'44 l
Ma.cropolicies in Transition to a Market
Economy: A Three-Year Perspective
Leszek Balcerowicz and Alan Gelb
Countries in transtion to market economies have had to implement macroeconomic
stabilizaton programs at the same time that they were engaged vi massiue changes of
their poliical isitutons and the systemic frameorks of therr economies. What has
been the interaction of stabilization with economic liberalzaon and dwep institutional
reform in the countries of Eastern Europe, in partcular, the reations among initial con-
ditons, political development, reform strategi, and outcomes? Expec  in Eastern
Europe suggests that when then is a political brakthrough (as n the countries under
review) a radical stablzation-liberalization strategy is probably the kast nsky
approac to re/om and will not constrain otput orstructural reform over the mediun
tenn. Even stabilization that is initially succfid in   ai     inflation  vill later
come under pressure because of soci policies and the strral transitions impelled
by reform. Several factors are identified that affect the credibility of reforms, and lessons
are derived for counties that have stabilized and those that yet face this task.
T       he collapse of parry and state domination of society and the economy left
the countres of Eastern Europe and the former Soviet Union facing a daunt-
ing dual challenge: to move toward competitive market economies while at
the same time maintaining and strengthening newly gained democracies. The eco-
nomic transition in these countries is viewed here as having three elements: macro-
economic stabilization; liberalization of prices, markets, and entry, and deep
institutional change. This article focuses on the problem of achieving macroeco-
nomic stability and sustaining macroeconomic balance through the transition.
Countries in transition must implement stabilization policies in the midst of deep
changes in political institutions and in the systemic frameworks of their economies.
In the context of such large changes outcomes usually ascribed to macroeconomic
policies can strongly influence systemic and political developments. Conversely,
Lesek Balccrowicz is professor of economics at the Warsaw School of Economics. Alan Gelb is division
chief Transition Economics Division, at the World Bank The authors grateflly acknowledge the con-
tributions of staff of the World Bank, the Intenational Monemry Fund, the Bank for International
Setdemcnts, and Planecon and the assistance of Raquel Artecona and Nikolay Gueorguie. Responsibility
for errors and shortcomings is that of the authors alone.
Proeedings of the World Bank Annual Conference on Development Eonomics 1994
01995 The International Bank for Reconstruction and Development / THE vORWl BANIC  21



:p ,4Z  ,, ,. -  -     MUacpolices in Transition to a Market Economy: A Thre-Year Perpedive
macroeconomic policies can have a large impact on outcomes traditionally attrib-
uted to structural or institutional policies. We therefore emphasize the pattern of
interaction between macroeconomic policies, systemic changes, and political devel-
opments-rather than the details of individual programs and outcomes, which are,
in anv event, subject to unusually large measurement problems. The core countries
in the analysis are those in Eastem Europe with longer post-1989 reform experi-
ence-Bulgaria, Czechoslovakia and its successors, Hungary, Poland, and
Romania-but comparisons are made with other countries as appropriate.
"Destroyed Capitalism": Initial Conditions
and Measurement Problems
"Theres no use trying," she said, "one can't believe impossible things."
"I daresay you haven't had much practice," said the Queen ... "Why, some-
times I've believed as much as six impossible things before breakfast."
-Lewis Carroll, Through the Looking Glass
The Common Legacy...
After forty or more years of communism the "destroyed capitalism" of Eastern
Europe and the Soviet Union differed from both the temporarily 'suspended capi-
talism" of postwar Germany and the "distorted capitalism" embraced by Latin
American and other dirigiste economies. In the economies characterized by
destroyed capitalism, institutions reflected a fundamentally different system of orga-
nization and incentives (Kornai 1992). With few exceptions private activity was
severely repressed, and private ownership of assets was limited to savings deposits
and part of the housing stock.
The restricted role of the private sector had important implications for economic
institutions, including the operation of factor markets. Medium-size and large state
enterprises dominated output and employment. Industry was overbuit, especially
machine building and heavy industry; services, particularly trade and distribution,
were underdeveloped and highly constrained. Government played a major financial
role, intermediating between enterprises and between households through subsidies
and transfer programs, and spending accounted for more than half of gross domes-
tic product (GDP). Much of the revenue base was provided by the enterprise sector,
where surpluses were concentrated among relatively few firms (in contrast, in mar-
ket economies state enterprises are usually a fiscal drain).
The features of destroyed capitalism shaped many institutions and professions,
including those of the legal system. Statistical systems were not adapted to dealing
with large numbers of small firms or individual taxpayers, and enterprises had prim-
itive marketing and accounting capabilities. But nowhere was the legacy of
destroyed capitalism more pronounced than in the financial sector. Banking systems
may have been "deep" (as measured by the ratio of balances to output), but finan-



Balcewouicz and Gelb                                                23
cial flows accommodated decisions on the real economy. There was no experience
of indirect, market-based monetary policy. Payments systems were primitive.
Passive, monopolistic state banks lacked the capability to evaluate creditworthiness,
and risk was socialized.
Foreign trade was dominated by a few monopolistic organizations. Autarkic trade
patterns emphasized bilateral exchange between the members of the Council for
Mutual Economic Assistance (CMEA), with increasingly adverse implications for
product quality, as shown, for example, by the low prices of Soviet automobiles on
Western markets (Roberts 1993). Relative prices were distorted; prices of energy
and essential goods and services were heavily subsidized so that cash wages (subject
to centralized norms) could be kept low and investment levels high. Official trade
margins were tightly controlled.
Repressed inflation was widespread, reflecting an absence of b .oad commodity
markets and a "shortage" economy, where demand for goods (and labor) often
exceeded supply at controlled prices. Queuing, rationing, and hoarding were com-
mon responses, although use of these practices differed among countries and over
time. The prevalence of "seller's markets" had profound implications for behavior.
Enterprise managers set production goals to meet the requirements of bureaucratic
bargaining processes (including spurious quality improvements) rather than those of
markets and clients. Parallel prices often were multiples of official prices. Input
stocks were hoarded, and inventories of unfinished investments were large.
...and Its Implications for Measurement
Even in established market economies standard statistical data provide only an
incomplete description of economic reality, but in countries in transition data defi-
ciencies and biases are much more serious (for more extensive discussions, see
UIpton and Sachs 1990; Berg and Sachs 1992; Berg 1993a,b; Bratkowski 1993; and
Balcerowicz 1993a). The statistical system inherited by the transition countries
focuses on the contracting public sector rather than on the expanding private sec-
tor, and while output was overreponed before transition, now there are strong tax
incentives to underreport. Conventional statistics fail to reflect the sharp improve-
ment in the quality and range of goods and in the composition of output stemming
from market-oriented reform, instead applying the same "welfare weights" to pre-
and postreform aggregates. And when prices are freed, conventional statistics over-
state increases in the price level relative to an initial situation with unsatisfied
demand at official prices. This bias distorts all deflated variables, particularly wages,
which appear to fall more, relative to a consumer price index (CPI), than do real
purchasing power and consumption. And conventional statistics do not measure
welfare gains from the elimination of queuing, which may be considerable.
Reported unemployment data also are problematic. They are heavily influenced
by incentives to report and typically are smaller than true labor redundancy, includ-
ing that hidden within enterprises; indeed, the relation between unemployment and
redundancy may vary a great deal among countries.' Cross-border transactions are



24                 Macropolicies n Transition to a Market Economy: A Three-Year Perspeaiue
poorly reported, giving rise to sometimes massive errors in trade data; there is still
no perfect way to compare ruble and hard currency trade, and so to distinguish the
impact of the collapse of CMEA markets from that of the reforms themselves.
Reported fiscal deficits can seriously understate true consolidated government
deficits, particularly where the central bank supports enterprises with cheap credit
(although, at the same time, enterprises may assume some functions of government,
including providing unemployment benefits and social services) for an extended
period. Finally, such basic statistics as stock-building and enterprise markups and
profits are severely distorted by inflation and rapid disinflation, and bank profits can
be spurious because of inadequate loan-loss provisioning.
These data biases are not neutral with respect to types of reform program.
Although many benefits of reform will be more seriously underreported the faster
the economy moves away from the inherited statistical system, the swifter are the
reforms the more rapidly will the inefficiencies of the old economic system be
revealed. These inefficiencies include hidden unemployment, inefficient invest-
ments, excessive input stocks, repressed inflation and the associated costs of queu-
ing, and "purely socialist" output for which demand can be maintained (if at all)
only in a socialist economy. Instituting market-based monetary policies will make
subsidies and losses more tranwarent Invariably, the fiscal budget becomes the
repository for the losses of the previous system.
Country-Specific Factors and the Stabilization Problem
Despite the commonalities of their heritage, the transition economies inherited very
different macroeconomic, structural, and systemic conditions. These conditions
determined the environment in which these countries began reform and thus the rel-
ative difficulty of achieving stabilization. The differences in those conditions reflect
longstanding differences in policies and in the political developments that pre-
ceded-and followed-the breakdown of the one-party state.
Even with conservative macroeconomic policies all socialist countries were
plagued by shortages resulting from supply rigidities and forced substitution in
demand at administratively fixed prices. Resolving such microeconomically induced
shortages is part of the liberalization problem, and a one-time correction of the price
level will be needed to accommodate large changes in relative prices, along with suf-
ficient macroeconomic discipline to preserve stability in the liberalized environment.
Countries also faced macroeconomically induced imbalances, reflected in a mix of
shortages and open inflation (depending on the degree of price control) and, to the
extent that overexpansive policies had been externally financed, high external debt.
Only some of the core countries faced an urgent internal stabilization problem at
the start of reform. In Bulgaria, Poland, and Romania the weakening of the commu-
nist regime was manifested in progressive loss of economic control and increasing
macroeconomic imbalances driven by growing consumption. But Czechoslovakia, in
keeping with a long tradition of macroeconomic conservatism, preserved spending dis-
cipline throughout its political transition. The stabilization problem was also less



Balcerowicz and Gelb                                               25
urgent in Hungary, where both the political and the economic transitions began in a
more stable and liberalized environment and involved far more continuity than the
transitions in the other countries. Like Bulgaria and Poland, however, Hungary had
accumulated large foreign debts. In contrast, Czechoslovakia was essentially debt-free,
and Romania had repaid previously contracted foreign debts by the end of the 1980s.
The magnitude of the initial stabilization problem the core countries faced is sug-
gested by the strength of open and repressed inflation in their economies. Open
inflation was most serious in Poland, and repressed inflationary pressures were
strong in Bulgaria, Poland, and Romania. Czechoslovakia's economy was macro-
economically balanced yet microeconomically distorted, with key relative prices way
out of line and a large foreign exchange premium on a small parallel markeL
The core countries faced structural problems of varied seriousness that were
reflected particularly by dependence on CMEA trade and the size of industry's share
in the economy. Other conditions shaping the reform environment induded the
extent to which the old economic system had been reformed and the strength of the
labor movement after reform: a strong labor movement tended to produce a wage
push, which complicated the stabilization effort. In addition to external debt
Bulgaria inherited perhaps the most difficult structural problem: its high depen-
dence on CMEA trade implied losses of 16 percent of GDP or more with the col-
lapse of that trade. The trade collapse was a large external macroeconomic shock
for all countries, with particularly severe effects for sectors that depended on the
Soviet market. Romania's industry was unusually concentrated, with a hydrocarbon-
based complex dependent on imports from the Soviet Union. Poland too had sen-
ous structural problems, as well as high foreign debt and strong trade unions;
however, following reform, its private agricultural sector outperformed state-domi-
nared systems elsewbere. Other than dependence on CMEA trade the main liability
of Hungary's relatively open and reformed economy was its high level of foreign
debt At the other end of the spectrum, debt-free Czechoslovakia had a weak trade
union movement, and the Czech Republic would later be able to separate away its
most serious problems of industrial structure with its split from Slovakia.
Also vital for the reform environment are political developments during eco-
nomic transition. Hungary has been dose to a model of political stability, with a
smooth transition and the government freely elected in March 1990 still in power
in 1994. Czechoslovalia's far more radical political transition involved elections in
June 1990 and June 1992, both held according to a predetermined timetable. The
country's split in early 1993 was a serious political shock, resulting in fiscal gains to
the Czech Republic and losses to Slovakia, yet in the Czech Republic the same eco-
nomic team has functioned since December 1989.
The other countries have experienced far less political stabifity. Poland formed its
first nonconmmunist government in September 1989 and launched its reform pro-
gram three months later. Following presidential elections in late 1990, a new gov-
ernment was seated in early 1991. Nevertheless, essentially the same basic team was
responsible for the economy until December 1991. After the parliamentary elections
in October, however, the pace of political change visibly quickened. By April 1994



26                 Maaopolices in Transition to a Market Economy: A Three-Year Perspecive
Poland had gone through three governments and five ministers of finance; three of
the ministers resigned. In Bulgaria, where a coalition government initiated an eco-
nomic program in early 1991, the period up to February 1994 saw two parliamen-
tary elections and three governments. Since early 1991 Romania has had as many
governments, and it held parliamentary elections in September 1992. Romania has
been the only country in which the government was forced to resign (in September
1991) under the pressure of widespread industrial unrest and riots.
For a stabilization plus transition program, perhaps even more than for stabiliza-
tion in a nontransition economy, political developments form an important part of
the conditions for implementation. Elections in an early stage of implementation tend
to raise inflationary expectations and reduce the propensity of state enterprises to
adjust because of expectations of a change of direction in economic policy. Frequent
tumrnover of governments and niinisters slows the implementation of structural and
institutional reforms. Thus the political framework for stabilization has been less
favorable in Bulgaria, Poland, and Romania than in Hungary and the former
Czechoslovakia As discussed below, experience shows that the period of "extraordi-
nary politics" that foliows a major political breakthrough and the discrediting of the
old order can create conditions conducive to effecting a determined stabilization. A
vital element of that breakthrough in Eastem Europe (as in the Baltics) was undoubt-
edly the sense that the long period of Soviet domination was at an end.
Economic Policy and Political Breakthrough
Given initial conditions, whether economic policy can shape outcomes in a con-
trolled way depends on political developments. Extreme political instability may
render economic policy uncontrollable, so that the outcome is a product of initial
conditions, external factors, and political chaos. 'With that in mind, consider three
important dimensions of economic policy:
* Launching speed: the interval between a political breakthrough and the
launching of a coherent economic program.
* Phasing: the timing of the launching and implementation of the main com-
ponents of the program.
* The implementation rate for each main component.
The importance of launching speed becomes clear when we analyze economic
transition from the perspective of political economy. The period following a great
political breakthrough, such as those experienced in Eastern Europe and the Baltics,
is characterized by a special mass psychology. In the interval between the discredit-
ing of the old political elite and the coalescing of new interest groups, conditions are
especially favorable for technocrats to assume positions of political responsibility.
There is also a greatly increasfA probability that the population will accept difficult,
normally controversial economic policy measures as necessary sacrifices for the
common good (Balcerowicz 1994b). But the experience of Eastern Europe suggests
that this period of "extraordinary, politics" usually lasts no longer than one to two
years. Then, "normal politics" reemerges, when political groups are much less will-



Bakcerowicz and Gelb                                                   27
ing to accept such measures-or their distributional implications. The timing of dif-
ficult economic measures can therefore be expected to affect their acceptance.
With respect to phasing, consider the three sets of policy reform measures:
macroeconomic stabilization, microeconornic liberalization, and deep institutional
restructuring. Stabilization policy involves fiscal and monetary restraint, exchange
rate management, and possibly wage or other controls. Liberalization policy elimi-
nates legal or bureaucratic restrictions on economic activity, including price con-
trols, quantitative restrictions on foreign trade, limits on foreign exchange
convertibility, rationing and the command mechanism, and barriers to setting up and
developing private firms.2 Liberalization policy primarily supports a rapid shift
toward a market economy and spontaneous growth of the private sector.
Institutional restructuring policy, which involves, for example, privatization of state
enterprises and reform of legal codes and tax administration, creates the fundamen-
tals of a capitalist economy and makes possible well-functioning financial and labor
markets. As discussed below, single policies alone do not determine economic out-
comes. Stabilization outcomes, for example, are the result of liberalization and insti-
tutional restructuring as well as the stabilization policy.
The implementation rate is the rate at which reformed polides are effected rela-
tive to the maximum possible speed for that type of changpe We term rapid policies
as more radical and more gradual polides as less radical. 3 Gradualism may reflect
initial conditions. For example, if a government had earlier undertaken substantial
price liberalization or started reform from a balanced macroeconomy, only limited
furithr decontrol or macroeconomic tightening may be needed. Or the initial eco-
nomic situation may allow radical approaches, but the actual policy may be gradu-
alist by design or because of political drift. Radical reforms, in contrast, are
invariably introduced deliberately.
Clearly, stabilizaton and liberalization policies can be implemented much faster
than most deep institutional changes. A radical strategy thus involves a two-stage tran-
sition to a market economy. In the fir stage the economy is largely 'marketized
thanks to lbealization policy, and stabilized, tianks to stabilization policy, but it
remains market socialist rather than capitalist In the second stage, if stabilization and
liberalization policies are successful, their gains are consolidated and deep institutional
change is completed under macroeconomic stability. Alternative scenarios involve
delaying or interrupting stabilizaton or liberalization, or both, or implementing them
over a longer period during which deep institutional change might be effected.
A comparison of the economic reform policies of the core countries, focusing on
the speed with which reform packages were launched and the scale of changes
achieved, suggests some general observations:
Bulgaria and Poland responded to severe macroeconomic imbalances with
radical programs introduced soon after the political breakthrough (three
months afterward in Poland but a year afterward in Bulgaria). Private sector
Liberalization and institutional change advanced much more slowly in
Bulgaria, however, and by the end of 1993 growing fiscal imbalance and ris-
ing enterprise losses and arrears threatened to derail reform. Romania, inher-



28                 Macropolices in Transition to a Market Economy: A Three-Year Perspectiuc
iting a less severe macroeconomic situation, adopted a less consistent, stop-
go macroeconomic policy but launched a renewed stabilization and liberal-
ization program at the end of 1993.
* With a far smaller stabilization problem but distorted relative prices,
Czechoslovakia implemented radical liberalization safeguarded by tough
macroeconomic policy.
* Hungary responded to a moderately difficult macroeconomic situation with
more gradualist stabilization and liberalization policies. The response largely
reflected initial conditions, induding a more liberalized price system, but it
was also a deliberate choice; the government could have opted for a more
radical program, which the main opposition parties favored. In contrast,
Romania's gradualism did not reflect favorable initial conditions but was dic-
tated by political factors.
* The countries that adopted radical stabilization policies also adopted radical
price liberalization, raised many administered prices, and largely liberalized
foreign trade. In contrast, Romania implemented a much more stepwise lib-
eralization and delayed decisive price and interest rate adjustments.
• Countries undertaking radical reforms also tended to take a radical approach
toward external debt inherited from the old regime. Among the heavily
indebted countries, only Hungary continued to fully service its debt.
Stabilization Outcomes
What were the outcomes of these moderately heterodox stabilization programs for
inflation, internal and external balance, and the restructuring of the economy? None
of the countries implemented an orthodox stabilization program that relied solely
on fiscal and monetary restraint. The severity of initial distortions ruled out broad
reliance on price controls (as in some of the heterodox programs implemented else-
where). But the dominance of the state sector and the absence of owners who would
resist excessive wage pressures called for temporarily maintaining, or even strength-
ening, inherited wage controls. All countries maintained tax-based wage controls, at
ieast through the initial phase of transition. Among the countries facing severe
macroeconomic problems, Poland devalued its currency and pegged the exchange
rate, Bulgaria floated its currency, and Romania floated its currency but intervened
heavily in foreign exchange markets and restricted convertibility. Czechoslovakia
devalued and pegged its exchange rate; Hungary devalued and moved to a crawling
peg. Thus there were two variants of the somewhat heterodox programs: one with
wage controls and one with both wage controls and pegged exchange rates.
Disinflation and Elimination of Shortages
In Bulgaia, Poland, and Romania, where repressed inflation was strongest, prices
and exchange rates might have been expected to overshoot longer-run equilibriums
at the start of reform; in the event, initial price increases greatly exceeded projec-



Blakerowicz and GeIb                                                  29
tions (Bruno 1993). Czechoslovakia's price increase also acceded projections. But
perhaps because inflationary expectations had not become deeply institutionalized,
except in Romania, the initial inflationary burst lasted only a few months.
Nevertheless, only Czechoslovakia-where initial macroeconomic imbalances had
oeen least serious-managed to bring inflation to below 20 percent after a year.
Hungary kept inflation moderate in 1990 and 1991 as it further decontrolled its
already relatively liberalized price system and raised administered prices.
Continuing to exert upward pressure on price levels, however, were the need to
finance fiscal deficits that reemerged after the first year of reform, the introduction
of new indirect taxes (notably valued added taxes), and the adjustment of key
administered prices (rents, energy, public transport) (Orlowski 1993a,b). Although
no country fully adjusted prices in one step, Poland effected especially large adjust-
ments at the beginning; Czechoslovakia and Hungary adjusted prices more gradu-
ally. At the end of 1993 most countries still subsidized residential rents and
household utilities, however.
Romanies inflation profile was an outlier. Despite an apparently tight macroeco-
nomic program and far slower price liberalization than in the other countries, infla-
tion remained above 200 percent after 1990 and reaccelerated after 1991. An initial
phase of decontrol in 1990 stalled and partially reversed in 1991; liberalization was
resumnrd in 1992 but became pervasive only in May 1993 with the lifting of restric-
tions on trade margins. Romania also maintained a number of export restrictions.
Price and trade liberalization and the elimination of restrictions on private busi-
ness, combined with restrictive' macroeconomic policies, were remarkably effectve
in eliminating shortages. Agricultural markets in Poland, for example, swung from
chronic shortage to excess supply in little more than a month as households and
farmers dishoarded stocks and food ceased to be used for animal fodder. Food aid
and export controls added to food surpluses (Kwiecinski and Quaisser 1993).
Business surveys and interviews with managers of industrial state enterprises in sev-
eral countries suggest that a demand barrier rapidly replaced input shortages as the
main constraint on output, at least for established firms. Surveys in Poland show that
most managers saw the new environment as credible and not as a short interlude
before a return to "business as usual." Sharp increases in nominal interest rates in
January 1990 appear to have played an important signaling role, even though real
rates did not immediately become positive. Many state enterprise managers began
to take initiatives toward restructuring, despite high uncertainty about the macro-
econormLic oudook, constraints on corporate governance set by enterprise councils
and trade unions, and limited information about their options (Gelb, Jorgensen, and
Singh 1992; Pinto, Belka, and Krajewski 1993). Whether managers in other coun-
tries took such initiatives is less dear.
Fiscal Deficits, External Financing, and Monetaiy Policy
Reform initially led to sharp reductions in reported fiscal deficits, particularly in
Bulgaria and Poland, where deficits had been large before reform. The largest turn-



30                  Macropolicies in Tramsition to a Market Economy: A Tbree-Year Perpectivw
around, more than 10 percent of GDP, came in Poland, which ran a sizable surplus
in 1990. In 1992, however, large fiscal deficits reemerged in all the countries except
Czechoslovakia, which, along with Hungary, initially maintained a conservative fis-
cal stance. But Hungary's deficit widened sharply in 1992 and 1993, to more than
7 percent of GDP. The increase reflected mainly an unexpectedly large decline in
corporate taxes-a surprising result considering that Hungary had by far the most
reformed tax system. Bulgaria saw its deficit explode to 13 percent of GDP in 1993.
The reemergence of serious fiscal deficits reflected depressed economic activity, as
well as longer-run processes of systemic change and costly social policies.
Current accounts, like fiscal balances, were at first unexpectedly strong. Bulgaria,
Czechoslovakia. Hungary, and Poland all ran surpluses in the early stages of reform,
although Bulgaria and Poland would not have been able to had they been fully ser-
vicing their foreign debts. By 1993, however, for all but the Czech Republic (which
gained a fiscal benefit of some 4 percent of GDP on the separation from Slovakia),
current accounts had swung back into deficit. The deterioration was especially sharp
for Bulgaria and for Hungary, which in 1992 had been able to finance its deficit
domestically because a sharp rise in household financial savings coincided with a
sharp fall in net domestic credit to enterprises.
Romania again followed a different pattern. A demand explosion combined with
a contracting economy to worsen the current account by 14 percent of GDP
between 1989 and 1990. Current account deficits continued through 1993 even
though the budget was in ostensible surplus until 1992; current account deficits and
sharply higher inflation therefore preceded an open fiscal deficit by three years.
Romania's failure to stabilize after 1990 reflected mainly a self-fulfilling monetary
policy failure that helped to undermine the credibility of the reform program and
made subsequent stabilization far more difficult. At the start of reform Romania had
a command, rather than a market-socialist economy, and strong ties between indus-
trial ministries and enterprise managers continued after 1990. Far more than in the
other countries, industrial lobbies, representing heavy and petrochemical industries
and regies autonomes (strategic sectors that were not to be commercialized) exerted
great political power, as did the agricultural lobby. Firms continued to have access
to credit at interest rates far below inflation. Real rates on bank loans averaged
about -58 percent in 1991-93, and National Bank rediscounts were available even
more cheaply until the end of 1993. National Bank losses contributed to a real con-
solidated government deficit of about 21 percent of GDP in 1992, which was
financed in part by monetary expansion.
In Romania, as in the other countries, deflated monetary aggregates and credit to
enterprises contracted at the start of reform. The enterprise sector responded to the
cut in real bank credit by increasing interenterprise borrowing, which soared to 190
percent of bank credit by the end of 1991. Surveys of Romanian enterprises left lit-
de doubt about their payment priorities: wages first, suppliers last (Calvo and
Coricelli 1993). That ranking of priorities suggests that weak credibility-not invol-
untary lending-caused the credit explosion. In contrast, interenterprise credit actu-
ally declined relative to bank credit in Poland in 1990, as the credibility of the



BaCerowia and GeIb                                                  31
hardened budget constraint and the high real cost of borrowing madc enterprises
less inclined to extend such credits. Interentcrprise credit was contained at 20 per-
cent of bank credit in Czechoslovakia and Hungary. At the end of 1991 Romafnia
eliminated the overhang of interentecrprise debts, which had made evaluating cred-
itworthiness impossible, through a global compensation scheme. Canceling out the
net debt required the injection of heavily subsidized credits from the National Bank,
-vhich added to inflationary pressures and undermined the credibility of future sta-
bilization policies. Further compensations have since been needed, suggesting a
'low-discipline arrears equilibrium."4
In addition to continued credit subsidies from the National Bank, Romania's
inflation in 1992 reflected the need to finance an open fiscal deficit of 7 percent of
GDP. This deficit was caused in large part by fiscal subsidies, which increased from
8.6 percent of GDP in 1991 to 11.4 percent in 1992 (subsidies in the other coun-
tries had by then been cut to an average 3.6 percent of GDP). By 1993 fiscal balance
ostensibly had been restored by sharp spending cuts, but off-budget interventions
were still significant, and the inflation tax was being levied on rapidly shrinking leu
balances. Romanians were building up foreign exchange deposits (a third of M2 by
early 1994) and fleeing the leu. Until initiation of its stabilization program in late
1993, Romania's monetary policy, and the erosion of its financial balances, was not
unlike that of Russia after 1992 (Easterly and Vieira Da Cunha 1994). Both coun-
tries formally "deregalated" deposit rates, but their state- and enterprise-dominated
banking systems have temporarily benefited from inflationary expropriation of
householders' deposits.
In contrast, the radical stabilizers that raised deposit rates to positive real levels
soon after the first shock of price liberalization saw sustained remonetization and
conversion of foreign exchange deposits into domestic assets as their currencies
reappreciated from initially very devalued levels. The cumulative dollar return on
Bulgarian deposits for the two years after May 1991 was about 60 percent
Stabilizaion and Restrcturing
What did the financial side of stabilization imply for the restructuring of the real
economy? In the three radical reformers-Bulgaria, Czechoslovakia, and Poland-
real interest rates were negative during the sharp initial spike of price liberalization,
so that the real burdens of domestic debt were suddenly written down (much like
external debt in Bulgaria and Poland). In Hungary, where price liberalization had
been under way far longer and reform involved more gradual adjustment in the
price level, the writedown was more modest and was soon reversed, but external
debt service was sustained, at heavy fiscal cost. Real borrowing rates subsequentLy
averaged about 15 percent in Poland and Hungary and 8 percent in Czechoslovakia.
With modest growth in deflated credit volumes net of enterprise deposits and
large interest rate spreads,s the net resource transfer from banks to enterprises
(defined as the expansion of net credit less the net interest bill due from the enter-
prises to the banks) was negative. Enterprises therefore had to proceed with physi-



32                 Macropolicies in 7iTnsirion to a Markt Eronomy: A Three-Year Pentpecuiw
cal restructuring in a highly resource-constrained financial environment and to rely
on cash flows for new investments. The main question about the banking system,
then, is how it redistributes scarce resources between firms: efficiently-toward
profitable, growing (private sector) firms-or perversely-rolling over loans to loss-
makers and postponing adjustment? This question is taken below.
Net transfers followed a different pattern in Romania. Negative real interest rates
offset the contraction of real credit, particularly in 1991, the year of the global com-
pensation. The net resource transfer from banks to firms was therefore positive but
unsustainable because of the erosion of leu deposits.
Exchange Rate Outcomes
Although the three radical reformers pursued different exchange rate management,
their exchange rate profiles after reform showed strong parallels. The initial deval-
uations in Czechoslovakia and Poland set their newly unified rates at about 4 times
the purchasing power parity (PPP) level; Bulgari.., with low reserves, saw the freely
floating lev briefly devalued to almost 10 times PPp.6 Nevertheless, independent of
whether exchange rates were pegged or floating, real rates converged within a year
toward levels of between 1.8 and 2.5 times PPP, with Czechoslovakia maintaining a
more competitive rate than Poland. The lev reappreciated strongly toward a market
to PPP ratio of 2 until October 1993, when the widening fiscal imbalance and an
increasingly chronic buildup of tax, social security, and interest arrears helped to
provoke a foreign exchange crisis and a series of devaluations that halved the lev's
external value. The expected ratio of market to PPP rates for countries with real per
capita incomes comparable to those of Eastern Europe is about 2, according to 1985
data of the United National International Comparison Programme (Ahmad 1992);
in this sense the exchange rates were converging toward 'normal' levels.
Hungary's real exchange rate had slowly depreciated by 25 percent over the
period 1985-90. More so than the other countries, Hungary pursued a deliberate
exchange-rate-based stabilization after 1990. It encouraged foreign borrowing and
foreign direct investment to offset its heavy debt service burden and at the same time
to build reserves. Hungary's policy of limiting devaluations considerably dampened
inflation, particularly for industrial goods (Solimano and Yuravlivker 1993), but
resulted in steady real appreciation, with the forint reaching a value of 1.6 times PPP
by early 1993.
In real exchange rate movements Romania again displayed a distinctive pattern.
With continued heavy intervention in goods and foreign exchange markets, the
exchange rate moved erratically, and the official rate tended to depreciate relative to
PPP rather than to firm at levels dose to expected longer-run values. By 1993 the
official rate was still more than 3 times PPP Grey and black foreign exchange mar-
kets were active, however, with the National Bank allocating limited foreign
exchange on a pro rata basis on the official auction and with premia between 25 per-
cent and 40 percent. The real effective exchange rate was therefore even higher than
the official rate. Strong depredation of the leu reflected the partial and contradic-



Balcerowicz and Gelb                                                  33
tory nature of Romania's reforms, particularly the subordination of monetary pol-
icy to cushion loss-making firms rather than to reestablish confidence in the cur-
rency. In late 1993 Rornania initiated renewed stabilization efforts, including sharp
increases in interest rates to positive real rates and exchange rate unification. These
efforts led to a substantial decline in inflation, to 5 percent a month by early 1994,
and a considerable firming of the leu.
Was Radical Reform Too Radical?
Monetary and fiscal policies thus have been powerful tools in containing inflation
and stabilizing exchange rates, possibly because inflationary processes were not
deeply institutionalized at the start. But were the radical programs too tough? Not
only were the initial price increases higher than projected, but deflated credit aggre-
gates and measured output levels were lower, and current accounts considerably
stronger, than expected (Bruno 1993). Considering data lags and the uncertainty
accompanying the programs, deviations from projections are not surprising. How
should policies have responded to new information? Should they have been relaxed
to take advantage of space on the external account? Or should they have been main-
tained (or tightened further) to reduce the likelihood that the unexpectedly strong
surge in prices would translate into sustained high inflation?
Such conundrums are endemic in stabilization programs. Both options have risks
that cannot be considered in isolation from the political situation. Policies could
indeed appear too tough to be cre&d--, particularly if there is no major political
breakthrough. But relaxation poses risks of rising inflation, progressive indexation,
loss of credibility, and cosl ier (and politically less acceptable) stabilization in the
future. The transition ecor'oaies generally were not highly indexed; that could
argue for speedy relaxation once the initial price spike had dissipated. But main-
mining macroeconomic stability in a liberalized economy requires fundamental
changes in behavior, particularly for enterprise managers long used to an accom-
modating budget constraint and a passive banking sector. This problem is com-
pounded by the inevitable lag of institutional reforms, including privatization,
behind radical stabilization and liberalization. Because the phase of current account
improvement was short, and because all the countries (with the possible exception
of Czechoslovakia) faced an external balance constraint again by 1992, the risks of
relaxation far outweighed any temporary benefits.
Liberalization, Relative Prices, and Competition
After price and trade liberalization, most relative price ratios appear to have
approached world levels relatively rapidly in Bulgaria, Czechoslovakia, Hungary, and
Poland (Berg 1993a). In some cases policy slowed the transmission of international
prices; in other cases, otably agricultural marketing and input supply, domestic
monopolies in processing and distribution prevented a full pass-through of world
prices. Although the ratio of market to PPP exchange rates is only a rough indicator,



34                 Macrpolicies in Transition to a Market Economy: A ThI-Year Perspecive
it suggests that prices for a wide range of goods should have converged to close to
their long-run levels within about a year of the start of radical reforms; comparative
costs would then play a more important role in determining real exchange rates. The
elimination of temporary "exchange rate protectione exposed finns to the full pres-
sures of domestic and international competition, squeezing industrial enterprises'
margins and creating pressures for selective protection (Schaffer 1993; ACED 1993).
The persistent undervaluation of the leu seems to have limited competitire pres-
sure from abroad in Romadia; although data are inadequate, the pattern of reported
profitability in the regies autonomes and the commercialized sectors suggests that
price controls were more important than foreign competition in determining
markups and profits.
Large formal devaluations did not simply imply a relative price shift in favor of
sectors producing traded goods. Exchange rate unification reduced the relative
prices of many goods (especially high-quality consumer durables and foreign travel)
that had been obtainable only through parallel exchange markets with high premit
Not surprising, consumption of high-quality consumer durables increased sharply at
the same time that the corresponding domestic industries faced a demand barrier.
Thus consuners benefited from unification, while state enterprises that had pur-
chased commodities at official prices and sold at artificially high market prices lost.
Interpreting inflation rates and exchange rate movements in transition is further
complicated by the divergence between consumer and producer prices in most tran-
sition economics. hlt Bulgaria, Hungary, and Poland consumer prices rose more
rapidly than industrial producer prices, while Romania showed, if anything, the
opposite tendenq. In addition to indirect taxes and technical factors that can cause
price indexes to diverge,7 the relative rise in consumer prices appears to reflect two
transitional processes First, relative prices of previously heavily subsidized essential
services are realigned, and second, trade and distibution margins widen from pre-
viously repressed levels as the service sector is privatized and Western marketing pat-
terns take hold. Indeed, the service sector has often led to CPI inflation- In contrast,
service prices lagged behind the CPI in Romnania until May 1993, when margns
were freed. Romania was also the only country in which budgetary subsidies
increased as a share of GDP after reform- These important differences in pricing
policies show up in household budgets, where the share of nonfood essentials rose
sharply in the other countries after 1989 but fell in Romania despite a possibly
deeper decline in living standards (Cornia 1994; UNICEF 1993).
A widening wedge between industrial producer prices and the CPI has several
implications. Part of the CPI inflation in the period following an inidal stabilization
comes from a relative price adjustment in favor of previously suppressed or heavily
subsidized nontraded sectors. This price adjustment increases financial pressures on
industry and agriculture-the traditional tradables sectors-which must contend
with wage demands boosted by rising consumer prices. Increases in the prices of
nontradables offset, in part; the impact of an initial devaluation on competitiveness.
To the extent that price increases reflect cuts in consumer subsidies, higher wages
represent a progressive monetization of the overall consumption bundle; to the



Bakerowcz and Gdb                                                   35
extent that they reflect growing trade and distribution margins, they are part of the
process of shifting resources away from state-dominated industry to build private
wealth in the service sectors. For these reasons, as well as the fiscal consequences of
systemic changes, it is unrealistic to expect inflation rates to fail to levels typical in
OECD countries imunediately after initial stabilizations. They need to be sufficiently
high for a period to accommodate relative price shifs.
Is Stabilization Sustainable? Feedbacks from Systemic Change
The larger process of transition involves many interrelated transitions: from public
ownership toward private, from industry toward services, from an economy domi-
nated by large enterprises to one with many small firms, from seller's markets to
buyer's markets (including the labor market, where open unemployment replaces
hidden unemployment), and from CMEA product standards toward world market
standards (Kornai 1993). Another fundamental change is to transform the financial
sector from a passive player to an active one and to desocialize risk-bearing.
Eastern Europe already has seen huge changes along these lines. The private sec-
tor's recorded shares in the economy have increased especially rapidly in the Czech
Republic and Poland, at first in trade and services but later in other sectors
(Rostowski 1993). Little of this private sector growth is directly attributable to the
privatization of state enterprises. Although private sector development has lagged in
Bulgaria and Romania, thriving private economies have nevertheless arisen in these
countries too. The share of industry in output and employment has fallen sharply in
all the countries in favor of services and, in Romania, agriculturet Layoffs and
breakups are transforming the structure of industial employment; for example,
employment in Hungary's largest industrial firms fell by half between 1989 and
1991 while small firms multiplied.
What do such structral and institutional transitions imply for the sustainability
of macroeconomic stabilization? We consider three areas important for the credibil-
ity and sustainability of stabilization programs: the supply response, fiscal sustain-
ability, and banking reform.
Supply Response
The asymmetric response of private and public sectors to reform argues for a dual-
sector framework for explaiinig the supply response (Aghion and Blanchard 1993;
Berg 1993c; Chadha and Coricelli 1993).8 Because of the composition of its output
the state sector became demand constrained after markets were liberalized, compe-
tition emerged, and CMEA trade collapsed. But the private sector has benefited both
from demand substitution effects and from the recognition and guarantee of private
property rights, which is equivalent to a growth-enhancing supply shock.
Cross-country comparisons suggest two possibly surprising conclusions. First,
radical stabilization is not associated with lower measured output over a period of
two to three years. Indeed, the association is positive in our sample, where the



36                 Macropolcies in Transition to a Ma7kfl Econmy: A Three-Year Perpecive
largest cumulative contraction is in Romania, the least successfil stabilizer, and the
smallest in Poland, the most successful stabilizer. Second, stabilization does not seem
to slow the transition to a private economy. Indeed, some supply-side stimuli to the
private sector have been direcdy related to the severe financial squeeze on the state
sector. In addition to freuing up labor for the private sector, cash-strapped public
enterprises have sold and leased assets to private firms. Where state firms have not
responded to a tightened budget constraint (as in Bulgaria), the private sector has
tended to be crowded out.
These two observations are related. Although Poland shows that state industry
can recover under credible hard budget constraints (Pinto, Belka, and Krajewski
1993), one clear message of the past three years is the importance of private sector
growth for the initial stage of output recovery and job creation-and hence for the
sustainability of stabilization.
Fiscal Sustainability
Before reform, fiscal revenues came mainly from three taxes collected through the
state enterprise sector: the profits tax, the turnover tax, and the payroll tax, which
funded social payments. Together these taxes accounted for almost 80 percent of
tax receipts in Czechoslovakia and Poland and about 50 percent in Hungary.
Because the enterprise sector concentrated surpluses and employment in large
units, governments did not need to develop the capacity to tax large numbers of
individuals and small businesses. But with competition, constrained demand, and
an end to cheap credit, enterprise sectors could no longer concentrate fiscal
resources for governments.9 Meanwhile, widespread private sector underreporting
is encouraged by high profit tax rates and payroll and wage taxes that can cause
gross labor costs to be double net pay. As would be expected, transition has eroded
tax revenues.
Fiscal sustainability is not essentially a revenue problem, however For the core
countries, excluding Romania (which followed a different fiscal profile), the aver-
age ratio of governnent expenditure to GDP was 61 percent in 1989. Of this, 16
percent represented subsidies and 13 percent social transfers, so that redistributive
current spending accounted for half of total government expenditure. Ratios of
government spending to GDP normally are far lower for market economies at com-
parable levels of PPP income: about 21 percent of GDP for revenue and 22 percent
of GDP for expenditure (Crunun, Milanovic, and Walton forthcoming). Recent
research suggests that, allowing for the level of income per capita, an increase of
10 percentage points in the ratio of fiscal expenditures to GDP is associated with a
1 percentage point drop in the growth rate (Easterly and Rebelo 1993). That sug-
gests a high premium on constraining fiscal spending-beyond simply containing
the deficit-during the transition.
Between 1989 and 1992 fiscal expenditure fell relative to GDP only in Bulgaria
and Czechoslovakia. The considerable increase in Romania- can be attributed to ris-
ing producer and consumer subsidies. In the other countries subsidies declined



Balccrowi:z and Gelb                                                37
sharply, so that the average ratio of nonsubsidy spending to GDP rose by S per-
centage points. Social spending accounted for almost all of this increase, rising on
average from 13 percent to 17 percent of GDP between 1989 and 1992. Although
the fiscal costs of registered unemployment have been limited, other social expen-
diture programs are comprehensive, with pensions by far the largest component.
Transition generally has led to a retirement boom; only a small part of the sharp
increases in the ratios of pensioners to population after 1989 has been due to demo-
graphic factors. Some early retirement has substituted for unemployment, but much
has occurred under the many special regimes offering full pensions far earlier than
even the low official retirement ages. As the number of pensioners increases, the
number of contributors to pay-as-you-go state pension schemes shrinks, raising
dependency ratios in Bulgaria to a high of 87 percent by 1992. Ratios of average
pensions to wages also have tended to rise; in Poland they climbed from 43 percent
to 63 percent between 1989 and 1992.
With minimum wages little above poverty lines and a need to drastically restruc-
ture employment, transition countries face a huge challenge in designing affordable
social protection systems that avoid perverse incentives for workers. Moreover,
because of low retirement ages and slow population growth, pensioners and aspir-
ing pensioners weigh heavily in voting-age populations. No country except Estonia
has been able to achieve radical reform of the benefit system, perhaps because
reformers did not see it as an early priority.
Transition also raises the issue of intergenerational equity. The time horizons of
the older generations since the onset of reform are short, and opportunities to accu-
mulate (nonhousing) wealth largely bypass them in favor of younger entrepreneurs,
often well educated, skilled in foreign languages, and able to earn high incomes in
the private sector. The age-specific distribution of wealth arising from transition will
thus differ markedly from the long-rn equilibrium distributions in market
economies, where wealth tends to be concentrated among older generations.
Transition thus increases the need for intergenerational income transfers through the
state for several decades. This need will be greatest in countres that have failed to
stabilize, because of the erosion of household savings by inflation.
Baking Reform and the Enterp     Sector
At the start of stabilization programs banks were typically passive and inexperienced
creditors. Rollovers of loans and capitalization of unpaid interest have led to a dual
financial system in which loans are heavily concentrated among a few; usually less
profitable enterprises. But considerable learning-by-doing in the banling sectors of
the more advanced reformers has resulted in a progressive tightening of financial
constraints on loss-making firms, although loan quality probably is stil detenorat-
ing in the other countries.
In addition to improving banling services, one major objective of financial
reform is to resolve the allocation of the stock of bad debt After stabilizadon this
stock of debt is no longer eroded by inflation but tends to expand because of posi-



38                  Macmpoliaes in Transition to a Market Economy: A Thee-Yar Parpctive
tive real interest rates. A second major objective is to ensure that new credits flow
to solvent, growing firms, to avoid creating another serious loan portfolio problem.
The two objectives are dosely related. Incentives for prudent commercial lending
cannot be established in the absence of properly capitalized banks: solving the flow
problem requires addressing the stock problem. Conversely, recapitalization, if
repeated as part of the process of unearthing and provisioning against bad debt, threat-
ens to undermine the credibility of the hard budget constraint. Some argue for a third
objective: to give banks an active role in corporate governance of the enterprise sec-
tor-not because they are an ideal choice but for want of other plausible outside strate-
gic owners and because they have the most knowledge of the enterprises-and,
through this arrangement, an active role in resolving the bad debt problem.
Countries are addressing this complex set of interrelated problems through vari-
ous strategies, usually including a mix of debt transfers to special institutions, debt-
equity swaps, and bank recapitalization and provisioning financed in part by
government and in part by large spreads.10 The ultimate fiscal costs of resolving prob-
lem loans still are not known, but they will be considerable for all countries (Dittus
forthcoming; Gomulka 1993). In Hungary, for example, the costs of provisioning
against bad loans accounted for 4 percentage points of the 10 percentage point
spread between lending and borrowing rates in 1992, and the effect of new provi-
sioning rules on banks' profitability cut profit tax revenues by 2 percent of GDP.
Hardening budget constraints and moving to sound finance is not likely to be
achieved quicldy. The institutional change in the banking sector and other financial
markets needed to fully privatize commercial risk will take time. Risk will remain
high, in part because of the preponderance of new firms (including those split off
from old firms) without a track record in the new econoniic environment. In addi-
tion, in most countries in transition legal processes still favor debtors over creditors.
Necessary yet abrupt measures to fully decentralize risk could encourage banks to
hold excess liquidity and lead to the withdrawal of credit from many potentially
viable clients. As a practical matter, government budgets therefore probably will
have to bear some of the costs of problem loans for some time, if spreads are not to
be so large as to inhibit good firms and furither weaken bank portfolios.
Ten Lessons
From the perspective of stabilization, postsoialist countries now fall into two
groups. Except for Romania, the core countries, together with Albania, Slovenia,
and two of the Baltic states, have had some success in containing inflation following
price liberalization and now must consolidate macroeconomic stability while deep-
ening institutional reforms. Most of the others, many with new currencies, have not
yet succeeded in holding inflation at even moderate levels. The experience of the
more advanced reformers offers lessons for both groups of countries.
A general lesson is that, in addition to economic policy, what matters in deter-
mining outcomes are initial conditions, political developments, and external factors.
On balance, the core countries that faced the most serious macroeconomic problems



Bakericz and Gelb                                                     39
may also have inherited the most difficult structural conditions for stabilization, as
well as high foreign debt and powerful trade unions. It is not dear that countries
that had earlier undergone a phase of market socialism were at an advantage. Their
institutions may have been somewhat better adapted to market conditions, but
decentralization of management to enterprises created difficult political problems
for further reform in Poland, and an apparently less urgent need for reform may
have encouraged policy drift in Hungary.
1. Radical is less risky. The countries of Eastern Europe and the former Soviet
Union that have managed to contain inflation to low or moderate levels have all
experienced major political breakthroughs combining democratization and renewed
national independence. The record suggests that when there is both high initial
macroeconomic imbalance and a major political breakthrough of this kind, a radical
approach involving forcefiul stabilization measures and rapid liberalization is almost
surely the least risky option. Initial stabilization has been most successfill when rad-
ical programs have been launched during an early period of extraordinary politics
following the breakthrough. The reversion to normal politics after an initial hiatus
complicates fiuther stabilization by raising inflationary expectations, slowing the
implementation of structural and institutional reforms, and reducing the propensity
of state enterprises to adjust-Bulgaria is an example. Normal politics is especially
problematic if it involves extreme political fragmentation. Successful stabilization
requires fundamental behavioral and institutional changes in an economy still dom-
inated by state enterprises. One function of radical reform is to signal the need for
such change through a fundamental shift in regime. If that dcange is not achieved,
chronic financial indisopline and mounting arrears can rapidly erode a reform pro-
gram and destroy the credibility of a financially disciplined equilibrium.
2. There is no simple link between type of reform and politial stabiity. Contrary
to popular opinion, the degree of political instability does not appear to have a sim-
ple relationship to the type of economic program: consider, for example, the Czech
Republic and Hungary (both stable but with different political transitions and very
different economic programs) or Poland and Romania (both unstable, but one sus-
taining a radical program and the other not).
3. Don't fine-tune at the start. It is better to err on the tight side than to try to
fine-tune policies in the early stages to take advantage of the temporary balance of
payments relief that appears as inflation is brought under control. Failure to stabi-
lize initially will lead to a need for further attempts, but from weakened economic
and political bases and with lower credibility. In any case, the inevitable uncertainty
makes successful fine-tuning impossible.
4. Monetary and fiscal policy can stabilize in transition. Monetary and fiscal poli-
cies can be powerful instruments for containing inflation to moderate levels (very
low inflation may be difficult to reconcile with the need for continued price realign-
ment) and for stabilizing exchange rates, possibly because inflationary processes are
not deeply institutionalized at the start. Conversely, stabilization can be elusive after
initial price liberalization if financial policies fail to support the holding of domes-
tic financial assets. The main failure in Romania (and in Russia and other countries)



40                  Macropolides m Tansition to a Aiket Econony: A Three-Year Perspectiv
has been monetary and interest rate policy; lack of credibility in this area may
encourage arrears. Minimum interest rate targets may be needed for an extended
period to protect uncompetitive, state-dominated banks from decapitalizing house-
hold balances.
5. Wage controls are vitaL All the countries relied on wage controls, which
become important as privatization inevitably lags behind stabilization and liberal-
ization policy. The enterprise sector needs to preserve retained earnings to finance
its restructuring and is likely to have to effect a net resource transfer to the bank-
ing system as stabilization is consolidated. Some countries, such as Romania and
Russia, have temporarily continued positive transfers to their enterprises, but these
transfers become unsustainable as financial balances erode. And because the trans-
fers are related to political pull rather than to performance, they also are likely to
be inefficient.
6. Liberalization reinforces stabilzation. In theory a government could achieve
stabilization while retainig extensive state control of the economy. But in practice
there has been a dose relation between stabilization and liberalization exr-pt on the
wage front. Policymakers who valued stabilization also valued liberalization. Cutting
fiscal subsidies and avoiding large central bank losses due to subsidized credits (a
major weakmess in Romania and many countries of the former Soviet Union) have
been important elements of macroeconomic adjustment. Most fundamental, stabi-
lzing a heavily controlled economy that retains extensive price controls and strong
insttutional links between government and enterprises is very difficult even for a
strong state because the enterprises can blame these controls for their losses. In addi-
tion, the enterprises always have better information than the government and can
take advantage of their lack of autonomy to claim subsidies. Romania shows how
problematic this option is.
7. The importance of exchange rate pegs depends on the nature of infationary
expecations. Some of the tansition countries pegged their exchange rates to pro-
vide another nominal anchor. The experience of Bulgaria, as well as that of Latvia
and Slovenia, suggests that the importance of fixing the exchange rate is a more
open question than the need for wage controls. It may hinge on whether deeply
embedded inflationary expecttions concern price increases (as in Latin America) or
mosdy shortages.
8. Radical stabilization and liberalization polxiy encourages recovery and transi-
tion to a private economy. Country experience suggests no adverse medium-run
tradeoffs between stabilization (with its accompanying liberalization) and either out-
put or the speed of transition to a private economy. Failure to stabilize wastes valu-
able time and will delay recovery. Measures to promote private sector development,
including privatizing small assets and providing access to commercial real estate,
should be essential components of a stabilization package because of their impor-
tance for the supply response.
9. After nittial stabilzation, credbk, sustainable reform  nill require a Strong
growth response by the private sector; fiscal reform, especially on the spending side,
and probably some external spport; and resolution of bad debts without redig



Balexrudz axd Gelb                                                      41
the crvedibility of budget constrains. The inevitable lag of institutional reform behind
stabilization and liberalization suggests that an extended phase of fiscal pressure is
to be expected in a radical reform program. In theory a fiscal balance condition can
set an upper limit to the speed of transition from a public to a private economy.
Although this tradeoff is appealing, in practice the aim of preserving tax receipts is
a poor argument for deliberately slowing this transition because revenues are in any
case not sustainable. It is counterproductive to try to maintain fiscal revenues by
supporting public enterprises with quasi-fiscal subsidies through the banking system:
rampant fiscal deficits will shatter the credibility of reform. On the other hand, very
high (actual or anticipated) taxation will slow the private economy, which will need
to finance rapid growth mostly out of retained earnings. Slow private sector growth
also will undermine reform.
The main fiscal reforms must, therefore, center on spending. Even after the elim-
ination of subsidies, ratios of expenditure to GDP are far above those usual in mid-
dle-income countries, largely because of social transfers. Severe resource c onstraints
will prevent economies from rapidly growing out of these ratios, particu&arly if the
budget also has to bear the costs of previously hidden inefficiencies. Spending analy-
ses, including detailed studies of the incidence of social benefits, are needed to show
where adjustments will be less painfuL But it is difficult to cut expenditures abruptly
(especially after the end of "extraordinary politics") when trying to preserve a
democratic consensus in favor of reform. Unlike in nontransition stabilizing
economies, fiscal deficts in transition economies must be seen in a medium-term
perspective, with greater emphasis placed on reducing government spending and
improving its composition. In the interim external finance will be needed to allow
larger fiscal deficits than conventionally acceptable to be financed without excessive
monetary expansion. For the same fiscal reasons, reductions in debt service obliga-
tions are critical to the success of radical reform in heavily indebted countries.
Especially because financial systems can transfer few resources to borrowers in
the stabilization phase, measures to limit credit rollovers to loss-mnaldng state firms
and to encourage lending to private firms need to be put in place relatively quicldy,
even if banks cannot immediately assume a leading role in initiating bankruptcies. It
is also vital to "activate" at least part of the banking system as fast as possible, to
support the reorganization of the enterprise sector without, however, creating
expectations of bailouts that weaken the credibility of reform. How best to resolve
the bad debt problem depends on the country's conditions; the Czech Republic,
Hungary, and Poland offer a range of experience suggesting that progress is possi-
ble. Across-the-board debt relief is problematic, however, and a simple one-off res-
olution of the bad debt problem is probably unrealistic.
10. Failre to stabilize at first does not awgze against continuing instional
restrucnt g and liberalization. Consider countries still grappling with major stabi-
lization problems in the absence of a major political breakthrough (or having failed
to stabilize initially in the aftermath of a breakthrough). How should these countries
proceed? Clearly, their task is difficult because of continued economic weakening
and the likelihood that indexation mechanisms will be strengthened. One lesson is



42                      Maropolicies in Transition to a Market Economy: A Three-Year Perspeaive
to exploit the credibility imparted by any major political discontinuity to implement
a detennined stabilization program. Even while not yet stabilized, however, these
countries should proceed with liberalization and institutional reforms as rapidly as
they can. These reforms will not realize their full potential until macroeconomic sta-
bility is achieved. But they can improve the underpinnings for future stabilization
and may strengthen a constituency in support of a future stabilization effort.
Notes
1. For example, in 1992 a third of Hungary's registcred unemployed workers (12 percent of the labor
force) may have been worling, but in Russia, where registered unemploymcnt is oniy 1 percent, actual
unemployment has been estmated at 10 percenL
2. For a more detailed breakdown of policies, see Fischer and Gelb (1991).
3. The term rdical is preferred to the more emotive big bang because the second term implies a possi-
bility of effecting all changes immediately-, see Balcerowicz (1994a,b).
4. The countries that have needed to resort to global compensation (RoMamnia and Russia) have also
achieved less policy credibility for stabilization. Bulgaria's experience in 1993 supports the view that
decreased credibility of reform (m Bulgia's case, the stalling of institutional restruacuring after initial
stabilization and liberalization) leads to explosive growth of arrears of all types (ACED 1993); if suffi-
dently widespread, these arrears create a low-discipline equilibrium (see Rostowski 1992).
S. Large spreads in transition countries reflect four factors: limited competition, cosdy reserve require-
ments, the banking systcm's need to generatc resources for modenization, and the need to provision
against loan losses and recapitalize the banks.
6. Becausc historical indicators of real exchange rates are not a useful guide to long-run equilibrium
market rates, a rato of nominal exchange rates to cstimated purchasing power parity (PPP) rates is used
to suggest the level of the rcal exchange rate
7. The consumer price index (ClI) is a Laspeyres index, and weighting may bc inauate; the producer
price index (PPr) is a Pasche index, and coverage of industralc products may be limitedL The continued
divcrgence between CPI and PPI, especially in Bulgaria, suggests that mcasurement anomalics are seious
8. Extensive discuion of the relative importance in explaining output of measurement errors, demand
shocks, supply shocks (including those effected through credit contractions), trade disruption, decline in
"information capital," and increased uncertainty cue to the change in cconomic systnm is beyond the
scope of this article (but see, for cxample, Bleer and others 1993).
9. This effect has been somewhat offset in the carly stas of reform by cuts in subsidies and deficient
inflation accounting, whih susaincd taxes by decapitalizing enterprises (ichaffer 1993; Barbone and
Marchetti forthcoming). For an ectcdsive treatuent of fiscal issues in transition, see Tlizi (1993).
10. It is not yet dclar how well the strateies have worked; for reviews of the stratcgies that differcnt
countries have adopted, see Pleskovic (1994); Transition (1994); BIS (1993); Dittus (forthcoming).
References
ACED (Agenc for Economic Coordination and Developmcnt). 1993. Bulgarian Economy in 1993.
Annual Report Sofia.
Aghion, Phillippe, and Olivicr Blanchard. 1993. 0O the Speed of Tansition in Central Europ-e.
Working Paper 6. European Bank for Reconstruction and Development, London.
Ahmad, Sultan. 1992. "Regrcssion Estimates of per Capita GDP Based on Purchasing Power Parities."
Policy Research Working Paper 956. World Bank, International Economics Department, Socioeconomic
Data Division, Washington, D.C.
Balcerowica, Lcszek 1993. 'Common Fallacies in the Debate on the Economic Tansition in Central and
Eastern Europe. European Bank for Reconstruction and Development, London.
- . 1994a. "Econoniic Tansition in Central and Eastern Europe: Comparisons and Lessons.
Austlian Economic Reiew (first quarter): 47-S8.
. 1994b. -Poland." InJohn Williamson, ed., The PoRa d Ecmy of Policy RefnraL Washington,
D.C.: Institute of International Economics.



Balcerowicz and Gelb                                                               43
Barbone, Luca, and Domenico Marchetti Jr. Forthcoming. "Transition and the Fiscal Crisis: Crisis in
Central Europe.' Economics of Transition.
Bcrg, Andrewc 1993a. "Does Macroeconomic Reform Cause Structural Adjustment? Lessons from
Poland." International Monetary Fund, Washington, D.C.
1993b. "Measurement and Mismeasurement of Econormic Activity during Transition to the
Market." In Mario L Blcjer, Guillermo A. Calvo, Fabrizio Coricelli, and Alan H. Gelb, eds., Eastern
Europe in Transition: From Recession to Growth? World Bank Discussion Paper 196. Washington, D.C.
1993c. "Supply and Demand Factors in the Output Decline in East and Ccntral Europt? Paper
presented at the International Confecrcnc on Output Decline in Eastern Europe, November, Vienna,
Austria.
Berg, Andrew, and Jeffrey Sachs. 1992. "Structural Adjustment and International Trade in Eastern
Europe: The Case of Poland." Economic Policy (April): 117-73.
BIS (Bank for International Settlcments). 1993. Sixiy-ThirdAimual Report Basel
Blejer, Mario I., Guiliermo A. Calvo, Fabrizio Coricelli, and Alan H- Gelb, edts. 1993. Eaern Europe in
Transitiomn From Recession to Growth? Proceedings of a Conference on the Macroeconomic Aspects of
Adjutment, Cosponsored by the Intemnational Monetary Fund and The World Bank World Bank
Discussion Paper 196. Washington, D.C.
Bratkowski, Andrzej. 1993. "The Shock of Transformation or the Transformation of the Shock? The Big
Bang in Poland and Official Statistics." Communwist Econonties and Economic Transfiormation S (1):
5-28.
Bruno, MichaeL 1993. "Stabilization and Reform in Eastern Europe: Prcliminary Evaluation? In Mario
L Bleler, Guillermo A. Calvo, Fabrizio Coricelli, and Alan H. Gelb, eds., Eastme Europe in Transition:
From Recession to Growth? World Bank Discussion Paper 196. Washington, D.C.
Calvo, Guillermo A., and Fabrizio Coricilli. 1993. "Inter-Enterprise Arrears in Economics in Transition."
Paper presented at the International Conference on Output Decline in Eastern Europe, November,
Vi'enna, Austria.
Chadha, Bankim, and Fabrizio CoricetiL 1993. "Fiscal Constraints and the Speed of Transition.?
International Monetary Fund and World Bank, Washington, D.C.
Cornia, Giovanni A. 1994. "Povcrt}y Food Consumption, and Nutrition during the Transition to the
Mlarket Economy in Eastern Europe." Americnr Economic Review 84 (2): 297-302.
Dittas, Pcte  1994. "Corporate Governance in Central Europe: The Role of Banks." BIS Economic
Papers 42. Basel, Switzerland.
Eastcrly, William, and Sergio Rebelo. 1993. "Fiscal Policy and Economic Growth: An Empirical Investi-
gation.? World Bank, Transition and Maao-Adjusutent Division, Washington, D.C.
Easterly, Wlliam, and Paulo Vieira Da Cunha. 1994. "Financing the Storm: Macroeconomic Crisis in
Russia, 1992-93." Poliqc Research Working Paper 1240. World Bank, Macraoeconomics and Growth
Division and Europe and Central Asia Country Operations Division IH, Washington, D.C.
Fischer, Stanley, and Alan Gedb. 1991. "The Process of Socialist Economic Transformation." Journl of
EFconomic Perspecives 5 (1): 91-405.
Gelb, Alan, Erika Jorgensen, and Indeijit Singh. 1992 "Lifc after the Polish 'Big Bang': Episodes of Pre-
Privatization Enterprise Behavior" In Aryc Hill man and Branko Milanovic, eds, The Dansition tium
Socialm  in Eastern Exrope: Domest Rest ructuzg and Foreign Trade Washington, D.C.: World Bank.
Gomulka, Stanislaw. 1993. "The Financial Situation of Polish Entcrprises (1992-93) and Its Impact on
Monetary and Fiscal Policies." Eastern Europe Research Paper Series 28. World Bank, Transition and
Macro-Adjustment Division, Washington, D.C
Kornai, Jinos. 1992. The Socialist System. The Political Economy of Commnism. Princeton, N.J.:
Princeton University Press.
. 1993. "Transforrmational Recession: A General Phenomenon Examined through the Example of
Hungary's Development? Paper presented at the Frangois Permoux Lecure, College de France, June, Paris
Krumm, Kathy, Branko Milanovic, and Michael Walton. Forthcoming. "Transfer and the Transition from
Socialism: Is a Radical Alternative Necessary?" World Bank, Europe and Central Asia Regional Office
and Transition Economics Division, Washington, D.C.
Kwiecinski, Andrzej, and Wolfgang Quaisser. 1993. "Agricultural Prices and Subsidies in the
Transformation Proccss of the Polish Economy.? Economsc System 17 (2): 12-54.



44                     Maaropo Sides in Transition to a Market Economy: A Thie-Year Perspective
Lipton, David, and Jeffrey Sachs. 1990. 'Creating a Market Economy in Eastern Europe: The Case of
Poland." Brookings Papers on Economic Activizy 1. Washington, D.C: The Brookings Institution.
Orlowski, Lucian T. 1993a. "Destabilizing Factors in the Economic Transition in Central Europe." Paper
presented at the Conference of the European Association for Evolutionary Political Economy: The
Economy of the Future Ecology, October 28-30, University of Barcclona.
. 1993b. "Problems of Corrective Inflation in the Transformation from Central Planning to a
Market Economy. The Experience of Poland." Sacred Heart University and Kiet Insttute of World
Economics, Kid, Germany.
Pinto, Brian, Marek Belka, and Stefan Krajewski. 1993. "Transforming State Enterprises in Poland.
Evidence on Adjustment by Manufacturing Firms." Brookings Papers on Economic Activity.
Washington, D.C.: The Brookings Institution.
Pleskovic, Boris. 1994. 'Financial Policies in Socialist Countries in Transition." Poiicy Research Working
Paper 1242. World Bank, Research Advisory Staff, Washington, D.C.
Roberts, Bryan. 1993. "What Happened to Soviet Product Quality? Evidence from the Finnish Auto
Market" Univcrsity of Miami, Miami, FLIt
Rostowski, Jacek. 1992. "The Inter-Enterprise Debt Explosion in the Former Soviet Union: Causes,
Consequences, Cures." Comnunist Economies and Economic Transformation 5(2): 131-59.
. 1993. "The Implications of Rapid Private Sector Growth in Poland." Discsion Paper 159.
London School of Economics and Political Science, Centre for Economic Performance.
Schaffer, Mark. 1993. "The Enterprise Sector and Emergence of the Polish Flsca Crisis, 1990-91." Policy
Research Worling Paper 1195. World Bank, Transition and Macro-Adjustment Division, Washington,
D.C.
Solimano, AndrEs, and David E. Yuravlivker. 1993. "Price Formation, Nominal Anchors, and
Stabilization Policies in Hungary: An Empirical Analysis." World Bank, Policy Research Department and
Europe and Central Asia Country Department 1}, Washington, D.C.
Tanzi, Vito. 1993. Transition to Market. Studies mi Fiscal Reform. Washington, D.C.: International
Monetary Fund.
'hnsidion. 1994. World Bank, Policy Research Department, Washington, D.C January.
UNICEF (United Nations Childrens Fund). 1993. Public Policy and Social Conduitons. Regional Moni tory
Report 1. New York.