Limitations of ISLM Both the simple income-expenditure model and the quantity theory of money engage in drastic aggregation. The quantity theory of money aggregates the economy into only two markets, the market for money balances and the market for all other things. The simple income-expenditure model involves an equally drastic but different aggregation, to a market for goods and services and a market for all other things. Then each, by Walras' Law, ignores the second market and focuses on only one market. These drastic aggregations make sense if the main source of problems in a market economy arise in only one sector and the problems that are evident in other sectors are simply reflections of problems that originate in that one sector. A year after the publication of the General Theory, John Hicks proposed ISLM as a less dramatic aggregation that he saw as a more general case of both the quantity theory and the simple income-expenditure model. Part of its popularity this model, which has reigned as the standard macroeconomic model for half a century, undoubtedly lies in its ability to present macroeconomics in terms of a model with exactly the same structure and mechanics as the model of supply and demand. Though the ISLM model is a fundamental model of macroeconomics, seldom do macroeconomists try to estimate the parameters of the model and use it to predict the future course of GDP. There has been at least one attempt to estimate the importance of fiscal and monetary policy with equations similar to the equation one attains when one solves the model for equilibrium income. However, many economists have argued that such an approach could not capture the subtleties of how the economy works and thus do not give reliable estimates.1 The fact that economists have not used the ISLM model in their attempts to numerically predict the effects of policy suggests that ISLM has weaknesses. The first of these is the question of whether or not ISLM is meant as a long-run or short-run model. If it is meant as a long-run model, then its prediction that equilibrium can exist at any level of output is controversial. ISLM aggregates the economy into three markets: goods, money, and all other. This aggregation makes sense if nothing interesting happens in the "all-other" market, if it simply adjusts passively to changes in the goods and money market. Included in the passively adjusting sector is the resource market, even though many economists point to the labor market as a sector that does not readjust rapidly. ISLM predicts the equilibrium can be at any level because it assumes, as does the simple income-expenditure model, a passive supply. Sellers produce whatever is demanded, and all adjustment to changes in demand are in the form of changes in output and none of the adjustment is in the form of changes in prices. Adjustment cannot be in the form of price changes because the price level does not enter the model. Since changes in prices are the primary way markets adjust in microeconomic theory, the failure of ISLM to say anything about prices is a serious weakness. If meant as a short-run model, the model is severely limited because it does not incorporate the rate of inflation. Inflation creates a difference between real and nominal interest rates. The nominal rate is the visible rate that people pay and receive, and the real interest rate is what is happening in terms of purchasing power. Thus if the rate of interest is 5% and the rate of inflation is 10%, a person who borrows $100 and pays back $105 in a year will pay back less in terms of purchasing power than he borrowed. He pays back 105 dollars, but each dollar can buy 10% less than it did a year earlier. Hence his rate of interest in real terms is actually negative. The real interest rate is computed as the nominal interest rate less the rate of inflation.2 The distinction between real and nominal interest rates is important in ISLM because investment spending should respond to the real interest rate and money demand to the nominal interest rate. To see why investment depends on the real rate, consider a situation of zero inflation in which a person buys a machine that will cost $100 and earn $105 one year later. This purchase will be worthwhile if the interest rate is below 5%. Now suppose that inflation jumps from zero to 10%. The same machine will produce the same output, but because of the increase in prices, the output will be 10% more valuable in terms of the money it brings. Thus the $100 machine will earn $115.05 (which is 10% more than 105) in one year. The investment is now worthwhile at any nominal rate under 15% (which is a real rate under 5%.) Investment will remain constant if the real interest rate does not change; change in nominal rates will not change investment if it does not change the real rate. To keep the demand for money constant (which means that the velocity of circulation remains constant), the nominal interest rate must remain constant. When people hold cash balances for transactions; they are concerned with purchasing power. If all prices double, the amount of money people want to hold will double, but the amount of purchasing power they want will remain constant.3 The interest rate is a cost of holding purchasing power. If the rate of inflation increases, and the rate of interest with it, holding money becomes more expensive and people will want to hold smaller amounts of purchasing power. Thinking of the demand for money in terms of purchasing power lets us ignore price level and is the key to seeing the effects of the rate of interest. It is the nominal rate, not the real rate, that matters. Further, the rate of inflation independently affects the demand for money by changing its desirability as a store of wealth. In cases of very serious inflation, such as the German hyperinflation of 1923, people try to spend money as quickly as possible because it is losing its value. As a result, the velocity of money increases. To some extent estimates of how sensitive money demand is to interest rates may be catching this sensitivity of money demand to inflation because rates of inflation and interest rates move together. One could graph the ISLM model assuming that the vertical axis measured the real rate. Then any time the rate of inflation changed (and thus the nominal rate), one could shift the LM curve. A more rapid inflation would shift the LM curve to the right, for example, reducing real interest rates and increasing income. The problem with this solution, however, is that it leaves the rate of inflation as autonomous, unrelated to what is happening to fiscal and monetary policy. Though there have been times when many economists considered inflation autonomous ("cost-push" theories that were very popular in the 1950s are an example--prices rise because costs rise), most economists believe that macroeconomic policy is by far the most important determinant of rates of inflation. Given these serious weaknesses, why has ISLM remained as a framework for so much macroeconomic thinking? A major reason is that no other simple model gives as much insight. ISLM suggests that economic disturbances can arise in either the money market or the goods market, a conclusion that predates ISLM. Economists want a simple model that concludes this. Also, ISLM can be expanded and made more complex in an effort to overcome its limitations. An recent alternative to ISLM is Aggregate Supply and Demand. 1 The study was done by economists at the Federal Reserve Bank of St. Louis and found that monetary policy was much more powerful than fiscal policy. It is interesting to speculate how positions on this study would have changed if the study had shown that fiscal policy was more powerful than monetary policy. 2 Actually this equation is only approximate. The exact formula is: real rate = (nominal rate - inflation)/(1 + inflation). When the rate of inflation is low, the denominator can be ignored. 3 It is important to realize that this last statement is discussing two situations of equilibrium, one with prices higher than the other. The rate of inflation is assumed to be the same in both situations. That people will hold constant amounts of purchasing power is the central tenant of the quantity theory of money. Copyright Robert Schenk