Bank Credit: A Leading Indicator
Whoever the culprit the government or the banks, a downturn in bank credit is bad news. And, whatever the causes credit policy or the low level of business confidence, a decline in credit growth is an indicator of the decline in economic activity. In the long run, lower bank credit to the commercial sector would also imply lower investment. In the short run, shortage of credit can cripple economic activity. It can dampen growth in small and medium-sized firms which are the first casualty of credit squeeze because of their dependence on its availability for their working capital requirements. It hurts that portion of consumption demand which is financed by borrowings. These factors make non-food credit a prospective candidate to serve as a leading indicator of activity.
An examination of data on bank credit suggests a contraction in the growth rate of bank lending to the commercial sector during 1996. The growth rate of non-food credit decelerated from 10.9 per cent during the period April to December 1995 to 2.5 per in the corresponding period in 1996. This deceleration in growth of bank credit despite a robust growth in bank deposits reduced the incremental credit/deposit ratio for the non-food sector from 135 per cent for the April to December period in 1995 to about 17 per cent in 1996. Demand for credit drops with poor business prospects.
The business confidence index based on the Business Expectation Survey conducted by National Council for Applied Economic Research (NCAER) also points to the lowered expectations of the corporate sector. Banks have also played it safe by preferring to invest in riskless government securities rather than lending to the private sector. Such behaviour on the part of banks was triggered by the new capital adequacy and risk management norms (aimed at enhancing the safety of banks) which they now have to follow. The paucity of funds for productive purposes can impact the real side sooner or later.
Ongoing research at NCAER involves an analysis of monthly data to construct a set of leading indicators for the Indian economy. The leading indicator approach is based on the view that market-oriented economies experience business cycles. The need for predicting cyclical behaviour of the Indian economy is emerging as the economy increasingly shows signs of upturns and downturns in industrial production. A popular rule of the thumb in developed market economies is to define a recession as a period of two consecutive quarterly declines in real GDP. A somewhat different concept of the business cycle is the growth cycle.
The growth cycle consists of fluctuations around the long-run trend. In other words, it is a trend adjusted business cycle. During the period 1950-51 to 1995-96, the Indian manufacturing sectors output actually declined only twice on an annual basis. The concept of a growth cycle appears more relevant for the Indian economy.
The leading indicator approach is essentially statistical and its forecasts are not based on an economic model. This may be considered a drawback as the analysis lacks an adequate theoretical structure. However, this `atheoretical approach is also its appeal as it avoids the criticism of being restrained by a large number of assumptions.
The first step in the construction of a leading indicator is to identify cycles in the economy. The `state of the economy is measured by the index of industrial production (IIP) for the manufacturing sector. The cyclical component was separated from the trend, seasonal and an irregular component using very simple statistical methods. We then tried to identify the cycles in the index of industrial production. To identify the lead, lag or coincident indicator, the cyclical component of each of the other variables including non food-credit to commercial sector was separated and peaks and troughs identified.
Monthly data for orders, stocks, retail trade, employment, profit etc, which act as leading indicators in industrial countries, are not available and we are restricted to mainly monetary and trade variables. Out of the 50 odd variables (monetary, trade and output) that are being analysed, a few did appear to lead the IIP. Non-food credit was one of these. It showed a lead of 10 months to the IIP peak in February 1991.
In other words, the cyclical component of non-food credit started showing a downturn after April 1990, 10 months before the cyclical component of IIP started declining (in February 1990). Whatever the causes of the fall in this component of bank credit high interest rates, low business confidence, lower orders etc a decline in it can be taken as an indication of decline in economic activity.
In the recent past, the cyclical component of bank credit started showing a downturn after March 1996. However, in an economy in transition the lead-lag relationship is expected to change. The lead period for the last downturn was 10 months. The non-food credit has shown signs of a cyclical downswing. This suggests that IIP manufacturing may soon start showing a cyclical downturn. Secondly, the lead-lag period is usually different for peaks and troughs.
Our analysis shows that the upturn in the cyclical component of the non-food credit were coincident with an upturn in the cyclical component in IIP in April 1994. So, while it starts going down before industrial activity does, it starts going up only with higher activity levels. One reason for this could be that a significant component of credit is for short-term transactions.
Even if this relationship holds it would still have useful information about an upturn in the movement of IIP. The data for non-food credit is available sooner than the data for industrial production. Therefore a consistent upward movement in the cyclical component for two or three quarters would be an indicator of a current or future increase in the growth rate of Industrial output.
The downturn in the cyclical component of non-food credit which started in March 1996, continues unchecked. Unless we observe a reversal of this trend over the next few months, the slower pace of industrial growth is likely to continue.
The authors are economists at NCAER. The views expressed here are their own.