During the 1970s inflation in Israel climbed towards 100% per annum, and by the first half of 1985 exceeded 500%. Now it is about 15%. Empirical research has failed to find a causal link between the money supply and inflation in Israel, and has interpreted the inflation as reflecting instead real phenomena, such as excessive real wages or excessive real devaluation. With inflation set off by such real forces, the demand for money rose, it is argued, and the authorities allowed the supply of money to rise to meet the higher demand. In Discussion Paper No. 290, Research Fellow Michael Beenstock and Michael Ben-Gad challenge this view by integrating real and monetary theories of inflation.

Beenstock and Ben-Gad estimate a two-equation model in which the money stock and the aggregate price level are endogenous variables, using quarterly data from 1970 to 1987. Stability tests reveal that the model did not break down following the Stabilization Programme launched in July 1985. The public is assumed to form its expectations `rationally', according to the estimated model itself, so that inflation will tend to jump up or down prior to expected policy changes. For example, if fiscal policy is expected to be expansive next year in a way that will induce excessive monetary growth, inflation will accelerate now. It can thus appear that monetary growth is caused by inflation, when the truth is the converse. Controlling and testing for the presence of rational expectations is essential to assess the causal effects of monetary policy. The paper's econometric technique is cointegration, a powerful method for estimating the dynamic interdependence between variables which are highly trended, as prices and money are in Israel.
The analysis suggests that fiscal policy did not directly affect the rate of inflation. Instead, inflation depended on the rate of monetary growth and the net excess supply of money. It also depended on expected inflation which, in the model, depends on expected monetary growth. This is a `monetarist' account of inflation, in the sense that only money appears to matter. But the analysis also suggests that monetary growth largely reflected the fiscal deficit, which reached 15% of GDP by 1985. The tendency for the authorities to accommodate inflationary shocks also reinforced underlying inflationary tendencies, both directly and indirectly via rational expectations.
Beenstock and Ben-Gad's policy conclusion is that control of the fiscal deficit is vital if inflation is to be stabilized. If expectations are indeed rational, the public will respond to credible policy pronouncements. This implies that inflation may be brought down by promulgating a `Medium-Term Financial Strategy' in which moderate fiscal deficits and consistent monetary growth targets are announced in advance. The alternative, of reducing monetary growth on an unannounced basis, will delay the reduction of inflation and may depress the economy unnecessarily.
It is implicit in this work that the exchange rate is determined by, rather than a determinant of, inflation. The authors do not test the hypothesis that excessive devaluations were an independent cause of inflation. This must await the integration of the exchange rate as a third endogenous variable into the model.

The Fiscal and Monetary Dynamics of Israeli Inflation: A Cointegrated Analysis, 1970-1987
Michael Beenstock and Michael Ben-Gad

Discussion Paper No. 290, March 1989 (AM)