Personal Website of R.Kannan
Students Corner - A Decade of Economic
Reforms in India - A Review

Against the backdrop of these monetary policy changes, this Section begins with some stylised facts followed by an assessment of the reforms in terms of the macroeconomic objectives of price stability and credit availability. In addition, the changes in the liquidity management framework are evaluated in terms of the management of capital flows and maintaining orderly conditions in the money market.

The 1990s opened with a severe balance of payments crisis and a high rate of inflation, reflecting the growing internal imbalances of the 1980s. The Reserve Bank responded swiftly with monetary and credit measures aimed at import compression and demand containment. Following the success of the initial stabilisation efforts and the liberalisation of the external sector, the management of capital flows posed a fresh challenge to monetary management. The Reserve Bank absorbed surplus capital flows in its balance sheet in order to maintain the export competitiveness of the economy and at the same time, attempted to sterilise the monetary impact (RBI, 1994).

With the downturn in economic activity beginning 1997-98, the revival of growth emerged as an overriding concern of monetary policy, although with a constant vigil on the inflation front. Besides, in contrast to sustained capital flows in the earlier years, the latter half of the 1990s saw periodic episodes, albeit short-lived, of capital outflows, leading to bouts of volatility in the financial markets, necessitating swift policy action to restore orderly conditions.

The pursuit of price stability was central to the process of financial sector reforms initiated in the 1990s. Although inflation is not targeted per se, monetary policy was formulated with the stated objective of curbing inflationary expectations. The projected inflation rate was gradually scaled down to 4-5 per cent in the last few years from around 7-9 per cent indicated in the early part of the 1990s.

The Indian inflation record over the past five decades can be considered satisfactory as compared with many developing economies. This can be attributed to relatively appropriate monetary management coupled with measures to contain the adverse effects of supply shocks through adequate buffer stocks of foodgrains and other sensitive commodities (Singh, 1982). The guiding principle in respect to inflation management continues to be that, in the medium to long-run, the increase in prices is largely sustained by monetary expansion. In the short-run, however, inflation could be affected by non-monetary, essentially supply-side, factors (RBI, 2001a).

The inflation rate (measured by variations in annual average wholesale price index, WPI) over the past five decades averaged 6.6 per cent per annum. However, decade-wise, the inflation rate accelerated steadily from an annual average of 1.7 per cent during the 1950s to 6.4 per cent during the 1960s and further to 9.0 per cent in the 1970s before easing marginally to 8.0 per cent in the 1980s. On the other hand, the volatility in the inflation rate, as measured by the coefficient of variation, which was fairly high in the 1950s (4.4 per cent), moved in a narrow range of 0.4–1.0 per cent in the subsequent decades. The pickup in inflation from 1970s onwards coincided with a sharp rise in M 3growth, partly reflecting fiscal dominance. At the same time, the inflationary dynamics also reflected supply shocks, such as the two wars (1962 and 1965), hike in crude oil prices (1973-74 and 1979-80) and crop failures. Demand pressures, emanating partly from the widening fiscal imbalances, sustained the inflation in the 1980s.

The first half of the 1990s saw a sharp increase in inflation. The decade began with double-digit inflation attributable to the large-scale monetisation of fiscal deficits. This was intensified by hikes in procurement prices as well as supply-demand imbalances in essential commodities like pulses, oilseeds and edible oils. The severe foreign exchange shortage that the country faced at the time constrained the ability of the authorities to augment the supply position through imports and also contributed to inflationary pressures. Increases in fuel prices and other administered items also fed higher inflation. The inflationary pressures continued during 1993-94 and 1994-95 in the wake of unprecedented capital inflows and the consequent higher monetary expansion. The focus of monetary policy, therefore, shifted to the moderation of money supply with the objective of containing inflation to create an enabling environment for the process of structural adjustment.

The second half of the 1990s was marked by a significant turnaround in the inflation outcome. The inflation rate declined from an average of 11.0 per cent during 1990-95 to 5.3 per cent during the second half of the 1990s. In the first three years of the current decade (2000-01 to 2002-03), the inflation rate continued to be benign, despite a continued volatility in the international crude prices. The actual inflation rates since the second half of the 1990s were, thus, broadly in line with the indicative projections set out in monetary policy statements.

The low inflation rate since the second half of the 1990s, despite an unchanged GFD/GDP ratio, sustained external capital flows and continued double-digit fuel price increases can be attributed to a number of factors:

The role of these factors in containing inflationary pressures also emerges from a formal econometric analysis. A common approach to modelling inflation is the short-run Phillips curve relating inflation to lagged inflation (assuming adaptive expectations) and output gap augmented by supply shocks (Gordon, 1998; Stock and Watson, 1999; RBI, 2002a). The estimated coefficients have the expected signs and are statistically significant.3 The estimates suggest that aggregate demand at one per cent above trend output level raises the inflation rate by around 40 basis points. Similarly, nominal money at one per cent above its trend level increases the inflation rate by around 86 basis points. In brief, the decline in the inflation rate since the second half of the 1990s reflected a confluence of factors. These included better monetary management facilitated by improved fiscal-monetary co-ordination, lower pressures from administered procurement price hikes, weakness of domestic aggregate demand, presence of excess capacities, ample food stocks and foreign exchange reserves. The decline in inflation was also synchronous with the current phase of disinflation characterising the global business cycle .

The objective of the monetary policy all over the world is not only to maintain low inflation but also a stable rate, i.e., to minimise its deviations from the target. Moreover, a consensus is emerging that monetary policy should also minimise deviations of output from its potential level although it is recognised that a trade-off exists between the variability of inflation and the variability of output. Increased credibility in the form of inflation expectations anchored on the inflation target can reduce the variability of both inflation and output (Svensson, 2003). Assessed against these goals, although the inflation rate witnessed a noticeable deceleration in the second half of the 1990s in India, the volatility of the inflation rate did not see a concomitant decline. This could be attributed to an increased volatility in fuel group inflation - the coefficient of variation of the fuel group increased from 16 per cent in the first half of the 1990s to 32 per cent in the period 1995-2002 -reflecting volatile international crude oil prices. The primary articles inflation too was more volatile during the second half of the 1990s. The persistence of volatility could hamper anchoring expectations at a low level. As regards output, its variability – as measured from growth in the index of industrial production, in the absence of high-frequency data on overall GDP growth – was lower in the post-1995 period as compared with the first half of the 1990s. The lower variability of output, however, has to be seen in the context of the industrial production getting caught in a low growth rate trajectory.

Retail inflation, as measured by variations in the consumer price index for industrial workers (CPI-IW or CPI for brevity), by and large, moved in tandem with WPI inflation. Although CPI inflation, at 7.7 per cent, remained lower than WPI inflation during the 1970s, it turned out higher at 9.0 per cent during the 1980s and 9.5 per cent during the 1990s. The divergence between the two inflation indicators widened during the second half of the 1990s with the CPI inflation at 8.6 per cent, almost three percentage points higher.

The wedge between the two inflation indicators largely reflects the larger weight of the food group in the CPI and a differential movement in the inflation rates of various sub-groups. For instance, in the WPI index, the ‘primary articles’ inflation at 6.5 per cent was higher than that of 4.1 per cent recorded by ‘manufactures’ in the second half of the 1990s. In addition, the divergence between the CPI and WPI could also reflect the movements in the prices of services, which are included in the CPI. The divergence between goods and services inflation appears to be a widespread phenomenon. The higher services inflation can be attributed to faster productivity growth in manufacturing (King, 2002).

In view of the divergence between the CPI and the WPI inflation, there may be a need for monitoring a host of other indicators to obtain a more accurate assessment of expectations of inflation for a forward-looking monetary policy. Pethe and Samanta (2001), for example, developed a ‘composite leading indicator’ for inflation and found manufacturing output, money stock, exchange rate, bank credit and raw material prices as possessing predictive capability for future inflation in India. Food prices, exchange rate and real money gap lead inflation expectations by one month each while yield spreads and fuel prices lead by two and six months, respectively (RBI, 2001b).

As the inflation rate in the short-run is often dominated by supply shocks emanating from the agricultural sector or fuel prices, the year-to-year headline (i.e., WPI) inflation rate does not necessarily reflect the underlying inflationary pressures in the economy. This brought forth the construction of ‘core’ measures of inflation to eliminate the noise or transient components of the headline, especially in terms of supply shocks. Two alternatives, exclusion-based and limited influence estimators (trimmed mean), have been examined for India (Samanta, 1999; Mohanty, Rath and Ramaiah, 2000). However, the loss of information content in the construction of core inflation and the relatively greater public acceptability of the headline inflation make the core measures useful only as indicators of the underlying inflationary process rather than as policy targets. Furthermore, in developing countries, a measure of core inflation excluding food items – which can account for more than half of the weight in the index – may not be very meaningful (Jalan, 2002b).

There is, by now, an emerging consensus in the literature, that while money is typically neutral in the long run, monetary policy generates short-run real effects in view of price rigidities and expectation errors. Recent research in the context of developed countries suggests that monetary policy should aim at an inflation rate of 1.5-4.0 per cent (Akerlof et al, 1996; Krugman, 1996). In view of the possibility of a inflation-growth trade-off, a growth-maximising threshold inflation rate has been placed at around 1-3 per cent for industrialised countries and 7-11 per cent for developing countries (Sarel, 1996; Khan and Senhadji, 2001)

The Chakravarty Committee’s observation of four per cent inflation rate as being acceptable can be regarded as the first illustrative benchmark on threshold inflation for the Indian economy. A number of studies have subsequently placed threshold inflation in the range of 4-7 per cent (Rangarajan, 1997; Kannan and Joshi, 1998; Vasudevan, Bhoi and Dhal, 1998; Samantaraya and Prasad, 2001; and RBI, 2002a). The estimate of threshold inflation has, however, a shifting perspective. With structural changes in the economy, prolonged price stability at the global level as well as in India and the credible anchoring of inflationary expectations at a lower level, the threshold inflation could also move downwards.

Finally, while there is little disagreement regarding the long-run benefits of low and stable inflation, a number of studies stress the possible short-run costs in the process of reducing inflation. This is discussed in the received literature in terms of the so-called sacrifice ratio, which is defined as cumulative output losses that an economy needs to endure to reduce average inflation, on a permanent basis, by one percentage point. The concept of sacrifice ratio focuses on reducing inflation on a permanent basis by the monetary authority rather than on a temporary reduction in inflation (say, due to beneficial supply shocks like lower food prices). Furthermore, the sacrifice ratio measures cumulative output losses, which could be spread over a number of periods. The output losses refer to deviations of actual output from its trend. Over time, as inflation expectations stabilise at a lower level, actual output gradually converges to its trend path. The sacrifice ratio measures such output losses in the process of disinflation. One study puts the estimate of the sacrifice ratio at around two for India (Kapur and Patra, 2000). It is necessary to emphasise here that for a proper interpretation of the sacrifice ratio, it is crucial to analyse the dynamics and sources of inflation. In particular, a clear distinction must be made between low inflation driven by benign supply factors and a tight monetary policy.