Who Paid the Inflation Tax?

There is a presumption in the literature that, in transition economies, state ownership of industrial enterprises will thrust the burden of inflationary finance onto households. Havrylyshyn, et. al. (1994) and de Ménil (1996) both argue that households in Ukraine paid all of the inflation tax, because, argues de Ménil, "the seigniorage collected from banks and nonfinancial enterprises is offset by a corresponding reduction in the real profits accruing to the state" (p. 10). Havrylyshyn and his collaborators are more blunt, writing that, "the government sector broadly defined clearly cannot finance its overall deficit by taxing itself!" (p. 360). There is nothing wrong with this view, on a priori grounds. However, throughout 1992-96, increasing numbers of enterprises were privatized; many more of them were subject to "corporatization," or were leased by workers' cooperatives. Further, there was nothing to prevent the state from denuding enterprises of cash, nor partially liquidating enterprise fixed assets. In fact, during 1992-94, the government was criticized repeatedly for attempting to re-value enterprise assets, with the intention of transferring the difference between the new and old appraisals to the state budget.[9] This indicates a willingness on the part of the government to "spend down" certain enterprises' net worth that is largely not reflected in the literature. Segregating M2 into its household and enterprise components (and attributing currency in circulation to households, a not unreasonable assumption given Ukraine's still quite rigid financial sector[10]) is instructive. Although cash holdings accounted for an increasing share of household financial assets from 1992-94, throughout 1993 the household share of broad money (M2) grew from just 30 to only around 50 percent, the balance being held in the form of enterprise deposits in banks.[11]

From Appendix Tables A-2 and A-3, it can be observed that, while households paid most of the inflation tax (as a percent of GDP) in some quarters, in others enterprises accounted for the preponderance of the inflation tax paid. It seems doubtful that household assets were sufficient throughout this period to finance such magnitudes of inflation; the numbers are simply too large. Graphical analysis of the relationship of real monetary holdings to the inflation tax paid, respectively, by enterprises and households reveals two very different processes at work in the two sectors. In Graph 7 it can be seen that the quadratic relationship which characterized the M2 inflation tax also pertains to the enterprise sector; if anything, it is even more pronounced. Graph 8, however, reveals a downwards-sloping curve for households (reading from right-to-left), with a very rapid decline after 1992. The functional form is also different, having a mildly curvilinear shape, but not the "Laffer Curve" shape which one observes in the case of broad money.

Graph 9 depicts quarterly household, enterprise and total seigniorage as a percent of real GDP from 1992-96. As can be seen, household seigniorage remains relatively flat during this five-year period, gradually falling to nearly zero by early 1995. Enterprise seigniorage, on the other hand, tracks much more closely with total M2-calculated seigniorage, spiking in mid-1993, but falling precipitously later that year when inflation consistently exceeded the maximum sustainable level. It appears that, with inflation having thoroughly depleted the household inflation tax base in 1992, the ill effects of inflationary finance fell onto the enterprise sector. Thus did the state's appetite for monetary expansion force it to begin "eating its own children." The government's opportunities to play this game after mid-1995 were severely curtailed, however, as enterprises accumulated ever-larger amounts of foreign currency, a development which also contributed to the accelerating velocity of money in the enterprise sector in 1994-95. (See Graph 10.)

The enterprise and household seigniorage levels appear to contrast even more sharply in Graph 11, which isolates the two in the absence of total M2 seigniorage. Such a pattern flies in the face of the conventional view that households pay the inflation tax in its entirety. Further evidence of a significant divergence between the effects of inflation on households and enterprises can be discerned through an examination of changes in their respective liquidity positions between 1992-96. The theoretical justification for examining sectoral liquidity is provided by the socialist theory of money, a theory to which many Ukrainian officials subscribed in 1992-94.[12]

Expressed succinctly by Peebles (1991), the theory of socialist money is that households in the socialist economy have recourse to the state for the discharge of debts owed to them, which are represented by the paper money issued by the state. Indeed, "the whole logic underlying this monetary system is that the currency has value as long as the state can supply all the goods people want in the correct assortment, at stable prices" (p. 29). Consequently, some economists have measured the money supplies in socialist countries in terms of the relationship of the money stock to the annual flow of retail goods supplied through state stores (Fogaras, 1978; Birman and Clarke, 1985; Winiecki, 1985, 1988; Peebles, 1986, 1991). Portes (1989), on the other hand, disagrees with this approach on theoretical grounds. However, there is nothing conceptually incorrect with the view that financial assets held by households in socialist economies are debts of the government and the government-owned banking system. Further, concerning its applicability in Ukraine, it is hard to imagine that policy makers in this recently socialist country would so quickly abandon their previous conceptions about the role of money in the economic system. Quite to the contrary.[13]

Movements in the ratio of household monetary holdings to retail sales (and the same for enterprises), therefore, may provide important indications of the general drift of government economic policy. Two measures of household liquidity are pertinent, and are labeled DSD Liquidity and HM Liquidity. DSD Liquidity measures household liquidity as a function of demand and savings deposits. Peebles (1991) estimates that, in Soviet times, there was rapid growth in DSD Liquidity, from around .15 of retail sales in 1960, to over .80 by 1989. Consistent with Peebles' estimation of the long-term trend, the present study finds that Ukrainian household DSD Liquidity was in the .95 to 1.00 range in 1989-90, on the eve of the Soviet collapse. (See Table 11.) The year 1991 appears as a watershed period, with household DSD Liquidity falling rapidly thereafter. Table 12 provides a quarterly analysis, based on a rolling four-quarters' worth of retail sales. Since 1991, changes have been dramatic. Household liquidity fell by half in 1992, reaching 1985 levels. It fell by half again by mid-1994, which approximates Peebles' estimate of the 1975 level in the USSR. Not surprisingly, the proportion of household monetary assets held in the form of cash rose during this period, from 18.5 percent at the time of independence to 80 percent in late 1993, where it has hovered ever since, in a narrow band of 82 3 percent.

DSD Liquidity provides only a partial view of household monetary assets, however, since it excludes cash holdings. HM Liquidity provides a much broader picture, incorporating cash as well as demand and savings deposits as a ratio of annual retail sales. According to Peebles, HM Liquidity reached 1.054 in the Soviet Union by 1988, having risen from .182 on an all-union basis in 1960. Peebles' figures are roughly consistent with the present estimate of 1.16 to 1.23 for 1989-90 in Ukraine. Here, too, HM Liquidity falls precipitously after 1991, more than halving by the end of 1993. Graph 12 tracks the decline in both household liquidity measures from 1991-96.

Regarding the trends in enterprise sector liquidity, three measures were developed: enterprise demand deposits (D), demand deposits plus savings deposits (S+D), and demand and savings deposits plus foreign currency deposits (S+D+F). For the lattermost measure, foreign currency deposits have been ascribed entirely to this sector on the basis that enterprises are in a much more favorable position than households to earn large quantities of foreign exchange from export trade. This is not an unreasonable assumption.[14] In any case, the intent was to examine whether the accumulation of foreign currency would have served to cushion adverse changes in enterprise liquidity during this period. It turns out that it was of only transitory benefit. Indeed, as Table 12 shows, enterprise liquidity, which initially rose during 1992 (the period in which household liquidity collapsed), after the hyperinflationary period of late 1993 enterprise liquidity also fell rapidly, reaching approximately household levels by late 1995. (Graph 13 depicts the trend in enterprise liquidity from 1991-96.) Enterprises were therefore late to suffer the ill effects of inflation; their liquidity remained at robust levels until the second half of 1993, but fell after mid-year. Hyperinflation, it seems, spared no one.

The general drift in Ukrainian monetary policy can be discerned from an examination of the liquidity trends depicted in Graph 14. Throughout 1992-93, the government sought to restore the previous balance between incomes and output levels, by consciously transferring liquidity from households to the enterprise sector. This served the perceived policy imperative of wringing inflation out of the retail trade sector while at the same time stemming the fall in output. In fact, the output collapse was viewed by many in the Ukrainian economic establishment as itself the problem, rather than the inevitable adjustment after decades of inefficiency under the former Soviet regime. The policy tools to pursue this objective remained available to the authorities throughout this period, owing to the still rather rigid walls of separation between the official household and enterprise sectors, which were reinforced by Ukraine's credit policies and heavy regulation of retail trade. That Ukraine should squeeze household liquidity so is not surprising in light of the traditional Soviet method of supressing inflation through manipulation of retail prices. Since in a socialist economy retail sales are the main channel for recalling excess currency, the Ukrainian government engaged in a delicate "tightrope act," attempting to strike the appropriate balance between wages and retail goods availability. This task was made all the more difficult by the new inflationary dynamic which had been set in motion in 1992-93.

Previous | Contents | Next

URI HomepageURI Working Papers Page