• PAGE ONE
• INDEX
• HOME

Course of real interest rates in US economy

P. R. Brahmananda

THE concept of real rate of interest is a most important theoretical construct in monetary policy. Monetary authorities use it as an instrumental measure to target the mandated or desired inflation rate. This concept was introduced by Irving Fisher in a book Appreciation and Interest. Alfred Marshall used to quote from the above book, contrary to the general impression that Marshall was insular. This is a tribute to Marshall, as he was old and Fisher was young. The former referred to the latter as ``one of the ablest of the younger school of American economists''.

If one can reasonably expect the level of prices to go up steadily at, say, 3 per cent per annum, and if the expected real rate of return on capital which governs the pure interest rate happens to be, say, 3 per cent, the loan contracts would be at a nom inal rate of 6 per cent per annum.

In an economy in which the price level is expected to be constant, the rate of interest on loanable funds would be equal to the real rate of return on capital. In the market for loanable funds, the lender would like to build in the expected inflation rat e in his terms for lending. The nominal interest rate under free market conditions will, therefore, have to make allowance for the expected inflation rate.

In calculating the real rate of return on capital -- say, the rate of profit -- the entrepreneur/ producer takes into account that on his sales, as the expected rate of inflation would be reflected in the prices of his products and the capital values wou ld also be marked up according to the inflation rate in prices. The rate of profit and the real rate of return would be unaffected by the expected rate of inflation.

It is a fundamental axiom in pure theory that the rate of profit is a pure number and is independent of the expected rate of inflation. This theorem is very difficult for lay businessmen and finance ministers to comprehend. They tend to compare the nomin al rate of interest with the rate of profit and go on pleading for lower and lower nominal rates, though in reality sale prices are always adjusted for inflation rates. It is the money rates of interest that have to be adjusted upwards and downwards when expected inflation rates become different and change.

The great Sraffa in his ``Production of Commodities by Means of Commodities,'' laid down the proposition that the rate of interest would be close to the rate of profit. Von Neumann, in the early 1930s, related the real rate of interest to the rate of uni form expansion of an economy under ideal conditions. Actually, the concept of steady growth rate, popularised by Solow, is close to the concept of the actual rate of uniformly expanding economy. Uniformly here means proportionately expanding.

In his recent brilliant lecture in memory of Sukhamoy Chakravarti, His Double Excellency Dr C. Rangarajan, Greenspan's peer in his achievements as a central bank governor, has characterised the real rate as an invisible element; he correctly states that the nominal interest rates minus the expected (note, not current) inflation rates yield the real rates. But there can be different perceptions regarding the expected inflation rates.

Another important point Dr Rangarajan emphasised is that the real rate of interest is generally close to the growth rate of an economy. I would restate the equation for an administered nominal rate as follows: The nominal rate of interest equals the real rate of return on capital plus the measure of the expected rate of inflation; the real rate of return on capital would be equal to the measure of the steady growth rate of an economy. If the latter is `g', and the expected rate of inflation is `px', the administered interest rate in a neutral economy would be `g' + `px'.

Now, Dr Rangarajan also has emphasised that there could be a premium on savings that a government may like to offer as, in a developing economy, the incentive to save should be strengthened. This means there is a third item; hence, the formula will have to be `g' + `px' + `s', where `s' is the premium in terms of a percentage point.

The problem is the fixation of appropriate numbers to correspond to `g', the steady growth rate that can reasonably be expected, and `px', the expected inflation rate. `s' can be a normative number determined by the authorities on the basis of studies on the elasticity of savings to the real rate of interest. In his earlier studies, Dr Rangarajan and Dr K. S. Krishnaswamy, a former Deputy Governor of the Reserve Bank, and several others have shown that savings are somewhat elastic to the real rate of in terest.

Note that `g' can vary, as is happening in the US economy. We can take as a proxy for `g', the trend rate of growth for the past five years in a sequential manner. `px' can be proxied either by the average inflation rate over the past three or five years or by the trend rate of rise in prices over the past five years, again sequential. By deducting from the nominal rate of interest, `px', we get a measure of the real rate of interest. The Graph shows the course of real rates of interest on 10-year bonds and five-year bonds in the US from 1965 to 1999. Also given are the values of the sequential five-year trend growth rates.

It may be noted that the measures of the real rates on five-year and 10-year bonds are both close to each other. They are naturally affected by the cyclical course of the US economy. The mean of the real rate on five-year bonds for the period 1965-1999 i s 3.03 per cent per annum; that of the 10-year bonds is 3.20 per cent per annum. As the Graph indicates, both five-year bond and 10-year bond real rate measures have high range of variations. The standard deviations are respectively 1.49 and 1.46 percent age points. The mean of the trend growth rates is 3.21 per cent per annum. This is almost equal to the mean of the real rate on 10-year bonds.

The hypothesis that over the long period the real rate of interest on long loans is close to the trend rate of growth cannot be wished away. The standard deviation of the sequential trend growth rate values is 1.35 percentage points, whereas that of the 10-year bond is 1.46 percentage points. Hence, the hypothesis that the real rate on long bonds varies in a similar fashion as the trend real growth rate is also empirically supported.

Irving Fisher would have been delighted at the results! We must remember that in pure theory the real rate of interest does not take into account the liquidity properties of bonds. Some types of savings, such as the savings through public provident funds and some items of small savings are perfectly liquid compared to government bonds. Further, when we think in terms of risk averters and risk-avoiders who like to place their funds in banks, the latter have to make provision for a minimum of real costs o f financial intermediation.

The same is true also for mutual funds. In any detailed exercise on administered interest rates in a developing economy, the measure of real costs of intermediation will have to be taken into account. Hicks thought that the minimum in a real rate could b e accounted by this factor. Keynes thought that the minimum is accounted for by compounding the lender's and borrower's risks. In theory, the real rate assumes away these risks. But it cannot assume that the real costs of intermediation and brokerage are zero.

We must note that in the US economy, the authorities do not place a premium on the thrift motive. Their premium is high on sustained productivity growth. Orthodox theory would place equal weights on both and would be reflected in better incentives for sa vings in the US.

In the latest Monetary Policy report, which Mr Alan Greenspan submitted to the US Congress just four days ago, he has presented strong evidence of the declining savings ratios in the US. He has also given evidence how some developing countries (not India , thanks to Mr Sinha), are keeping the band of difference over US rates high enough to attract capital to their countries. An implicit premise in his Report is that one may not continuously depend only on productivity growth to sustain high growth in the US economy. I was delighted to find that the measure of real rate of interest in his Report over the recent period is close to the measure given above in the Graph.

The next article will discuss the course of the differentials in the US between the real rate of earnings on capital in the US non-financial corporate sector and in our measure of the real rate of interest. The differentials are quite high in the US. Tha t is a strength of the US, but also the Achilles' heel of the US economy.