Seigniorage and the Inflation Tax

We turn next to the question of how and to what extent the Ukrainian government benefited from inflating the karbovanets. John Maynard Keynes wrote in 1923 of the insidious ability of governments to raise revenues by inflating their currencies. In an oft-cited passage, Keynes wrote about a country's ability to "live for a long time . . . by printing paper money;" indeed, it can "live by this means when it can live by no other" (Keynes, 1923). This maneuver involves nothing short of deceiving the public as to the intrinsic value of its money holdings. The government's advantage in this regard derives from its exclusive right to issue new money. The command over societal resources, or the purchasing power obtained by issuing new money, is known as seigniorage. The reduction in value to holders of existing money balances due to the issuance of new money is termed the inflation tax. The sum of the inflation tax and seigniorage is equal to the real (i.e., inflation-adjusted) change in monetary holdings, or real balances. The relationship between them is:

 Real Balances = Seigniorage + Inflation Tax

The Ukrainian government benefited extensively from its inflationary policies, especially in the period from 1992-94. Appendix Table A-1 provides computations of real balances, monetary velocity, seigniorage and the inflation tax, for the five-year period from 4Q:1991 - 4Q:1996. During the early period of Ukraine's independence, roughly from late 1991 through mid-1994, the government derived a significant amount of revenue from its monopoly in supplying base money.[6] In fact, during this period, the government's command over economic resources reached magnificent proportions, in some periods exceeding 100 percent of quarterly GDP. This obviously means that the government not only confiscated vast amounts of the social product (i.e., output), but effectively liquidated a portion of net social assets (i.e., wealth) as well.

Starting in the first quarter of 1995, both seigniorage and the inflation tax declined precipitously. This raises the question of whether such reduction resulted from a conscious policy choice to reduce inflation, or from a rapidly declining ability to engage in inflationary finance. A related question concerns the proportion of the outstanding money supply which constitutes the "inflation tax base." In some sense, the inflation tax "base" represents the government's potential to raise revenues through inflation. We adopt here the conceptually straightforward definition of the inflation tax "base" as real money holdings as a percent of GDP. Given the decline in the inflation tax yield as a percent of GDP after 1994, we are compelled to examine whether Ukraine's appetite for inflation exceeded the maximum potential yield from the inflation tax. It turns out that this stage was reached in the second quarter of 1993, reducing drastically the demand for Ukrainian karbovanets (i.e., experiencing a real decline in the inflation tax base), and pushing the country into hyperinflation by the end of 1993. Let us examine the process by which this occurred, and in so doing, derive a sense of the government's rapidly narrowing set of policy options in 1994-96.

It is a widely-understood feature of the inflation tax that it will rise with inflation up to a certain point, beyond which it actually can decrease (Dornbusch, 1987; Bruno and Fischer, 1990). This occurs when inflation rises so high that the societal demand for money falls. This point is reached, for instance, when people place vast amounts of their wealth into hard currency (i.e., a "dollarization" of the economy occurs); or when cash incomes are rapidly converted into real goods, hoarding critically needed goods for self-consumption, or for trading at a later date (perhaps through barter, in order to avoid cash transactions altogether). To one extent or another, Ukraine showed evidence of such karbovanets-avoiding behavior throughout this period.

The flight from money -- effectively a reduction in demand for money balances -- reduces the inflation tax base, which necessitates greater and greater monetary emissions in order to reap an equivalent amount of inflation tax revenue. A disastrous consequence would occur where inflation continues to rachet upwards without limit. Falling real balances require larger infusions of cash, and higher inflation levels for the government to finance the same level of fiscal deficits. This could precipitate an "upwards spiral," as the inflation tax rises even as the tax base falls, with the effect becoming its own cause in the monetary equivalent of the "dog who chased his own tail." Hyperinflation can be the result; should the public become convinced that the government is incapable of bringing inflation under control, then Roberts (1993) has argued that a complete collapse of the currency can take place instantaneously. Even in circumstances that fall short of hyperinflation, however, at a certain point the revenue-producing potential of the inflation tax base may be exceeded, and incremental amounts of new inflation will produce incrementally less inflation tax revenue (Bruno and Fischer, 1990).

Beyond a certain point, as real balances fall the tradeoff between inflation and inflation tax revenue becomes increasingly unfavorable to the government. The point at which inflation tax revenues turn downwards depends upon the semi-elasticity of real balances to changes in inflation.[7] On the basis of a limited number of quarterly observations, Havrylyshyn, et. al. (1994) took a "first cut" at the semi-elasticity, estimating it at around 3 months, implying a maximum sustainable inflation rate of 33 percent per month at a maximum seigniorage tax yield of 12 percent of GDP. Cheikbossian (1994) estimated the maximum yield at 21 percent of GDP, on the basis of monthly data from February 1992 - March 1994. De Ménil (1996) employed data not available to Havrylyshyn and his collaborators or to Cheikbossian, to estimate the maximum sustainable rate of inflation at the enormous level of 100 percent per month, and the maximum seigniorage yield to be 37 percent.[8] Despite divergencies in these estimates, it is clear that the threshold of sustainability probably was passed in Ukraine in the third quarter of 1993, lasting for a period of at least six months. This tracks closely with the trends in the last two columns of Appendix Table A-1.

Econometric analysis reveals that, in the case of Ukraine, the money demand function assumed a nonlinear form, involving a change of sign from positive to negative at some critical level of inflation. This can be observed in the regression results in Table 10, which tests the effects of changes in real balances, plus changes in real balances expressed as a quadratic term, and the "deposit money banks" deposit interest rate, on the M2 inflation tax, and on enterprise sector money holdings. The interest rate was included as an independent regressor for control purposes, because for most of this period interest rates were subject to an administratively-controlled ceiling set independently of the inflation rate. It is clear that the inflation tax in 1992-96 was a nonlinear function of the demand for real balances. Using these results, we are able to estimate that the turning point in the inflation tax yield was reached at approximately 125 percent of GDP, which was surpassed in the second half of 1993. At the time, this corresponded to an inflation rate of 100-110 percent. This is broadly consistent with the findings referenced above.

Graph 6 depicts the nonlinear relationship between the M2 inflation tax and real broad money balances. It is clear that the inflation tax yield peaked in 1992-93, falling steadily throughout 1993 and into 1994. It remained relatively stable in the 1-7 percent range from mid-1995 to the end of 1996. From Graph 6, we can discern that the inflation tax exceeded its maximum yield after mid-1993, never again providing as fruitful a basis for the government's obvious preference for inflationary finance. We may infer from these results that the government's policy options narrowed considerably in 1995-96, increasingly excluding inflation as an option to finance the fiscal deficit.

Previous | Contents | Next

URI HomepageURI Working Papers Page