Search IMF Staff Papers

Advanced Search
Journal Description

Editorial Committee

How to Subscribe

Forthcoming Articles

Staff Papers Archive

Copyright Information

Research at the IMF

Free Email Notification

Receive emails when we post new items of interest to you.

Subscribe or Modify your profile

IMF Staff Papers Logo    Last updated: September 2005
Volume 52, Number 2
Comments on "From 'Hindu Growth' to Productivity Surge: The Mystery of the Indian Growth Transition"
T. N. Srinivasan*

This is a disappointing paper. It sees a mystery and fails to convince through analysis why it does. Had the authors been familiar with Indian economic literature, they might not have written it! The literature has not only noted the growth acceleration in the 1980s but has also questioned its sustainability on the grounds of its possibly being debt-led and fueled by employment and real wage expansion in the public sector.1, 2, 3

The authors' basic claim is simply that it is not the systemic reforms of the 1990s but a mysterious shift in the 1980s that changed the growth process of the economy once and for all. A direct test of this claim would be to compare the growth process of the 1980s with that of the 1990s. If the two are no different, it would support the authors' claim. Instead of this direct test, the authors "investigate" and reject a number of other implausible hypotheses, such as that growth acceleration occurred only after the 1991 reforms. No one in India subscribed to this view; it was widely recognized that a relaxation of the severities of the control regime in the 1980s, though it stimulated growth, was not extensive and systemic enough to have accelerated the growth rate of the economy in a sustainable manner.4 The authors do not embed their hypotheses into an overarching model or framework that is firmly rooted in the realities of the Indian economy. These are:5

(1) Growth rate of real GDP in the 1980s at 5.6 percent per year exceeded the rate of 3.6 percent per year in the preceding three decades. It further accelerated to 7.0 percent per year from 1992 to 1997, only to slow down to an average of 5.6 percent from 1997 to 2004.

(2) The fiscal deficit rose from 7.5 percent of GDP in 1980–91, to 9.4 percent in 1990–91, and 10.1 percent in 2002–03. Total external debt more than tripled, from $19 billion in 1980 to $62 billion in 1990, with debt to private creditors rising tenfold (from $2.4 billion to $25 billion) and short-term debt rising fivefold (from $0.9 billion to $4.5 billion).

(3) The gross domestic savings rate marginally increased from 21 to 24 percent of GDP in the 1980s, while the rate of investment climbed steeply from 23 to 27 percent of GDP. From 2003 to 2004, the savings rate rose to 28 percent, while the investment rate fell marginally to 26 percent of GDP. The savings of the public sector steadily declined from 3.4 percent in 1980 to –2 percent of GDP in 2001–02. It is estimated at –0.34 percent in 2003–04. Public investment has also declined from 8.4 percent of GDP in 1980–81 to 5.6 percent in 2003–04, after reaching a peak of 11.2 percent in 1986–87.

(4) Roughly 60 percent of India's labor force is engaged in agriculture, and of that 60 percent, about half are self-employed, and more than 60 percent are employed in agricultural activities. In the 1980s, employment in manufacturing marginally declined. Clearly, with declining employment and better utilization of capacity in manufacturing, it should not surprise anyone if labor productivity is increased.6

(5) The real exchange rate has depreciated since the mid-1980s.

The economy in the 1980s was virtually closed to imports of manufactured consumer goods, and high tariffs and quotas restricted imports of intermediates and capital goods. Until the mid-1980s, there was no real exchange depreciation either. There was excess capacity in the industrial sector (though it is hard to quantify), which in itself was a reflection of the draconian control regime. Liberalizing the use of this capacity at the margin7 as well as imports of intermediates created the potential for output growth. However, the demand for output had to be generated largely from domestic resources, since export growth was limited. Fiscal expansion raised domestic demand. The external resources needed for additional demand for imports of intermediate goods and replacement equipment as capacity utilization increased was financed by borrowing abroad from private creditors.8 This way of generating growth was obviously not sustainable. The authors dismiss this explanation of 1980s growth by derisively calling it "Keynesianism run amok." Their dismissal is "ignorance run amok" of the essential features of the Indian economy of the 1980s.

Had the authors absorbed the facts of Indian growth experience, they would have asked why the rising fiscal deficit since the mid-1990s has not been reflected in any symptom of a macroeconomic crisis, and why growth slowed after 1996–97. They would have found part of the explanation in the slackening of reforms in the second half of the 1980s and the worsening of the climate for private investment. For example, the authors find a lagged effect of public investment on growth in their regressions. Until the 1990s, India's infrastructure industries were all in the public sector, and, as such, public investment, being largely on infrastructure, crowded in private savings and investment, by reducing the infrastructural bottlenecks to output and raising the rate of return for private investment. Clearly, gestation lags in infrastructure investment are long so that public investment in the late 1970s contributed to growth only in the 1980s. As the 1980s wore on, the pressure on public expenditure from rising deficits constrained public investment: from 1986 to 1987, public investment in GDP began a steady and deep decline from 11.2 percent to 5.6 percent of GDP in 2003–04. It is no surprise, therefore, that the authors find growth effects only in the 1980s and not in the 1990s.

The only new idea in the paper is the anecdotal hypothesis that around 1980 there was an "attitudinal shift" toward private business on the part of the national government,9 which triggered a permanent change in the growth process. Without any direct evidence for the shift, the authors infer its existence from the correlation between growth rates of states ruled by the same party as at the Centre and from an alleged "striking shift in the early 1980s in private investment toward corporate sector investment (and away from the household sector), comprising largely unincorporated enterprises." First, there is no evidence that household investment consists largely of investment by unincorporated enterprises. Second, private corporate investment fluctuated between 3.4 and 3.7 percent of GDP in the 1980s with no trend.

The story that the attitudinal shift at the Centre percolated to the states ruled by the same party is unconvincing. In the 1980s, the states did not have any significant policy choices to make relating to manufacturing, trade, or foreign investment. The finding that when they "introduce state-level registered manufacturing shares in the growth regression and allow the coefficients to vary by decade, not only are the shares for the 1980s and 1990s highly positive, but these variables can 'knock out' the pure period dummies" is easily explained. At independence, manufacturing was concentrated in a few states. After independence, the government steered industrial investment to other states through licensing. Because reforms since the 1980s affected manufacturing much more than they did, for example, agriculture, period dummies are knocked out once share of manufacturing is introduced in the regression.10

The distinction between promarket and probusiness orientation is overdrawn and incoherent. The authors consider trade liberalization an archetypal market-oriented policy without considering the implication of the facts that trade liberalization covered only imports of intermediate goods and equipment, and that industrial licensing remained in place. Clearly, such liberalization, by allowing greater flexibility in the use of their licensed capacity, favored incumbent producers but not consumers and potential entrants. By the same token, any policy (for example, reduction of corporate taxes) that raises the profits of incumbents also raises those of potential entrants, if they are allowed to enter. In short, the distinction made by the authors has no economic logic behind it.

The authors claim that, based on their (dubious) cross-country "regressions of income on the deep determinants," India appears "to be far inside the possibility frontier." India's economic history during and after the colonial era demonstrates this more convincingly. At independence on August 15, 1947, India had an incorruptible political leadership committed to development and an efficient, independent civil service. By the mid-1960s, infrastructure had been built up, schools of higher education in engineering and management had been established, and appropriate and inexpensive import substitution in consumer goods had been completed. Had the brief liberalization and opening that followed the 1966 macroeconomic crisis been in place thereafter, and had greater attention been paid to the agriculture and social sectors, I believe India would have replicated and exceeded the performance of the East Asian miracle economies that switched to an oriented strategy at about that time. By sticking to a dysfunctional development strategy for far too long, India's policymakers ensured that the Indian economy performed far below its potential and that India's poor remained poor for an avoidably long time.


Acharya, Shankar, 2004, "India's Growth Prospects: Revisited," Economic and Political Weekly, Vol. 39, No. 41.

Central Statistical Organisation (CSO), 1995, National Accounts Statistics 1995 (Faridabad, India: Government of India Press).

———, 2004, "Press Note: Quarterly Estimates of Gross Domestic Product for the Third Quarter (October–December) of 2003–04." Available via the Internet:

———, 2005, "Press Note: Quick Estimates of National Income, Consumption Expenditure, Saving and Capital Formation, 2003–04." Available via the Internet:

Ghosh, Arun, 1991, "Indian Development Strategy: An Exchange of Views," Economic and Political Weekly, Vol. 26, pp. 2234–36.

Kelkar, V., and R. Kumar, 1990, "Indian Industrial Growth in the Eighties: Emerging Policy Issues," Economic and Political Weekly, Vol. 25, No. 4, pp. 209–22.

Ministry of Finance, 2004, Economic Survey 2003–2004 (New Delhi: Government of India Press).

Ministry of Labor and Employment, 2002, Report of the Second National Commission on Labor (New Delhi: Government of India Press).

Minocha, A. C., 1991, "Indian Development Strategy," Economic and Political Weekly, Vol. 27, No. 9, pp. 430–32.

Panagariya, Arvind, 2004, "Growth and Reforms During the 1980s and 1990s," Economic and Political Weekly (June).

Reserve Bank of India, 2003, Handbook of Statistics on the Indian Economy 2002–03 (Mumbai, India).

Rodrik, Dani, and Arvind Subramanian, 2004, "From 'Hindu Growth' to Productivity Surge: The Mystery of the Indian Growth Transition," paper presented at the International Monetary Fund's Fifth Jacques Polak Annual Research Conference, "Policies, Institutions, and Instability," Washington, November.

Rudra, A., 1991, "Privatisation and Deregulation," Economic and Political Weekly, Vol. 27, No. 9, pp. 2933–36.

Singh, Nirvikar, and T. N. Srinivasan, 2004, "Fiscal Policy in India: Lessons and Priorities," Table 1, from a paper presented at the NIPFP/International Monetary Fund Conference on Fiscal Policy in India, New Delhi, January.

Srinivasan, T. N., 1991a, "Development Strategy," Economic and Political Weekly, Vol. 26, No. 52, p. 2966.

———, 1991b, "Indian Development Strategy: An Exchange of Views," Economic and Political Weekly, Vol. 26, No. 32, pp. 1850–2.

———, 1991c, "Reform of India's Trade and Industrial Policies," Economic and Political Weekly, Vol. 26, No. 37, pp. 2143–5.

Virmani, Arvind, 2004a, "Economic Reforms: Policy and Institutions: Some Lessons from Indian Reforms," ICRIER Working Paper No. 121 (New Delhi: Indian Council for Research on International Economic Relations).

———, 2004b, "India's Economic Growth: From Socialist Rate of Growth to Bharatiya Rate of Growth," ICRIER Working Paper No. 122 (New Delhi: Indian Council for Research on International Economic Relations).

———, 2004c, "Sources of India's Economic Growth: Trends in Total Factor Productivity," ICRIER Working Paper No. 131 (New Delhi: Indian Council for Research on International Economic Relations).


*T. N. Srinivasan is Samuel C. Park, Jr., Professor of Economics at Yale University. This is a revised and expanded version of comments the author made as a discussant at the IMF's Fifth Jacques Polak Annual Research Conference, "Policies, Institutions, and Instability," November 4–5, 2004, at the International Monetary Fund in Washington, D.C.
1See Ghosh (1991), Kelkar and Kumar (1990), Minocha (1991), Rudra (1991), and Srinivasan (1991a, 1991b, and 1991c).
2Per capita emoluments of public sector employees grew by 3.45 times between 1980 and 1981 and 1990 and 1991, while the consumer price index grew by only 2.37 times during the same period (Ministry of Finance, 2004, Table 3.4).
3The really interesting question is not whether the 1980s growth acceleration was mysterious or why it was followed by a crisis in 1991, but why the crisis led to systemic (and thus far unreversed) reforms. As in 1991, India experienced a severe macroeconomic crisis in 1966, went to the IMF and World Bank for assistance, devalued the rupee, and relaxed import restrictions and liberalized the economy. But the liberalization was reversed within 18 months, and this did not happen in 1991.
4For a detailed discussion of the growth and reforms during the 1980s and 1990s, see Panagariya (2004). He sees a greater role than I do for liberalization under way in the 1980s in stimulating growth.
5These data are from the Central Statistical Organisation (1995, 2004, and 2005), Ministry of Labor and Employment (2002), Acharya (2004), Ministry of Finance (2004), and Singh and Srinivasan (2004).
6The authors cite some, but not all, of the total factor productivity (TFP) estimates in the literature, such as those of Virmani (2004a, 2004b, and 2004c). Taken at face value (a dubious proposition) most of them show a significantly higher TFP growth in the 1980s and a slowdown in the 1990s, which is consistent with the hypothesis of unsustainability of 1980s growth acceleration without systemic reforms.
7The so-called "broad-banding" allowed capacity licensed for the production of one good to be used for the production of a closely related good. An increase up to 25 percent of licensed capacity of production was also allowed under certain conditions.
8The current account deficit rose from $2.80 billion in 1980–81 to $9.68 billion in 1990–91. Net commercial borrowings and deposits of nonresident Indians rose from $0.25 billion and $0.22 billion to $2.25 billion and $1.5 billion during the same period (Reserve Bank of India, 2003, Table 135).
9I do not wish to dismiss the fact that Prime Minister Rajiv Gandhi and several young former employees of the World Bank whom he appointed to senior positions in the economic bureaucracy had favorable attitudes toward business, markets, and the value of openness. However, compared with Mrs. Gandhi's long reign, they did not bring about a fundamental change in the regime of economic management.
10A more pertinent interpretation of interstate differences in growth would be based on differences in physical and human (education and health) infrastructure and in, for want of a better term, "governance." Reforms opened up opportunities that were absent. Those in a better position to take advantage of the opportunities get ahead of others not so favored initially. Thus, an initial widening of disparities across states, households, and individuals after reforms is to be expected. The policy-relevant issue is not so much the widening of disparities but whether a process is or could be put in place for the initially disadvantaged to catch up with the advantaged.