nt model than that of the non-accelerating inflation rate of unemployment. If zero inflation is optimal, it must be that inflation is costly and causes lower output and more unemployment, which is why zero inflation is preferred to inflation. This theory implies a positive sloped Phillips curve (Graph 2). Increased inflation causes a northeasterly movement along the Phillips curve that results in higher unemployment. As a result, profitability, output, and employment are all reduced.An alternative theory of inflation is the Keynesian Phillips curve (Graph 3). According to this theory, there is a negative long-run relationship between unemployment and inflation. Unemployment can be reduced at the expense of a little more inflation. In this case, the Phillips curve is convex so that as the unemployment rate falls, further reductions in unemployment cost more in terms of inflation. This is significant because the Fed now has a policy choice that involves deciding between inflation and unemployment. The logic of this theory is that inflation helps adjust the labor market. In a multisector economy different sectors are subject to random shocks. Thus, there is a need to adjust prices and wages in these sectors. In sectors where a positive shock occurs, prices and wages increase; in sectors receiving negative shocks, wages and prices fall. However, it is difficult to encourage wages and prices to fall and it may, therefore, be easier to accomplish adjustment by having prices rise in sectors with full employment. In this way, a little inflation helps the economy adjust, thereby lowering unemployment and raising output.The Keynesian Phillips curve has a number of important policy implications. First, inflation now has a positive effect on output. Second, there is nothing optimal about zero inflation. Third, there is no automatic optimal rate of inflation. Instead, the optimal rate depends on society's preference for low unemploy...
