Rana Hasan Neemrana

Do Constraints to Growth Vary by Firm Size? Evidence from India Over a Period of Trade Liberalization* Rana Hasan Asian...

0 downloads 82 Views 563KB Size
Do Constraints to Growth Vary by Firm Size? Evidence from India Over a Period of Trade Liberalization*

Rana Hasan Asian Development Bank

Based on previous and ongoing research with Poonam, Gupta, Utsav Kumar, Karl Jandoc and Niny Khor.

Services has been the main engine of growth in 60 India, not manufacturing 50

Services

40

Agriculture

30 20

10 0

Manufacturing

Sectoral contribution to GDP Growth

Gupta et al (2009)

Surprising given the focus of reform measures on the manufacturing sector Cumulative Share of Delicensed Industries

100

150

Average (output) tariff rates, %

Source: Staff estimates.

0

50

Cumulative Share of Industries Delicensed

1985

1990

1995 year

2000

2005

Source: Aghion et al and Gupta et al.

Some comparisons with China … G r os s va l u e a d d e d , 1 9 8 0 - 2 0 0 4 ( in th o u s a n d s )

1990

1995

2000

2 00 5

1980

1985

1990

year I n d ia

1995

C h in a

I n d ia

C h in a

150

Capital stock per labor, 1980-2004

0 1980

2000

year

100

1985

50

1980

0

0

1 00 0

2 00 00

2 00 0

4 00 00

3 00 0

4 00 0

6 00 00

E m p l oy m e n t , 1 9 8 0 -2 0 0 4

1985

1990

1995

2000

2005

year India

China

Capital stock per labor in thousands of 1995 Chinese Yuan per labor.

2 00 5

In the meantime, Indian firms continue to remain small  Microenterprises (1-4) and small enterprises (5-49) account for 84% of total manufacturing employment in India (37.5 million out of 44.6 million in 2005)  This is a very high share in comparison to many comparators in the region (for which detailed size distribution data is available).

Employment by employment size groups

OAME, NDME, DME, and ASI establishments

NDME, DME, and ASI establishments

4.0e+06 2.0e+06

0

0

1.0e+07

mean of gplabor

6.0e+06

Employment by employment size groups

2.0e+07

3.0e+07

In fact, very small…

1-5

6-10

11-50

51-100

101-200

Source: Authors' estimates based on NSSO 2005-06 (R62) and ASI 2004-05

201+

1-5

6-10

11-50

51-100

101-200

Source: Authors' estimates based on NSSO 2005-06 (R62) and ASI 2004-05

201+

Why does this matter? Case of apparel  Many Indian apparel producers operate at very small scales  It is not profitable to utilize modern production methods at *Size is defined in terms of number of employees. very low scales  “Productivity is low not Spreading Machine because tailors are using the wrong technology given their size, but because tailoring firms are too small to benefit from the best technologies….” (Banerjee & Duflo) 8

What could be constraining (relative) dynamism of Indian manufacturing?       

Infrastructural deficiencies Labor regulation Financing constraints Hysterisis (e.g., small-scale industry reservations) Land acquisition Labor quality/skills Coordination failures and/or learning related externalities (e.g., electronics a-la Hausmann and Rodrik)

Quick aside on labor regulations  Many regulations at Central and State level  Industrial Disputes Act  Requires permission of government for laying off workers (for firms with 100+ workers since 1982)  Sets conciliation, arbitration, and adjudication procedures to be followed in the event of a dispute  Requires 21 days notice for changes to service conditions

Econometric studies on the impact of potential constraints on Indian manufacturing  Labor regulation: Besley and Burgess (2004); Aghion et al (2008); and Ahsan and Pages (2007)  Financing constraints: Banerjee and Duflo (2008)  Combination of constraints (infrastructure, labor regulations, and access to finance): Gupta et al (2008) and Li et al (2011)

 Aghion et al and Gupta et al examine the effects of potential constraints in the context of delicensing reforms.  Both exploit state level variation in the business environment (regulatory characteristics and infrastructure)  Gupta et al further exploit industry characteristics – e.g., if the bite of labor regulations is greater for labor intensive industries, do we see labor intensive industries doing relatively poorly in states with pro employee regualtions?

21 20.5 20 19.5

Log of Real Gross Value Added

20.5

21

21.5

22

21.5

22.5

Example: Value added across state type and industry type

1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 Year GVA in Flexible Labor Markets GVA in Inflexible Labor Markets

1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 Year GVA in Labor Intensive Industries in Flexible Labor Markets GVA in Labor Intensive Industries in Inflexible Labor Markets

Methodology yist = αis dis + βst dst+ θi trendi + γ (delicensingit ) + δ (industry characteristici * delicensingit) + π (state characteristics * delicensingit) + τ (state characteristics * industry characteristic i * delicensingit) + μ other controls + εist

Key results….  Post-delicensing: industries dependent on infrastructure, dependent on the financial sector and the labor intensive industries have grown less  Infrastructure, financial sector imperfections, labor regulations emerging as bottlenecks on growth?

 Post-delicensing: states with more developed infrastructure, and financial sector have grown faster.  Labor intensive industries have grown slowly, particularly in states with pro labor regulations.  Employment generation has been slower in states with pro labor regulations

An ongoing extension  Introduce informal manufacturing into the picture  Use establishment level data for formal and informal firms  Allow the effects of the business environment on firms to vary by firm size  Introduce trade liberalization into the analysis  Are the effects of trade liberalization contingent on the business environment and firm size?  In particular: Does growth in VA and employment vary by state-level differences in “business environment” characteristics and enterprise size?

Econometric specification (1)  Baseline: ln Yjkst = j + k + s + TRENDjt + γ (Zs * TRENDjt ) + εjkst  Yjkst: measure of industrial performance (gross value added or employment)  Z the set of state policy environment and characteristics  j:size group a firm belongs to, k: industry, s:state, t :time  G: size groups -- micro enterprises (1-5 workers), small firms (6-49 workers), medium-sized firms (50-199 workers), large firms (200 or more workers).  TREND: linear time trend, varies by the size group j  j , k ,s respectively denote size group, industry and state fixed effects.

Econometric specification (2)  Augmented ln Yjkst = j + k + s + θTARIFFkt + (Gj*TARIFFkt) + γ (Zs * TARIFFjt ) + τ(Zs * TARIFFjt * Gj) + εjkst

 Triple interaction term captures heterogeneity of effects of tariffs across firm-size groups in various business environment

Data: firms  Annual Survey of Industry:  Covers firms that are registered under the Factories Act (firms that use electricity and hire more than 10 workers)  Survey of registered firms with less than 100/200 workers  Census for registered firms bigger than 100/200 workers

 NSSO survey:  A once in five-year survey of “unorganized” (or informal) enterprises

 3 rounds in 1994, 2000, 2005  Build panel of state-industry data on value added and employment for four types of firms: micro enterprises (1-5 workers) and small (6-49 workers), medium (50-199 workers), and large sized (200 or more workers) firms.

Descriptive Statistics T a ble 1 . N um be r o f F irm s in A S I a nd N S S O , 1 9 9 4 , 2 0 0 0 a nd 2 0 0 5 1994 D a ta s e t A SI

S a m p le

2000

P o p u la tio n

S a m p le

2005

P o p u la tio n

S a m p le

P o p u la tio n

4 7 ,1 2 1

9 7 ,8 4 6

2 6 ,6 1 1

1 0 6 ,2 0 5

3 3 ,8 3 8

1 1 0 ,8 7 3

1 4 2 ,7 8 0

1 1 ,5 7 5 ,7 4 5

1 9 6 ,3 8 5

1 6 ,3 0 6 ,6 9 6

7 2 ,1 0 9

1 6 ,4 9 6 ,2 8 5

OAME

1 1 0 ,8 9 9

9 ,9 0 8 ,9 4 5

1 2 9 ,9 2 1

1 4 ,1 6 3 ,0 7 5

4 8 ,0 4 9

1 4 ,1 8 2 ,5 7 6

NDME

1 9 ,0 1 0

1 ,1 1 2 ,8 8 5

4 2 ,3 8 4

1 ,5 5 6 ,9 7 9

1 5 ,3 1 1

1 ,6 6 9 ,4 5 4

DME

1 2 ,8 7 1

5 5 3 ,9 1 5

2 4 ,0 8 0

5 8 6 ,6 4 2

8 ,7 4 9

6 4 4 ,2 5 5

N SSO o f w h ic h :

n o t e : A SI = A n n u a l Su r v e y o f I n d u s t r ie s ; N SSO = N a t io n a l Sa m p le Su r v e y O r g a n is a t io n Su r v e y o f U n o r g a n is e d M a n u f a c t u r in g E n t e r p r is e s O A M E = o w n - a c c o u n t m a n u f a c t u r in g e n t e r p r is e s ; N D M E = n o n - d ir e c t o r y m a n u f a c t u r in g e n t e r p r is e s ; D M E = d ir e c t o r y o f m a n u f a c t u r in g e n t e r p r is e s So u r c e : A u t h o r s c o m p u t a t io n s b a s e d o n A SI ( v a r io u s y e a r s ) a n d N SSO ( v a r io u s y e a r s )

State-level characteristics  Physical infrastructure indices (principal components): 

Kumar (2002) and Ghosh and De (2004)

 Financial development:  



Kumar (2002), Ghosh and De (2004); Proportion of firms in each state reporting “shortage of capital” (from NSSO); Proportion of firms in each state acquiring loans from any formal institution (from NSSO)

 Labor Market Flexibility:   



Besley and Burgess (2004)  Flex1 Hasan, Mitra, Ramaswamy (2006)  Flex2 Gupta, Hasan and Kumar (2009)  Flex3 [drawing upon Besley and Burgess, OECD (2007), and Bhattacharjea (2006 and 2008)] Share of contract workers in formal sector

 Product Market Regulations 

Gupta, Hasan and Kumar (2009) [drawing upon OECD (2007) and World Bank (2004)]

State-level characteristics P roduc t M a rke t P h y s ic a l I n f r a s tr u c tu r e

F in a n c ia l D e v e lo p m e n t

L a b o r R e g u la tio n s

R e g u la tio n s

H a s a n , M itr a

G u p ta ,

D if f ic u lty o f

B e s le y a n d

a nd

H a sa n a nd

L oa ns from

O b ta in in g

B urge ss

R am asw am y

K um a r

G u p ta H a s a n a n d

G hosh a nd D e

K um a r

G hosh a nd D e

K um a r

F orm a l S ourc e s

C a p ita l

( f le x 1 )

( f le x 2 )

( f le x 3 )

K um a r

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

A ndhra P ra de sh

1

0

1

0

0

1

1

1

1

0

A ssam

0

0

0

0

0

0

0

0

0

0

B ih a r

0

0

0

0

0

0

0

0

0

0

G u ja r a t

1

1

1

1

0

1

0

1

0

0

H a rya na

1

1

1

1

1

1

0

0

0

1

K a r n a ta k a

0

0

1

1

1

1

1

1

1

1

K e r a la

1

1

1

1

1

1

1

0

0

0

M a dhya P ra de sh

0

0

0

0

0

0

0

0

0

0

M a h a r a s h tr a

1

1

1

1

1

1

0

1

0

1

O r is s a

0

0

0

0

1

0

0

0

0

0

P u n ja b

1

1

0

1

1

0

0

0

0

1

R a ja s th a n

0

0

0

0

1

0

1

1

1

0

T a m il N a d u

1

1

1

1

0

1

1

1

1

1

U tta r P r a d e s h

0

0

0

0

0

0

0

0

1

0

W e st B e nga l

0

0

0

1

0

0

0

0

0

0

n o te : s e e te x t f o r d e s c r ip tio n s o f e a c h m e a s u r e ; F o r in f r a s tr u c tu r e a n d f in a n c e : 1 = m o r e d e v e lo p e d ; 0 = le s s d e v e lo p e d F o r la b o r r e g u la tio n s : 1 = p r o - e m p lo y e r ; 0 = p r o - e m p lo y e e F o r p r o d u c t m a r k e t r e g u la tio n s : 1 = c o m p e titiv e ; 0 = r e s tr ic tiv e

Results on growth in VA and employment  Relative to large firms, micro enterprises tend to grow slower, and small and medium-sized firms tend to grow faster in states with a better business environment.  These results are consistent with the notion that business environment related constraints to growth impinge the most on small and medium-sized firms.

Results on tariff reductions  Reductions in tariff rates are associated with increases in value added and employment among micro enterprises in states with better business environments.  In these states, small and medium-sized firms tend to experience lower growth of value added and employment relative to larger firms on account of reductions in tariffs.  It is difficult to say why the results for trade liberalization differ from those involving just the analysis of trend growth in value added and employment.  However, the results on trade liberalization are consistent with the recent work of Nataraj (2009)  To the extent that many of the elements of the business environment that we consider in this paper make for a more competitive environment, the effects of trade liberalization are likely to be felt more quickly and/or more fully in states with a better business environment.

Tying the results to the policy debate  Results on infrastructure are the least controversial  General acceptance of the notion that a “good” business environment is key for manfacturing dynamism  Results on labor regulation attract the most attention and criticism  Problems with the state-level coding  The issue of contract labor as a way of getting around labor laws

 As does the idea that (very) small firms may not be very dynamic

Aside: Can the coding be all that bad? Regular workers in labor-intensive industries, 2005

0

0

.05

.05

.1

.1

.15

.15

.2

.2

Regular and contract workers in labor-intensive industries, 2005

0

2

4

6 ln tot workers

Flex2=0

8 Flex2=1

10

0

2

4

6 ln reg workers

Flex2=0

8 Flex2=1

10

Perhaps implementation of the National Manufacturing Policy will give us the experiment we need…  National Investment and Manufacturing Zones  Fund for paying severance to workers  Third party inspections for compliance of both environment and labour norms.  A single window clearance mechanism to cut red-tape  Fiscal incentives, especially for the micro, small and medium enterprises.  Tax breaks for skill development institutes.  A Technology Acquisition Fund  Special support to employment-intensive industries to ensure job creation

Thank you Rana Hasan India Resident Mission Asian Development Bank

www.adb.org

Some features of the data  OAME dominate the national landscape. There were 14.2 million of such enterprises compared to around 1.4 million of the other two combined in 2005.  However, 76% of such enterprises are based in rural areas.  Once we think in terms of urban areas only, the distribution is more balanced (3.4 million OAME versus 1.4 million of the other two)  In fact, the other two generate more employment in urban areas than OAME -- 7.2 million versus 5.6 million, respectively. (The corresponding rural employment figures: 5.2 million versus 17.5 million!)

Growth by Size Group and Product Market Regulations (Trend) Dependent Variable: Log of Value Added (5)

Dependent Variable: Log of Employment

VARIABLES

(5) VARIABLES

MICRO SMALL MEDIUM TREND TREND*MICRO TREND*SMALL TREND*MEDIUM PMR*TREND PMR*MICRO*TREND PMR*SMALL*TREND PMR*MEDIUM*TREND Constant

Industry Indicators State Indicators Observations R-squared Robust standard errors in brackets *** p