performance in the registered manufacturing sector

 Labour regulation leads to the rise of an informal

sector.

Robustness Test: Implications  Two states at the extreme ends:  Andhra Pradesh (A Pro-Employer State) would have

registered manufacturing output, 72% of what she had in 1990 and 73% of her employment rate in 1990.

 West Bengal (A Pro- Worker State) would have had

manufacturing output 24% higher than her 1990 levels and employment, 23% higher had she not passed pro-worker amendments.

Potential Endogeneity  States with larger vested interests in manufacturing

may have experienced more pressure to pass proworker amendments hence slower growth  political insiders can see development as a threat to

their rents and hence lobby for protection- Krusell and Rios-Rull (1996)  Reverse causality possibility

Vested interests measure  The average level of union membership (union

members divided by population) before 1977

 Labour regulation is then regressed on the

difference between registered manufacturing in a ‘treatment’ state and that in its matched ‘control’ state while also including match dummies in the regression.

 The results confirm baseline results (TBS), against

initial condition effects concerns

An extension: IV Approach  Most of the labour regulation changes take place

after 1977 Following Prime Minister Indira Gandhi’s declaration of a state of emergency (suspending democratic institutions)  The direction that post-1977 amendments took would depend, in part, on the importance of the initial vested interests (as proxied by union membership)  Interact Union Membership with a Dummy=1 for post 1977

IVs  Historical patterns of land tenure as an IV  There is a possibility that it is correlated with

contemporary patterns of political development

 Areas dominated by non-landlord based revenue

collection have larger concentrations of regional parties today...benefited from SOE

 Interact such districts with a Dummy variable which

equals one after 1977 to mark the persistent shift in political control after the state of emergency

Disaggregated Evidence  Indian states have quite different manufacturing

bases  Hence, there might be a suspicion that patterns of

specialization affect the direction of regulation in ways that could bias the results  E.g early industrializing states might specialize in

slow growing labour intensive industries

 In response, the results on disaggregated data

which look at the impact of labour regulation at the 3-digit industry level for the period 1980-97 are presented  Investigate the link between performance and labour regulation

Welfare Consequences

The Effects of the Labour Regulations on Poverty in India. Rationale:  It may give a sense of where the burden of the previously identified effects have been felt.

Welfare Consequences  Poverty Data from Ozler, Datt, and Ravallion (1996)

is used  Overall Poverty headcount in India is decomposed

into two : Urban Poverty and Rural Poverty  The direct effect of Poverty is expected to depend on

the extent to which the earnings of the poor are derived from Registered Manufacturing

Welfare Consequences

 State Output is also disaggregated into Agricultural,

Registered Manufacturing, and “Other” ( NonAgricultural and Non-Manufacturing) outputs  The correlation between Poverty rates and the

different components of state output in India is looked at

Welfare Consequences  For Urban Poverty: There is a look at the effect of

regulation on Registered Manufacturing, and “Other” outputs on poverty since Agricultural and Unregistered Manufactured outputs are not significantly correlated with urban poverty.  For Rural Poverty: A look at Unregistered output and

not Registered manufacturing output since it has no significant correlation with poverty.

Welfare Consequences: Results

 It is apparent that the coefficient on labour

regulation in Urban poverty headcount regressions remain high across all specifications but for that which includes State time trends.

Welfare Consequences: Implication of results  Andhra Pradesh: Urban Poverty of 112% of her 1990

levels. ( i.e. 640,000 more Urban Poor People in 1990 without the reform.)  West Bengal: Urban Poverty of 11% lower in 1990. (i.e. 520,000 less urban poor people without the reform)  Therefore No Evidence that Pro-worker Labour market policies redress the unfavorable balance of power between Capital and Labour and thus a progressive effect on income distribution.

Conclusion: Recap of Findings  Pro-worker regulation is associated with lower levels

of Investment, Employment, Productivity, and Output in registered manufacturing.

 Pro-worker regulation increased informal sector

activity.

 Pro-worker regulation is associated with rising urban

poverty.

Conclusion  As such, there is little evidence that Pro-worker

regulations have Promoted the interest of Labour. INSTEAD, it has been a constraint on growth and poverty alleviation.  Intention behind the Labour reform in India was

wrong!!

Remarks  India Manufacturing Today: 16% of GDP

 Improved by 3 places to 6th in largest manufacturing

countries  “Make in India” Drive by President Modi (target 25% of GDP by 2025 and 90mil domestic jobs)  (US$ 225.32 billion in investment commitments at

MII initiative in Feb 2016  In Sept 2016 FDI in electronic manufacturing has reached an all-time high of US$ 18.36 billion

The End. THANK YOU