India: Selected Issues and Statistical Appendix
Author: International Monetary Fund
Publication Date:
03 Sep 2002
eISBN:
9781451818567
ISBN:
9781451818567
Language:
English
Keywords:
ISCR; CR; bank; India; deficit; banking sector; infrastructure investment; investment rate; investment decision; service sector; government investment; Private investment; Labor force; Production growth; Total factor productivity; Agricultural sector; Global
Download PDF (858.4 KB)

This Selected Issues paper presents an analysis of trends in growth and investment in India in the 1990s, with a focus on the slowdown in growth during the second half of the 1990s. The paper discusses the fiscal situation, outlining the key reasons for the deterioration in fiscal balances, how the fiscal situation compares with other developing countries, and the key lessons from countries that managed successful fiscal consolidation. The paper also contains an assessment of India’s opening to global trade and factors that may be affecting India’s export performance.

Abstract

This Selected Issues paper presents an analysis of trends in growth and investment in India in the 1990s, with a focus on the slowdown in growth during the second half of the 1990s. The paper discusses the fiscal situation, outlining the key reasons for the deterioration in fiscal balances, how the fiscal situation compares with other developing countries, and the key lessons from countries that managed successful fiscal consolidation. The paper also contains an assessment of India’s opening to global trade and factors that may be affecting India’s export performance.

II. Recent Trends in Growth and Investment1

A. Introduction

1. Following the policies implemented in response to the 1991 balance-of-payments crisis, economic growth in India accelerated in the mid-1990s (Figure II.1).2 Annual GDP growth (at factor cost) in the five years to 1996/97 was 6¾ percent, the highest five-year average (based on a moving average) recorded in India since 1950/51. Economic strength during this period has been largely ascribed to the fiscal consolidation and structural reforms that were initiated after the 1991 crisis (e.g., see Callen, et al. (2001) and Chopra, et al (1995)). Reforms included delicensing and deregulation of the industrial sector, liberalization of private and foreign investment and trade, tax reforms, and measures to liberalize and strengthen the supervision of the financial sector. The benefits of reform were most evident in private fixed investment growth, which surged to an average of 15¼ percent in the period.

Figure II.1
Figure II.1

Annual GDP Growth Since 1950s

(At factor cost, in percent)

Citation: IMF Staff Country Reports 2002, 193; 10.5089/9781451818567.002.A002

Source: Central Statistical Organization (CSO).

2. In the late-1990s, however, economic activity weakened substantially. Growth in 2000/01 was only 4 percent, and in the five years to 2001/02 averaged 5¼ percent (Figure II.2).3 Moreover, growth during this period may have been overestimated by about ¼rcentage point a year on average because of the impact on growth estimates of civil service wage hikes related to the Fifth Pay Commission awards.4 While this growth rate still compared favorably with most other developing countries, it fell short of the government’s 6½–7 percent target (based on the objectives of the 9th Economic Plan) and the estimated 8 percent annual growth needed to meet the government’s ambitious poverty reduction objectives.5 The slowdown during the late-1990s was also broad-based across sectors, particularly in agriculture and industry. On the demand side, it largely reflected lackluster private fixed investment growth—which plummeted to an average of only 3¾ percent.

Figure II.2
Figure II.2

GDP Growth

(At factor cost, in percent)

Citation: IMF Staff Country Reports 2002, 193; 10.5089/9781451818567.002.A002

Source: CSO.

3. This chapter examines the slowdown in output and private investment growth during the late-1990s. The analysis suggests that while cyclical and exogenous factors—such as poor weather conditions, a weak external environment, and natural disasters—contributed to the deceleration in activity, there was also a slowdown in trend growth, perhaps reflecting lingering structural distortions—including high real interest rates (partly associated with large fiscal deficits), severe infrastructure bottlenecks, and remaining industrial and agricultural controls. Private investment growth, particularly, appears to have been affected adversely by the deteriorating quality of public expenditures, in addition to structural factors, such as the legal and regulatory framework and labor market rigidities. Results based on production function and filtering methods indicate that trend or potential output growth may have fallen to 6 percent or less, even assuming some rebound in productivity and investment growth under the premise of modest fiscal reform and consolidation and a pickup in the pace of structural reform.

4. The rest of this chapter is organized as follows. The next section examines the slowdown in GDP growth in the Iate-1990s, particularly some of the factors behind it and to what extent the deceleration in activity was cyclical versus structural. Section C presents an analysis of the slowdown in private investment. Section D examines future growth prospects, primarily by estimating potential output growth using a production function framework. Section E provides some concluding remarks.

B. The Growth Slowdown

The Composition of the Slowdown

5. On a sectoral basis, the slowdown in growth was most apparent in the agriculture and industrial sectors, while growth in the service sector was comparatively resilient (Figure II.3a and Table II.1). Agricultural growth slowed to an average of 2 percent in the late-1990s—less than half the rate in the mid-1990s—and industrial sector growth also decreased substantially to 4½ percent. Growth in these sectors was undermined, inter alia, by increasing infrastructure constraints and remaining agriculture and industrial controls, including on storage, production, and movement of agricultural products; on bankruptcy, restructuring, and labor market rules; and with small-scale industry reservations. Service sector growth remained buoyant, although annual growth, excluding government services, declined slightly. The slowdown in the second half of the 1990s more than reversed the improvement in agriculture and industrial sector growth made in the earlier part of the decade compared to 1980s, making the economy even more reliant on the service sector (which accounted for almost 50 percent of GDP in 2001/02). It should be acknowledged, nonetheless, that the comparative strength of the service sector may reflect the limited applicability of product and factor market regulations on firms in the sector and greater progress with reform (e.g., deregulation of telecommunications and insurance), which also allowed India to develop comparative advantage in providing IT services.

Figure II.3a.
Figure II.3a.

Contributions to Annual GDP Growth Output Components

(In percentage points)

Citation: IMF Staff Country Reports 2002, 193; 10.5089/9781451818567.002.A002

Table II.1.

Growth of Real GDP and Components

(Annual averages, in percent)

article image
Source: Staff estimates based on data provided by the CSO.

Includes statistical discrepancy on consumption.

Includes pro-rated errors and omissions on investment.

6. On an expenditure basis, private domestic demand was particularly weak, while net exports—notwithstanding the turbulent external environment—and public spending supported growth (Figure II.3b). Although private fixed investment contributed the most to the decline in GDP growth, the growth rate of private consumption also fell by about a third to an average of 4 percent, and its contribution to the overall slowdown is comparatively large because it accounts for about 65 percent of GDP. In contrast, estimated real export growth increased from 9½ percent to 16½ percent, and overall, net exports contributed an average of ½ percentage point to GDP growth in the late1990s compared to negative 1 percentage point in the earlier period. Growth in public expenditure also increased substantially—in particular, the growth rate of real public consumption growth more than doubled to an average of 10½ percent. As this surge in expenditure had only limited spillovers to the private economy, it raises questions about the inefficiency of public expenditures, and also suggests that public sector dissaving in recent years may have crowded out the private economy.

Figure II.3b.
Figure II.3b.

Contributions to Annual GDP Growth 1/

Expenditure Components (In percentage points)

Citation: IMF Staff Country Reports 2002, 193; 10.5089/9781451818567.002.A002

Source: Staff estimates based on data provided by the CSO.1/ Private consumption includes statistical discrepancy on consumption. Private investment and inventories include pro-rated errors and omissions on investment.

Factors behind the Slowdown: Are they Cyclical or Structural?

7. An approach used in the macroeconomic literature and applied in this chapter to estimate trend growth is to smooth the underlying series using the Hodrick-Prescott (1997) filter, which is designed to filter out business cycle fluctuations.6 One difficulty with this approach is that trends become poorly defined at the beginning and end of the sample period. To address this problem, trend growth was also estimated by extending the sample period based on forecasts of GDP growth.

8. The results suggest that the slowdown had both cyclical and structural factors (Table II.2):

  • Trend GDP growth fell to under 6 percent—and perhaps as low as 5 percent—in 2001/02.

  • Trend GDP growth reached a peak in the mid-1990s, but at 6.1 percent was substantially below the average growth rate of 6½ percent in that period.

  • On a sectoral basis, trend growth fell most significantly in the industrial sector. As noted above, estimates of trend growth are poorly defined at each end of the sample period, particularly if there are sharp changes in growth around the end points. As such, the relatively low estimate of services sector trend growth based on the unextended series (compared to the extended series) reflects this end-period problem, along with the sharp drop in service sector growth in 2000/01.

Table II.2.

Trend Growth Based on Hodrick-Prescott Filter 1/

(In percent)

article image
Source: Staff estimates.

Estimated using λ = 100, the standard value for annual observations. Sample period was 1950/51 − 2001/02 (or 2007/08 for the extended series).

9. One important exogenous factor affecting agricultural growth that could have contributed to the slowdown was weather conditions, although weak growth in the agriculture sector may also reflect structural factors. Agriculture remains very vulnerable to the monsoon. The correlation (from 1970/71 to 2001/02) between rainfall during the monsoon season and agriculture growth was 0.70, and econometric analysis suggests that rainfall was a significant explanatory variable for agriculture growth (Box II.1). Moreover, although agricultural output accounts for only about a quarter of GDP, the impact of agricultural activity is magnified through its effect on rural incomes and consumption, as more than 70 percent of the population is rural. Thus, the overall correlation between rainfall and GDP growth (at factor cost) was 0.67. It is notable, nonetheless, that rainfall during the monsoon season has been considered normal (or within 10 percent of the long-run average) for the past 13 years. This suggests weakness in the agriculture sector was also partly structural in nature, reflecting weak rural investment, high food stocks, limited agricultural diversification, and continuing regulations on production, storage, and transport.

Agriculture Sector Growth in India

Agricultural growth remains vital to overall economic performance in India. Spillovers from growth in the sector to other components of domestic demand continue to be substantial, even though the share of agriculture in GDP fell from over 55 percent (in real terms) in 1950/51 to about 24 percent in 2000/01. In particular, the long-run correlation between private consumption growth and agricultural growth was over 0.75, reflecting the large share of the population that is still rurally based (Table).

This box examines the determinants of agriculture sector growth—especially the influence of rainfall during the monsoon season. Growth in the sector was particularly weak in a number of recent years, with the poor performance often attributed to weather conditions. Indeed, over the long run, rainfall and agricultural growth were highly correlated (Figure). To assess the importance of rainfall and other factors in explaining agricultural growth, a model (all variables are in logs) was estimated on annual data from 1950/51 to 2000/01, using ordinary least squares with White heteroskedasticity-consistent standard errors (T statistics are given in parenthesis). The estimation result was:1/
Agr=-0.81−0.47(-7.55)(-5.53)Agr(-1)+0.41(8.07)Rain+0.21(3.82)Pagr(-1)-0.84(-1.90)Stock(-1)R-squared=0.69;AdjustedR-squared=0.67;Durbanh-statistic=1.78;

where Agr was agricultural output growth, Rain was an All-India index of rainfall during the monsoon (in cm.), Pagr was the relative price of agriculture (defined as the agriculture output deflator in the national accounts divided by the GDP deflator), and Stock was the change in agriculture stocks (or inventories).

The results indicate that rainfall and the relative price of agriculture had significant and positive effects on agriculture output growth, while lagged agricultural growth and agriculture stocks had negative effects. While these findings confirm the importance of weather conditions, they also highlight the negative impact of high food stocks and the importance of food prices in explaining growth in the sector.2/ Although this equation may be too simple to provide definitive conclusions on the impact of the Public Distribution System (PDS) on agricultural growth, the data does suggest that increased food stocks (which are likely to be mainly foodgrains and not other commodities that are not covered by the PDS) tend to depress growth in the next period.

Correlations with Agriculture Sector Growth

article image
Source. Staff estimate.

Excluding government services.

A02ufig01

Rainfall and Agriculture

(In percent)

Citation: IMF Staff Country Reports 2002, 193; 10.5089/9781451818567.002.A002

Sources: CSO for agriculture growth r: B. Parthasarathy (March 2001) and the Economic Survey 2001–02 for rainfall.
1/ The Q-statistic and LM test indicated no significant serial correlation of the residuals.2/ The spatial and temporal distribution of rainfall could also affect agricultural growth, but these cannot be tested in this estimation framework.

10. Another factor that could have contributed to the deceleration in economic activity was the external environment, although the direct impact on total exports was limited. During the late-1990s, global trade was adversely affected by a number of shocks—including the East Asian crisis, a significant oil price shock, weak commodity prices, and the recent global growth slowdown (Figure II.4a). Notably, though, real export market growth (real merchandise and services import growth of India’s trading partners weighted by their share in India’s merchandise exports) declined only modestly through 2000/01, partly reflecting the comparatively robust growth of India’s main export markets—the United States (21½ percent of India’s merchandise exports) and the United Kingdom (6 percent).7 In addition, services export growth surged, reaching an average of 26 percent (in dollar terms) in the late-1990s to 2000/01 (Figure TT.4b).8 Thus, the dollar value of total exports declined only slightly, and in volume terms, exports are estimated to have increased.

Figure II.4a.
Figure II.4a.

The External Environment 1/2/

(Percent change)

Citation: IMF Staff Country Reports 2002, 193; 10.5089/9781451818567.002.A002

Figure II.4b.
Figure II.4b.

Exports Growth 1/3/

(In dollar value, percent change)

Citation: IMF Staff Country Reports 2002, 193; 10.5089/9781451818567.002.A002

Sources: World Economic Outlook database; and Reserve Bank of India (RBI).1/ Data through 2000/01.2/ IMF estimates. Export and oil prices in dollar terms. Real export market is estimated based on real goods and services imports of trading partners weighted by their share of India’s exports, and terms of trade and export prices on trade prices weighted by the commodity composition of India’s trade.3/ Balance-of-payments basis.

11. The external environment might, nonetheless, have negatively affected the economy through terms of trade and exchange rate effects. These effects may have been most felt in the industrial sector. In particular, over the long run, industrial sector growth was positively correlated with changes in the terms of trade and negatively correlated with changes in oil prices (Table II.3). In the 1990s, industrial sector growth was also highly correlated with export prices. These relationships may reflect the impact on industrial sector profits of changes in these prices. In addition, merchandise export growth, which decreased in dollar terms from 13¼ percent in the mid-1990s to 7 percent in the late-1990s (through 2000/01), could have been adversely affected by the surge in services exports during the latter part of the decade, as the strength of these exports might have kept the rupee stronger than would have been otherwise.

Table II.3.

Correlations with Industrial Output Growth

(In percent)

article image
Sources: RBI; and World Economic Outlook database.

1977/78-2000/01

C. The Slowdown in Private Investment Growth

12. Following a secular rise starting in the early 1950s, the domestic investment rate in India stagnated in the 1990s (Figure II.5a). The investment rate peaked in 1995/96 at 27 percent of GDP and subsequently fell to 24 percent of GDP in 2000/01. In real terms, the decline was smaller (from 27¼ percent of GDP to 26¼ percent of GDP) during the same period, reflecting the decrease in the relative price of investment goods. In particular, the private fixed investment rate (in real terms) fell by 1¼ percent of GDP to under 18½ percent of GDP in the late-1990s, and the private corporate fixed investment rate dropped from 11 percent of GDP to 6¾ percent of GDP (Figure II.5b). This section of the chapter examines some of the potential causes of the slowdown in investment.

Figure II.5a.
Figure II.5a.

Domestic Investment Since the 1950s 1/

(In percent of GDP)

Citation: IMF Staff Country Reports 2002, 193; 10.5089/9781451818567.002.A002

Figure II.5b.
Figure II.5b.

Real Private Investment in the 1990s 1/

(In percent of GDP)

Citation: IMF Staff Country Reports 2002, 193; 10.5089/9781451818567.002.A002

Source: CSO.1/ Inventories pro-rated errors and omissions on investment.

13. The literature suggests a number of macroeconomic, microeconomic, and structural factors that could influence investment decisions and aggregate investment rates. Macroeconomic factors include domestic and foreign output growth (and expectations of growth), direct costs related to funding investment (the real interest rate, relative price of investment goods, and other input costs), credit availability (real growth of private sector credit), leverage levels (debt in relation to GDP or equity), and macroeconomic policies and policy uncertainty (including related to political stability).9 Microeconomic and structural factors include the regulatory and legal framework (including entry and exit policies), labor market flexibility, openness to trade, infrastructure, and transactions costs related to the regulatory burden, governance, and corruption. Many of these structural factors cannot be measured directly (and even indirect measures are generally not available on a time series basis), but surveys provide strong evidence that they are impediments to investment—particularly in India.10

14. A model of private investment growth in India was estimated starting with a broad set of potential variables. The regressions were estimated based on annual data from 1970/71 to 1999/2000 and using ordinary least squares with White heteroskedasticity-consistent standard errors. A number of the variables—including lagged output growth, lagged investment growth, inflation, real interest rates, real credit growth were found to be insignificant. The final estimation result was:11

IP=0.07−0.92(0.07)(-3.05)IG+4.56(1.81)WGDP+0.93(3.41)IGinfra(-1)−0.70(-1.73)ExG(-1)−0.07(-2.03)VINFLR-squared=0.46;AdjustedR-squared=0.35;DWstatistic=2.30.

Where IP was private investment growth (in logs), IG was public sector investment growth (in logs), WGDP was world output growth (in logs), IGinfra was public sector infrastructure investment growth (in logs), ExG was public expenditure growth excluding infrastructure investment (in logs), and VINFL was the monthly variance of WPI inflation. Private investment included inventories and errors and omissions; government investment included inventories; and infrastructure investment was investment in agriculture, electricity, gas, and water, and transportation, storage, and communication.

15. The estimation results indicate that:

  • Of all public sector expenditures, only public sector investment in infrastructure had a positive effect on private investment behavior.12 Higher growth in public consumption or other public investment inhibited private investment growth.

  • World GDP growth had a positive effect on private investment growth, while uncertainty related to inflation volatility had a negative effect.

  • Almost 70 percent of the slowdown in private investment in the late 1990s was attributed to a deterioration in the composition of public expenditures, which shifted towards public consumption and non-infrastructure investments after 1995/96 compared to the earlier part of the decade (Table II.4). This estimate is based on the regression results for the model of private investment and made by multiplying the regression coefficients by the change in the explanatory variables between the early part of the 1990s (before 1995/96) and the later part of the decade. The estimated impact is roughly divided equally between the negative effect of weaker growth of public infrastructure investments (lagged IGinfra) and of faster growth of other public spending (IG and lagged ExG).

  • As noted above, many variables, particularly structural factors, that influence the investment climate are not directly observable and thus could not be tested. This may explain the relatively low explanatory power of the estimation results.

Table II.4.

The Private Investment Slowdown

article image
Sources: Staff estimates; CSO; World Economic Outlook database; and International Financial Statistics database.

Based on the regression results for the model of private investment and made by multiplying regression coefficients by the change in the explanatory variables (in logs, except for VINFL) between the early part of the 1990s (before 1995/96) and the later part of the decade.

Lagged one year.

Excluding expenditures on public sector infrastructure investment.

D. Future Growth Prospects

16. Estimates of potential output growth are regularly used to access growth prospects in an economy. A difficulty, however, is that potential output is not well defined and also problematic to measure.13 In broad terms, the economic literature proposes two distinct definitions of potential output—one in the Keynesian tradition and the other in the neoclassical one. In the former, business cycles are related to changes in resource utilization as output deviates from its underlying potential due to movements in aggregate demand in relation to a more slowly-moving level of aggregate supply. In the latter, exogenous productivity shocks to aggregate supply determine trend and long-run growth, and the lagged impact of rational agents reacting to unexpected shocks determines short-run or business-cycle fluctuations. The literature also presents a number of methods to estimate potential output, including purely statistical ones—such as smoothing filters (e.g., the Hodrick-Prescott filter) and unobservable components methods—and structural methods—such as aggregate production functions, structural vector autoregressions (e.g., Blanchard and Quah (1989)), and demand-side models related to direct measures of spare capacity in the economy (Bayoumi (2000)).

17. In this chapter, potential output for India was estimated based on the aggregate production function methodology. As a first step, estimates of total factor productivity (TFP) were derived residually using historical data. It was assumed that the production function has a Cobb-Douglas specification, in which output, Y, depends on the level of technology (or TFP), A, and factor inputs L, the labor force, and K, the stock of physical capital:

Y = ALα Kβ

Where α and β sum to one and are, respectively, the labor share of income and the capital share of income.14 Under this specification, it was assumed that the aggregate production technology has constant returns to scale, labor and capital are homogeneous and fully employed, and (factor and product) markets are competitive.15 To the extent that these assumptions are incorrect, inputs and outputs are mismeasured, and hours per worker change over time, the derived level of TFP will be an inaccurate measure of the underlying or trend level of technology. In this chapter, as an alternative, the specification was also extended to account for changes in human capital or the education-level of workers. For this specification, the benefits of education were assumed to be embodied in labor.

18. Table II.5 summarizes the historical data underlying the estimates of TFP.

  • Annual labor force growth accelerated in the 1990s to about 2¼ percent, even though annual population growth fell to under 2 percent.

  • The comparative increase in labor force growth reflected a decreasing dependency ratio, as the labor force participation fell during the 1990s compared to the 1980s.

  • While net capital stock growth increased in the first part of the 1990s, the growth rate decelerated in the latter part of the decade, reflecting a lower total investment rate.

Table II.5.

Selected Macroeconomic Data

(Annual averages, in percent)

article image
Sources: Labor force and dependency ratio from World Bank World Development Indicators database; population and capital stock from the CSO; and average years of schooling from Barro and Lee (2000).

Labor force as a percent of working-age (ages 15–64) population.

Dependents as a percent of working-age population.

19. Table II.6 summarizes the derived estimates of TFP growth.

  • By most measures (labor force only or labor force augmented for average years of schooling), TFP growth was roughly unchanged (or increased slightly) in the post-crisis 1990s compared to the 1980s.16

  • The derived TFP growth rate decreased substantially in the late-1990s. It is important to note, however, that this decline, which occurred over a short-time horizon, may have partly reflected changes in the utilization rates of labor and physical capital, and thus may have been partly cyclical in nature.17

Table II.6.

TFP Growth 1/

(Annual averages, in percent)

article image
Sources: Staff estimates.

Specification: L = labor force or labor force multiplied by average years of schooling in the production function equation above.

20. Based on these calculations of TFP growth and the corresponding underlying production function model, potential output growth was estimated. The estimates were based on long-run (post-crisis 1990s) averages and, alternatively, short-run (late-1990s) averages of TFP growth and capital stock growth.18 Labor force growth was assumed to average 2.0 percent, based on World Bank projections of annual working-age (ages 15–64) population growth in India during 2000–05.19 The growth rate of human capital or schooling was assumed to remain the same as in the 1990s.

21. Estimates of potential output growth that were based on these assumptions are presented in Table II.7 and lie in the range of 5 to 6 percent.

  • These estimates were essentially insensitive to the inclusion of the human capital or schooling variables (not shown). When human capital is excluded in deriving the estimate of TFP, the contribution of human capital to growth is included in TFP growth (as TFP is derived residually), and thus generally does not affect the calculation of potential growth.

  • The estimates were also only slightly sensitive to different assumptions of the labor share of income.

  • The sensitivity of the estimates to different assumptions of capital stock growth was limited. The capital stock growth rate assumptions varied by only about ¼ percentage point, and the contribution of capital stock growth to potential growth is weighted by the income share of capital (or a maximum of 0.4 in this exercise). So at most, the differences in capital stock growth assumptions explained only about a tenth of a percentage point of the variation in potential growth estimates.

  • The estimates of potential output growth were most affected by assumptions of TFP growth. Since the assumptions based on short-run averages were more likely to be affected by cyclical factors (as discussed above), estimates based on the longer-run averages may better reflect the underlying structural growth rate. However, to achieve growth rates near 6 percent, TFP and investment growth rates would need to accelerate from current levels.

  • The estimates of potential output growth based on an aggregate production function were similar to estimates of trend growth made by smoothing historical data (e.g., using the Hodrick-Prescott filter). This is not surprising as estimates of TFP growth were derived residually and essentially smoothed.

Table II.7.

Potential Output Growth 1/

(In percent)

article image
Sources: Staff estimates.

Specification: Labor share as noted; Capital stock and TFP growth are based on averages for the post-crisis 1990s or the late-1990s.

E. Concluding Remarks

22. Economic activity weakened in India during the late-1990s, with average growth rates falling to 5¼ percent compared to 6¾ percent during the earlier part of the decade. The slowdown was broad-based across sectors, and on the demand side, largely reflected weak growth in private fixed investment.

23. This chapter examined the growth and private investment slowdown in India during the late-1990s. The analysis indicated that both cyclical and structural factors contributed to the deceleration in growth. In particular, poor weather conditions, along with weak rural investment, high food stocks, limited agricultural diversification, and continuing regulations on production, storage, and transport, led to lackluster agriculture growth with spillovers to the rest of domestic demand. The direct impact of a weaker external environment was limited; however, changes in relative trade prices could have adversely affected growth, especially in the industrial sector. Private investment appeared to have been inhibited by the deteriorating quality of public expenditures—namely, the shifting of expenditures from infrastructure investments to consumption and other investment—in addition to structural factors, such as the legal and regulatory framework and labor market rigidities.

24. Results based on production function and filtering methods indicate that potential output or trend growth may have fallen to 6 percent or less. Even this estimate presumes some rebound in productivity and investment growth under the premise of modest fiscal reform and consolidation—including improved composition of public sector expenditures—and a pickup in the pace of structural reform.

References

  • Acharya, Shankar, 2001, “Macroeconomic Management in the Nineties,” unpublished, (March), (New Delhi: Indian Council of Research on International Economic Relations).

    • Search Google Scholar
    • Export Citation

  • Barro, Robert J., and Jong-Wha Lee, 2000, “International Data on Educational Attainment: Updates and Implications,” unpublished, Harvard University, (February), available at http://www.cid.harvard.edu.

    • Search Google Scholar
    • Export Citation

  • Bayoumi, Tamim, 2000, “Where are We Going? The Output Gap and Potential Growth,” in Tamim Bayoumi and Charles Collyns (eds.) Post-Bubble Blues: How Japan Responded to Asset Price Collapse, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Blanchard, Olivier J., and Danny Quali, 1989, “The Dynamic Effects of Aggregate Demand and Aggregate Supply.” American Economic Review, 79(4), pp. 655–73.

    • Search Google Scholar
    • Export Citation

  • Callen, Tim, Patricia Reynolds, and Christopher Towe (eds.), 2001, India at the Crossroads, (Washington: International Monetary Fund).

  • Cerra, Valerie, and Sweta Chaman Saxena, 2000, “Alternative Methods of Estimating Potential Output and the Output Gap: An Application to Sweden,” IMF Working Paper 00/59.

    • Search Google Scholar
    • Export Citation

  • Chopra, Ajai, Charles Collyns, Richard Hemming, and Karen Parker, 1995, India: Economic Reform and Growth, IMF Occasional Paper 134, (December).

    • Search Google Scholar
    • Export Citation

  • Collins, Susan M., and Barry P. Bosworth, 1996, “Economic Growth in East Asia: Accumulation versus Assimilation,” Brookings Papers on Economic Activity, No. 2, pp. 135–203.

    • Search Google Scholar
    • Export Citation

  • Goswami, Omkar, David Dollar, and others, 2002, “Competitiveness of Indian Manufacturing: Results from a Firm-Level Survey,” (January), Confederation of Indian Industry and the World Bank.

    • Search Google Scholar
    • Export Citation

  • Government of India, 2002. Economic Survey, 2001-02, (New Delhi: Ministry of Finance, Economic Division).

  • Hodrick, Robert J., and Edward C. Prescott, 1997, “Postwar U.S. Business Cycles: An Empirical Investigation,” Journal of Money, Credit, and Banking, Vol. 29, pp. 1–16.

    • Search Google Scholar
    • Export Citation

  • Lee, II Houng, and Yougesh Khatri, 2001, “Potential Output and Inflation,” Chapter III in Malaysia: From Crisis to Recovery, IMF Occasional Paper 207.

    • Search Google Scholar
    • Export Citation

  • Nitsure, Rupa Rege, and Mathew Joseph, 1999, “Liberalisation and the Behaviour of Indian Industry,” ICICI Working Paper (July), available at ICICIresearchcentre.org.

    • Search Google Scholar
    • Export Citation

  • Parthasarathy, B., 2001, “All-India Monsoon Rainfall Index,” (March) available at http://grads.iges.org/india/allindia.html.

  • Reserve Bank of India, 2002, Report on Currency and Finance, 2000-01 (Mumbai).

  • Scacciavillani, Fabio, and Phillip Swagel, 1999, “Measures of Potential Output: An Application to Israel,” IMF Working Paper 99/96.

  • Serven, Luis, and Andres Solimano (eds.), 1993, Striving for Growth after Adjustment: The Role of Capital Formation, (Washington: World Bank).

    • Search Google Scholar
    • Export Citation
1

Prepared by Ranil Salgado (x34182), who is available to answer questions.

2

In this chapter, the mid-1990s are defined as 1991/92–1996/97 (the period of India’s 8th Five-Year Economic Plan), the late-1990s as 1996/97–2001/02 (the period of the 9th Five-Year Economic Plan), and the post-crisis 1990s as the combination of the two periods. Data presented in this chapter reflect official data released through May 2001.

3

National accounts data for 2001/02 are based on advance estimates (released in January 2002) from the Central Statistical Organization (CSO) and are only available for GDP at factor cost and some output components. Data of GDP at market prices and expenditure components are through 2000/01 (the latest available). As national accounts data on an expenditure basis in India are very limited, the analysis is partly based on staff estimates.

4

The CSO estimates government services (identified as public administration and defense in the national accounts) using the government’s wage bill (with arrears counted in the year that they are paid) deflated by the consumer price index for industrial workers Acharya (2001) calculates that this incorrect deflation of government services created a “spurious” addition to growth of about 1/2 percentage point a year during the three years from 1997/98 to 1999/00. This estimate is roughly consistent with growth excluding government services, which averaged 5 percent during the late-1990s.

5

The initial target for the 9th Plan, approved by the National Development Council in January 1997, was 7 percent. This target was subsequently revised down to 6 ½ percent in 1998/99, based on the weak outcome for the previous year—the first year of the Plan.

6

This method is particularly useful when there are data limitations. For India, time series of GDP and its components are generally limited to annual observations. Quarterly data of GDP at factor cost and some output components are available, but only for data since 1996/97.

7

The importance of the United States and the United Kingdom in India’s trade may be underestimated in the estimate of the export market growth as the weights used in the calculation are based on merchandise trade. According to the National Association of Software and Services Companies (NASSCOM), an umbrella organization for IT software and services companies in India, over 60 percent of exports of IT-related services in 2000/01 were to the United States, while over 10 percent were to the United Kingdom.

8

In first three quarters of 2001/02 on a balance-of-payments basis, merchandise exports fell by 1 percent (in dollar terms, compared to the same period in 2000/01), while services exports increased by 17 percent.

9

For example, see Serven and Solimano (1993).

10

See Goswami, et al (2002) for a recent study on the investment climate in India, which covers many of these factors.

11

T statistics are given in parentheses. Tests of serial correlation—the Q-statistic and LM test—indicated no significant serial correlation of the residuals. Data were from the CSO, the Reserve Bank of India (RBI), the World Economic Outlook database, and the International Financial Statistics database.

12

The Report on Currency and Finance, 2000–01 also found a similar result.

13

See Scacciavillani and Swagel (1999), Cerra and Saxena (2000), and Lee and Khatri (2001), including for details on other methods to measure potential output.

14

Factor income shares are not available or are difficult to measure in most developing countries, including India. In the literature, estimates or assumptions of the labor share of income range from 0.6 to 0.7 (Collins and Bosworth (1996)). In this chapter, the results were estimated with the labor share parameter assumed to be in this range.

15

Alternatively, if markets are not competitive, the parameters α and β could be assumed to be factor elasticities of output.

16

By construction, the differences between TFP growth estimates based on labor force only and labor force augmented for schooling are equivalent to the contributions to growth of increased schooling or human capital. Also it is notable that average TFP growth rates during the 1990s, including the 1990/91 crisis year (averages not shown), were lower than during the 1980s.

17

Directly measured time-series data of unemployment, hours per worker, or capacity utilization are unavailable in India on an aggregate basis. The Report on Currency and Finance, 2000–01 presented estimates of indirect measures of capacity utilization (using the Wharton Index and the Minimum Capital-Output Ratio Measure) in the industrial sector that suggest that capacity utilization decreased substantially in manufacturing, but increased in electricity and, to a lesser extent, in mining and quarrying during the late-1990s. Nitsure and Joseph (1999) estimated capacity utilization of the private corporate industrial sector for five years in the 1990s based on data from 802 medium and large-scale companies. These estimates indicated that industrial sector capacity utilization increased from 1993/94 to 1996/97 before falling slightly in 1997/98.

18

Note that TFP and capital stock trend growth rates based on the Hodrick-Prescott filter and the peak decadal growth rates fall within the range of these assumptions. For the growth rate of the capital stock, the assumptions are also broadly consistent with projected investment growth rates.

19

A regression of the labor force participation rate on a constant and trend from 1960 to 2000 indicated that labor force participation in India declined by about ¼ percentage point a year. However, the labor force participation rate rose slightly during the 1990s. For the exercise in this chapter, the participation rate was assumed to be constant.

Same Series

Other IMF Content

Other Publishers

Cover IMF Staff Country Reports
India: Selected Issues and Statistical Appendix
Author: International Monetary Fund
Volume/Issue:
Volume 2002: Issue 193
Publisher:
International Monetary Fund
ISBN:
9781451818567
ISSN:
1934-7685
Pages:
143
DOI:
https://doi.org/10.5089/9781451818567.002

Annual GDP Growth Since 1950s

(At factor cost, in percent)

GDP Growth

(At factor cost, in percent)

Contributions to Annual GDP Growth Output Components

(In percentage points)

Contributions to Annual GDP Growth 1/

Expenditure Components (In percentage points)

Rainfall and Agriculture

(In percent)

The External Environment 1/2/

(Percent change)

Exports Growth 1/3/

(In dollar value, percent change)

Domestic Investment Since the 1950s 1/

(In percent of GDP)

Real Private Investment in the 1990s 1/

(In percent of GDP)