VoxEU Column Development International trade Productivity and Innovation

Learning vs stealing: How important are market-share reallocations to India's productivity growth?

It is broadly agreed that trade liberalisation can increase productivity. The question is how. Earlier literature emphasises the role of firms “learning” to be more productive, whereas recent studies suggest that more productive firms are “stealing” market share from less productive ones, thus raising overall productivity. Presenting evidence from India’s trade liberalisation since 1991, this column finds evidence for both but argues that learning outweighs stealing.

There is a general consensus that trade liberalisation has the potential to increase the productivity of firms. But through what mechanisms does this occur? The earlier literature emphasised the idea that trade can cause existing firms to “learn” to be more productive by increasing managerial effort, accessing high-quality intermediate inputs, increasing economies of scale, and so on (Corden 1974, Grossman and Helpman 1991, Helpman and Krugman 1985). In contrast, the “new, new” trade theory stresses the importance of market-share reallocations. That is, the most productive firms “steal” market share from less productive firms, while the least productive firms are forced out (Bernard et al. 2003, Melitz 2003).

In recent research (Harrison et al. 2011), we use a major trade liberalisation episode in India to test the importance of each of these mechanisms in driving overall manufacturing productivity growth.

India’s trade liberalisation: A policy experiment

Prior to 1991, India had a highly restrictive trade regime, with average final goods tariffs on manufactured products of approximately 95%. In 1991, a combination of economic and political shocks – namely, a rise in oil prices, a decrease in remittances, and lower demand from abroad, and an unstable political climate – created a balance-of-payments crisis (Topalova and Khandelwal forthcoming). The Indian government called in the IMF, whose help was granted on the condition that India undertake several reforms.

Between 1990 and 2004, the tariff rates on manufactured products (final goods tariffs) were reduced and harmonised, and the average final goods tariff rate dropped from over 95% in 1990 to 30% in 2004. Similarly, the tariff rates on intermediate inputs used by the manufacturing industry (input tariffs) also fell sharply during this time. Several other reforms were enacted at the same time, including FDI liberalisation and the removal of the requirement for operating licenses (“delicensing”) in many industries. Table 1 shows the evolution of these policy changes over time.

Table 1. The evolution of trade, FDI and delicensing policies, 1985-2004

Year

Final Goods Tariffs (%)

Input Tariffs (%)

Fraction of FDI-Liberalised Industries

Fraction of Delicensed Industries

 

 

 

 

 

1985

88.98

58.38

0.00

0.34

1986

95.72

60.86

0.00

0.35

1987

95.08

59.16

0.00

0.35

1988

95.19

59.88

0.00

0.35

1989

95.90

59.99

0.00

0.36

1990

96.04

60.00

0.00

0.36

1991

96.04

59.99

0.36

0.84

1992

63.50

39.98

0.36

0.84

1993

63.92

38.69

0.36

0.85

1994

64.44

37.43

0.36

0.85

1995

53.41

30.20

0.36

0.85

1996

42.09

22.84

0.36

0.85

1997

34.01

18.45

0.43

0.89

1998

34.58

19.14

0.43

0.93

1999

35.60

20.24

0.43

0.93

2000

35.05

21.29

0.93

0.93

2001

34.18

20.63

0.93

0.93

2002

30.67

18.84

0.93

0.93

2003

30.81

18.94

0.93

0.93

2004

30.82

18.96

0.93

0.93

Source: Authors’ calculations based on data from Aghion et al. (2008), the UNCTAD-TRAINS database, and various publications of the Government of India.

Testing the role of reallocation in India

Figure 1 shows the evolution of total-factor productivity in India’s organised manufacturing sector between 1985 and 2004. The organised manufacturing sector is composed mainly of firms with more than 10 employees, and accounts for approximately 80% of manufacturing output. Figure 1 shows three measures of productivity, based on the decomposition suggested by Olley and Pakes (1996):

  • aggregate productivity (the sum of firm-level productivity, weighted by each firm’s market share),
  • average productivity (the unweighted sum of firm-level productivity), and
  • reallocation (the covariance between firm-level productivity and market share).

Productivity is normalised to zero in 1985, so that changes in productivity levels can be interpreted as growth since 1985.

Figure 1. All-India total-factor productivity, 1985 to 2004.

Note: The vertical line shows the start of the 1991 reforms. Source: Authors’ calculations, based on data from the Annual Survey of Industries. Data for 1995-1997 are omitted due to lack of availability of comparable firm-level data for these years.

Between 1985 and 2004, aggregate productivity grew by 19%, which implies an average annual increase of nearly 1% per year. When we consider the time period as a whole, nearly all of this increase (17%) can be attributed to growth in average productivity, rather than reallocation. However, Figure 1 suggests that there are three distinct phases between 1985 and 2004.

  • First, from 1985 to 1990, average productivity rose by over 8%, while the reallocation component actually fell by more than 6%, indicating that more productive firms lost market share to less productive firms.
  • Starting in 1991, this trend was reversed: average productivity fell, while reallocation productivity rose sharply.
  • By 1998, however, average productivity improvements were once again the more important driver of aggregate productivity growth. Reallocation productivity remained at approximately the level it achieved between 1992 and 1993, but rose no further.
How much productivity growth can the policy reforms explain?

We then run a “horse-race” between the different reforms to explore the extent to which they can explain productivity growth. We exploit the fact that the reforms varied across industries using a difference-in-differences approach. For example, we compare the changes in productivity across industries that received relatively large tariff cuts and those that received relatively small tariff cuts. This strategy allows us to control for macroeconomic changes that affected all industries in the same way at the same time. In addition, the fact that the reforms were driven by external requirements reduces the chance that industries were selected into the reforms for political reasons.

Table 2 summarises our results, showing the change in each type of productivity (aggregate, average, and reallocation) that can be explained by the policy reforms from 1985 to 2004. The results suggest that trade liberalisation, in particular the decline in input tariffs, is largely responsible for aggregate productivity growth. The average decline of 60 percentage points in final goods tariffs implies increases in aggregate and average productivity of approximately 3%, while the average decline of 40 percentage points in input tariffs implies increases in aggregate and average productivity of nearly 22%. The FDI liberalisation also plays a role, implying a 2.2% increase in aggregate productivity and a 4.6% increase in average productivity. The variation in policies across industries cannot explain the gains in reallocation productivity that were observed in the initial years following the reforms. However, the policies do explain the gains in average productivity, which was the more important driver of aggregate productivity growth during this period.

Table 2. The role of policy reforms

 

Decline in Final Goods Tariffs

Decline in Input Tariffs

FDI Liberalisation

Delicensing

Aggregate Productivity

3.2%

21.8%

2.2%

-0.4%

Average Productivity

2.6%

21.7%

4.6%

0.2%

Reallocation

0.6%

0.2%

-2.3%

-0.6%

Note: Productivity growth implied by policy changes. Bold values indicate that the underlying regression coefficients are statistically significant.

It is important to note that average productivity can increase if existing firms become more productive, but can also increase if relatively unproductive firms exit, or if relatively productive firms enter. Although firm identifiers are not available for the organised sector data during most of the time period we study, we construct a panel dataset by matching individual firms from one year of the survey to the next. The panel, though not representative of the entire organised manufacturing sector, allows us to isolate within-firm changes in productivity in a subset of relatively large manufacturing firms. Among these firms, we find that the final-goods tariff, input tariff, and FDI reforms can account for increases in within-firm productivity of 2.4%, 6.6%, and 2.9%, respectively.

Our results lend support to the importance of market-share reallocations in productivity growth. In the case of India, however, we show that such reallocations were only important at the beginning of the major trade liberalisation period, and that over the 20-year period from 1985 to 2004, average productivity improvements played a larger role in determining aggregate productivity growth.

References

Aghion, Philippe, Robin Burgess, Stephen J Redding, and Fabrizio Zilibotti (2008), “The Unequal Effects of Liberalisation: Evidence from Dismantling the License Raj in India”, The American Economic Review, September, 98(4):1397-1412.

Bernard, Andrew B, Jonathan Eaton, J Bradford Jensen, and Samuel Kortum (2003), “Plants and Productivity in International Trade”, The American Economic Review, 93 (4):1268-1290.

Corden, W Max (1974), Trade Policy and Economic Welfare, Clarendon Press.

Grossman, Gene and Elhanan Helpman (1991), Innovation and Growth in the Global Economy, MIT Press.

Harrison, Ann E, Leslie AMartin, and Shanthi Nataraj (2011), “Learning versus Stealing: How Important are Market-Share Reallocations to India's Productivity Growth?”, NBER Working Paper 16733.

Helpman, Elhanan and Paul R Krugman (1985), Market Structure and Foreign Trade: Increasing Returns, Imperfect Competition, and the International Economy, The MIT Press.

Melitz, Marc J (2003), “The Impact of Trade on Intra-Industry Reallocations and Aggregate Industry Productivity”, Econometrica, 71(6):1695-1725.

Olley, G Steven and Ariel Pakes (1996), “The Dynamics of Productivity in the Telecommunications Equipment Industry”, Econometrica, 64(6):1263-1297.

Topalova, Petia and Amit Khandelwal (forthcoming), “Trade Liberalisation and Firm Productivity: The Case of India”, The Review of Economics and Statistics.

1,050 Reads