Anti-Inflationary Policies

The distinction between demand-pull and cost-push inflation is very important for regulatory purposes. If inflation is considered cost-push in origin, arising from institutional labor agreements, then the only way to change the rate of inflation is to change the institutional framework within which these agreements are made. If expectations are the cause of inflation, then in the long run those expectations must be changed if inflation is to be curbed. Both Keynesian and monetarist approacues suggest that inflation should be combated by reducing demand, but they disagree on how: Keynesians would reduce demand through fiscal policy (increase taxes / decrease government expenditure), monetarists through monetary policy (restrict the money supply). This having been said, it is in practice quite difficult to determine the cause of inflation.

Worker productivity (and hence potential output) increases with time. If produtivity is increasing by 2% per year, then a 2% money wage increase per year will be consistent with price stability. In other words, price stability results when D(Money Wages)+D(Worker Productivity)= D(Price Level). As a result, the Phillips Curve, which is normally shown as inflation vs. unemployment, can also be shown as change in money wages vs. unemployment.

One weakness of the Phillips Curve is that it is possible to interpret the empirical results as showing either a demand-pull or a cost-push explanation of inflation. As a demand-pull explanation: As unemployment decreases, excess demand for labor increases, and vice versa, in a stable and predictable fashion. The higher the excess demand for labor, the greater the rate of increase of money wages, and vice versa, again in a stable and predictable fashion. As a result, there is a stable and predictable inverse relationship between unemployment and the rate of change of money wages. As a cost-push explanation: At high levels of unemployment, trade unions and employee groups would be less likely to demand money wage increases because the reality of layoffs and unemployment would be more visible. As unemployment decreases, these same groups would become steadily more militant, and at the same time firms would be more willing to allow costs to rise and to pass on these additional costs in the form of prices, since at low unemployment (in a ‘hot’ economy) their sales are less likely to suffer as a result of the price increases.

While empirical evidence confirms the validity of the Phillips Curve in the short run, it is not at all clear if it is valid in the long run. It has been suggested that the trade-off between unemployment and the rate of change of money wages is a transitory phenomenon resulting from the failure of expectations to adjust immediately to price changes. Once expectations adjust to the new price level, according to this theory, the trade-off effect between inflation and unemployment disappears entirely. This analysis has gained credibility in recent years because of the evident breakdown in the historical relationship between the price level and the unemployment rate. The 1970s and early 1980s witnessed ‘stagflation’ – a sustained simultaneous increase in both the rate of inflation and the rate of unemployment. Advocates of the Phillips Curve argued that the curve had simply shifted upwards on an ongoing basis because of expectations and exogenous events.

Monetarists have a more fundamental objection to the Phillips Curve. They argue that labor is concerned with the rate of change of real wages, rather than money wages. If this is so, a tradeoff between unemployment and the rate of change of money wages will only exist when labor expectations are that the rate of inflation should be zero or close to it. As soon as labor expects to see a noteworthy rate of inflation, the short run relationship between unemployment and money wages will shift to the right, with the magnitude of the shift depending on how high the inflation rate is expected to be. In the long run, therefore, there is no tradeoff between unemployment and inflation. By this analysis, policy makers are not able to select combinations of unemployment and inflation rates; in fact, macroeconomic policy is unable to affect the long run rate of employment at all. In the long run, only the rate of inflation can be controlled and therefore the policy maker should choose a zero rate of inflation.

Milton Friedman, the leading monetarist, postulates a ‘natural rate of unemployment’ which is similar to the previously-considered full employment rate of unemployment. Friedman suggests that the actual rate of unemployment can only be reduced below the ‘natural’ rate in the short term by the creation of inflation beyond expectations, and it can only be held below the ‘natural’ rate by continuing to accelerate inflation so that there is always a ‘gap’ between expected inflation and actual inflation. To combat unemployment in the long run, the ‘natural’ rate must be reduced, and this has nothing to do with macroeconomic policy. The ‘natural’ rate of unemployment depends on factors such as the efficiency of information flow in the job market, the rate of structural change in the economy, the costs of undertaking additional worker training, etc.

Monetarists consider that the demand for money is a stable function of a number of variables such as the level of income, the expected rate of return on investments, and the rate of change in prices. This implies that the velocity of circulation of money (V) will be quite stable. Keynesians believe that changes in V may offset and frustrate monetary policy; for example, in a depression, increases in the money supply will find themselves largely falling into precautionary and speculative balances, so that V falls and no overall change is seen to the level of aggregate demand.

Monetarists consider that the demand for and supply of money are largely independent of each other; the supply of money is determined exogenously by the regulatory authorities. It follows that given a stable demand function for money, exogenous changes to the money supply will result in predictable changes to aggregate demand and therefore inflation rates. Modern Keynesians believe that the money supply may be determined, at least in part, endogenously; it may be responsive to economic variables. For example, if the level of money wages rises due to union bargaining, the commercial banks and/or central money authority may expand the money supply to ‘underwrite’ these changes—in effect the money supply has responded to a change in the price level.

Keynesians further suggest that the central monetary authority may be unable to control the money supply effectively. The commercial banks may be able to frustrate the desires of the regulatory authority by finding effective substitutes for money (such as credit card balances), or by finding more efficient ways to use money (income tax deducted at source, for example, reduces the demand for money and therefore increases the ability to use what money exists).

When it all comes down, monetarists maintain that changes in the money supply have been the chief cause of substantial fluctuations in national income/output, causing both major inflations and major recessions. The monetarist view is that monetary policy can have a major impact on the level of real income and employment in the economy. Therefore, the monetarists maintain, monetary policy should follow an automatic rule, allowing an annual change in money supply to match the long-run growth rate of the economy. As a result, whenever the economy is operating at less than its potential, the overly-large money supply will fuel additional output and income; conversely, whenever excess demand exists beyond potential output, the overly-restricted money supply will ‘put on the brakes’ and return the economy to Q. Government attempts to expand or contract the money supply during the business cycle will simply result in heightened oscillations.

The following observations are from empirical evidence and would be acceptable to nearly all economists:

1.       There has never been any major inflation occurring without an accompanying substantial increase in the money supply.

2.       There has never been a substantial increase in the money supply which has not been accompanied by a major inflation.

3.       Given 1 and 2, a high rate of inflation cannot be sustained unless the money supply is expanded.

In examples of major inflation, such as Germany in 1923, America in the 1970s, or the sustained high rates of inflation currently observed in Latin American nations, the monetarist explanation fits empirical data better than Keynesian theory. However, in milder inflations, it can be argued that changes in the money supply are permissive but not causal.

In terms of the circular flow of income, if a large trade union is successful in negotiating a wage increase, then the firms in that industry will charge higher prices to compensate and money national output will rise. However, all output eventually accrues as income to resource owners. As a result, the total money income of all resource owners will rise. Given that all resource owners have some marginal propensity to hold money for precautionary and transactional purposes, a net increase to the total demand for money will occur. If the money supply is held constant by the regulatory authorities, then the amount of money available for speculative purposes will fall. In order to ensure that households and businesses are content to hold this reduced amount of speculative money, interest rates must rise. However, this will discourage investment, resulting in a fall of aggregate demand. In short, unless there is an increase in the money supply, then the process of passing on higher costs in the form of higher prices cannot be sustained indefinitely without a serious adverse effect on aggregate demand and consequent unemployment. While it is true that sustained inflation is impossible in the face of a sustained and determined attempt to restrain the money supply, this attempt will also generate undesirable consequences for the economy.

Keynesians and monetarists both accept that if the money supply does not increase, there will be higher unemployment. They disagree on the amount and duration of unemployment which would be necessary to contain inflation, and in their assessment of the long-term benfits as compared to the short-term costs of a restrictive monetary policy. Keynesians would argue that modern governments have a strong responsibility for ensuring full employment; a departure from full employment would involve heavy economic, social and political costs in the short term which must be weighed more heavily than possible long-term benefits. Monetarists, on the other hand, believe that such output and employment losses are temporary phenomena that the long-term benefit of price stability more than makes up for. Moreover, monetarists believe that any attempt to increase employment beyond its ‘natural’ rate will be self-defeating and simply result in more costly problems in the long run. In fact, monetarists would argue that price stability will reduce unemployment in the long run, since price stability improves the efficiency of markets, including the labor market.