The Perils of Polarization: Economic Performance in the Postcommunist World
Author(s): Timothy Frye
Source: World Politics, Vol. 54, No. 3 (Apr., 2002), pp. 308-337
Published by: Cambridge University Press
Stable URL: http://www.jstor.org/stable/25054190
Accessed: 18/12/2009 09:12
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=cup.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to World
Politics.
http://www.jstor.org
THE PERILS OF POLARIZATION
Economic
Performance
Postcommunist
in the
World
ByTIMOTHY FRYE*
visions of the politics of economic reform dominate the liter
ature.
The most widespread approach relies on the logic of the
TWO
of reform is temporal: re
can be
economic
gains in the future but
large
overcome
resis
reforms
achieved
today. To
only by imposing
painfiil
as labor,
tance from groups
in
from
the
short
reform
such
term,
losing
and state-sector
need to concen
governments
pensioners,
employees,
are
to reform,
trate power
in executives who
committed
ideologically
and
insulated
from
international
financial
backed
by
organizations,
J-curve,1 which
forms promise
holds
that the central dilemma
to generate
this view recendy,
many have criticized
popular pressure. Although
on
economic
still guides many
academic
and policy debates
reform.2
A second view holds particular
for the postcommunist
relevance
world. Argued most persuasively by Joel Hellman,
view
winners
is that the main
obstacles
from distortions
to economic
in the transition
the "partial reform"
transformation
economy
it
who
are the
early
then use their
*
I would
like to thank Andrew Barnes, David Bearce, Carles Boix, Valerie Bunce, Greg Caldeira,
Phil Roeder, Kira Sanbonmatsu,
Kurtz, Ed Mansfield,
Joel Hellman, Marcus
Anthony Mughan,
reviewers for excellent comments. Previous ver
and three anonymous
Kevin Sweeney, Kuba Zielinski,
sions of this essay were presented at the Davis Center for Russian Studies at Harvard University,
the
of the American
annual meeting
Political Science Association,
Cambridge, Mass., February 2001, and
of Slavic Studies, Washington,
the annual meeting
of the American Association
for the Advancement
2001.
D.C., November
1
Przeworski, Democracy and theMarket: Political and Economic Reforms inEastern Europe and Latin
America
Press, 1991). Przeworski
(New York Cambridge
University
cogendy describes the logic of
"In Search of aManual
for Technopols,"
the J-curve but does not endorse this view. John Williamson,
in John Williamson,
ed., The Political Economy ofPolicy Reform (Washington D.C.: Institute for Inter
inDe
national Economics,
"Introduction: The Politics of Economic Adjustment
1994); Joan Nelson,
in Joan Nelson,
in
ed., Economic Crisis and Policy Choice: The Politics ofAdjustment
veloping Countries,"
the Third World (Princeton: Princeton University Press, 1990); Stephan Haggard, Pathways from the Pe
inNewly
Countries
(Ithaca, N.Y.: Cornell University
Industrializing
riphery: The Politics of Growth
Press, 1990).
2
see
Most
(Manu
Joseph Stiglitz, "Whither Socialism? Ten Years of the Transition"
prominendy,
(New
ed., Sustainable Democracy
D.C.,
Bank, Washington,
1999); Adam Przeworski,
script, World
York: Cambridge
University
Press,
1995).
WorldPolitics 54 (April2002), 308-37
PERILSOF POLARIZATION
gains to block further reform.3According
economic
reform
309
to this view, in the short term
to
tends
concentrated
for well
benefits
produce
costs
while
the
of
transformation
groups
positioned
dispersing
the rest of society. The greatest
threat to economic
reform
throughout
comes therefore
from a narrow group of industrial managers,
bankers,
state for their per
and natural resource tycoons bent on
the
capturing
interest
sonal benefit.
diverse
from
view suggests
that robust political
and
competition
are essential
to
coalitions
the
winners
governing
prevent
early
state and
further reform.
taking control of the
sidetracking
This
This
article
an alternative
advocates
and anticommunist
nomic
resources.4
tween
and political
ex-communist
effect
on economic
have
It treats economic
approach.
as a reflection
of the political
factions
engaged
formance
struggle
in a war
per
ex-communist
between
of attrition
over
be
Specifically,
political
polarization
and anticommunist
factions has had a devastating
these competing
growth. Where
political
over
to
continue
and
power
equal
struggle
had
eco
roughly
nomic rules of the game?as
has been slow.Where
either
factions
the eco
or Ukraine?economic
in Bulgaria
growth
factions
or anticommunist
ex-communist
scene?as
or Uzbekistan?eco
in Estonia
the political
nomic performance
has been considerably
better.
Political
outcomes
economic
in two ways.
has
polarization
shaped
it
has
about
economic
future
conditions
First,
uncertainty
heightened
have dominated
because
businesses
expect
a
turnover
potential
to
in government
bring
sharp swings in policy. Facing this possibility, businesses have shied
away from productive
long-term
investments,
preferring
asset
instead
stripping, intensive lobbying of state officials and highly profitable but
deals. More
semilegal business
broadly, political
polarization
to make
it difficult
for governments
credible commitments
and future property
existing
rights.
has made
to respect
Second, political polarization has led to awar of attrition inwhich
ex-communist
herent
measures
and anticommunist
to address
anticommunist
tries,
neoliberal
reforms,
tempted
some version
factions
and
traditional
of gradual
failed to agree on co
coun
In the polarized
some
of
version
attempted
factions
the economic
have
have
crisis.
ex-communist
reform,
but neither
factions
has been
at
have
able
to
impose its preferred policy. The resulting war of attrition has led to in
coherent
policy
and slow growth.
3
"Winners Take All: The Politics of Partial Reform
Hellman,
Politics 50 (January 1998).
4
Alberto Alesina and Allen Drazen,
"Why Are Stabilizations
81 (December
1991).
in Postcommunist
Delayed?"
Transitions,"
World
American Economic Review
WORLD POLITICS
310
on an
Based
analysis
of economic
growth
munist countries during the period 1990-98,1
in twenty-five
postcom
present three findings.
economic
has hindered
First, political
growth,
controlling
polarization
in polarized
countries
for a range of factors. Second,
economic
growth
in such coun
has followed
the electoral calendar: as elections
approach
a
rates
in
increase
and
economic
the
odds
of
tries,
policy
growth
change
of
elections
In
the
effect
countries,
by contrast,
plummet.
nonpolarized
on growth is absent. Third, political polarization
greater
policy
volatility,
a
that
finding
is consistent
is associated with
with
the argument.
These findings have implications for studies of the postcommunist
the political
transformation,
nomic reform more generally.
This
nomic
business
cycle,
and
the politics
of eco
on the
of eco
differs from existing
literature
politics
in three respects. First,
it
in the postcommunist
world
Lib
economic
rather than economic
liberalization.5
growth
article
reform
measures
is an important
of transformation
but is not an
component
on cross-sectional
in itself. Second, while many works
analysis
rely
or focus on a subset of years within
this article takes full
the last decade,
eralization
end
time pe
of available data and pools that data over a ten-year
advantage
an
of
literature
scholars have developed
riod.6 Finally,
body
impressive
outcomes
in
often lack
but
these
works
countries,
explaining
particular
a
comparative
perspective.7
a
In Section
In Section
the argument.
I, I develop
II, I conduct
and
of
effects
the
the
of
quantitative
analysis
polarization
political
electoral
calendar
on economic
impact of political
growth.
polarization
In Section
III, I examine
on policy volatility.
Section
the
IV
concludes.
5
Hellman
nist World,"
in the Post-Commu
of Economic
Reform
(fn. 3); M. Steven Fish, "The Determinants
and David A.
East European Politics and Society 12 (Winter 1998); Jeffrey S. Kopstein
of the Postcommunist World," World Politics 53
and the Transformation
Reilly, "Geographic Diffusion
(October 2000).
from Post-Com
Peter Boone, and Simon Johnson, "How to Stabilize: Lessons
6Anders Aslund,
no. 1 (1996); Fish (fn. 5);
munist Countries," Brookings Papers on Economic Activity,
Stanley Fischer,
inTransition Economies: The Early Expe
and Growth
Ratna Sahay, and Carlos Vegh, "Stabilization
and Reilly (fn. 5); Olivier Blan
rience," Journal ofEconomic Perspectives 10 (Spring 1996); Kopstein
Transition
Press, 1997).
(Oxford: Oxford University
chard, The Economics of Post-Communist
7
MIT Press, 1995); Anders Aslund,
to aMarket
Jeffrey Sachs, Poland's Jump
Economy (Cambridge:
How Russia Became aMarket Economy (Washington, D.C.: Brookings
Institution,
1995); Laszlo Bruszt
and David
Stark, Post-Socialist
Pathways: Transforming Politics and Property in East Central Europe
Press, 1997); David Bartlett, The Political Economy ofDual Tran
(Cambridge: Cambridge University
Reform and Democratization
Shleifer and Daniel Treisman,
1997); Andrei
Russia (Cambridge: MIT Press, 2000).
sitions: Market
of Michigan
inHungary
Press,
(Ann Arbor: University
Without aMap: Political Tactics and Economic Reform in
PERILSOF POLARIZATION
I.The Political
Roots
of Economic
311
Performance
over
have received more
attention
politics
the politics of economic
reform. The wave of
neoliberal
reforms that swept Latin America, Western
Europe, Africa,
and eventually
the postcommunist
world
interest in
renewed
sparked
Few
in
issues
comparative
years than
the last fifteen
on economic
the impact of
The postcommunist
performance.
politics
with
their
diverse
and economic
institutional
countries,
arrangements
a
environment
offer
rich
for
the
outcomes,
topic. Indeed, de
exploring
the homogenizing
effects
the
differences
globalization,
of Soviet-style
socialism
and economic
across
in economic
post
performance
are
to the
Bank
striking.8 According
European
spite
communist
countries
for Reconstruction
the size of the economy
and Development,
in Hun
gary increased by 2.5 percent annually from 1992 to 1998; inMoldova
it fell by 8.5 per year during the same period?a contraction of historic
to
proximity
countries
Many
proportions.9
Europe,
have
defied
Given Ukraine's
predictions.
educated workforce,
and balance
its large market,
of agriculture and industry,Deutsche Bank rated it the Soviet republic
with
the greatest
economic
decline
the economic
economic
basket
in 1990.
potential
over
the subsequent
case of the 1980s,
But Ukraine
decade.10 Meanwhile,
as a
regional
emerged
saw vast
Poland,
economic
power in the 1990s.
on the
economic
reform points to a range of
politics of
economic
view
factors
that may
influence
The J-curve
performance.
are
to
elite
central
that
and
power
suggests
presidential
partisanship
are
com
executives
who
economic
strong
By insulating
performance.
The
literature
to reform and backed
countries
community,
by the international
to
withstand
exerted
the
may
pressure
by groups bearing
change policy
costs of reform. In contrast,
the partial reform view empha
short-term
sizes the importance
of democracy
and dispersed
power for
political
it difficult
economic
These
the short
features
make
for
performance.
mitted
term winners
from
the state and derail
distortions
reforms.
in the transitional
Others
have pointed
economy
to state
to capture
in
spending,
8
of Post-Socialism,"
Slavic Review 58 (Winter 1999).
Valerie Bunce, "The Political Economy
9
1 for descriptive statistics on polarization
and growth. The EBRD's Transition Report
See Appendix
1999 notes laconically that growth rates in the region "can lack precision." It also notes that some coun
tries incorporate estimates of the size of the informal economy into their growth rates. The growth data
are
likely the best available and have been widely used. European Bank for Reconstruction
presented here
Transition Report (London: EBRD, 1999), 188. In the quantitative analysis, I introduce
and Development,
a correction to the
growth data that attempts to account for the size of the informal economy. See fh. 60.
10
in Bartlomiej
in the Post-Soviet
"Economic Transformation
Gertrude
Schroeder,
Republics,"
in Russia and theNew States of Eurasia
(Armonk, N.Y.: M. E.
Kaminski,
ed., Economic Transition
Sharpe,
1997).
312
WORLD
POLITICS
and ties to international
legacies, geography,
and the IMF, as important
such as the European
Union
in transition
societies.
economic
performance
stitutional
organizations,
of
determinants
I adopt a different approach inwhich I analyze how the balance of
in postcommunist
power
political
mance. More
I argue
specifically,
countries
affects
that political
economic
polarization
perfor
exacerbates
the distributional effects and credible commitment problem of eco
nomic
transformation
and
leads
to slower
growth.
Before
developing
the argument, I briefly discuss my treatment of political polarization.
the term
in many ways.11
"political polarization"
as used in this article, the two
in
factions
major political
Conceptually,
are ex-communist
and
anticommunist
fac
systems
polarized
political
Scholars
have
used
tions. Leaders of the former have typically held responsible positions
within
mitted
the state apparatus or communist
a dominant
to
state
supporting
com
party, have been largely
with
sector, have campaigned
the pos
and have emphasized
the backing
of the ex-communist
party,
to 1989.
itive side of communist
policy prior
In contrast,
the leaders
of
anticommunist parties have either left or never held positions in the
communist party prior to 1989, have been highly critical of the activi
run
party prior to 1989, have
against the largest
a
role for the private
dominant
ex-communist
party, and have favored
sector. This
it reflects diver
because
is
important
cleavage
particularly
over
structure
of
For
the
views
the
the
economy.
gent
example, during
the
ex-communist
Com
in Russia,
the traditional
party,
past decade
ties of the communist
munist
Party
of the Russian
Federation,
promised
continued
state own
ership of land and a leading role for the state in finance and heavy
a mas
led by President Yeltsin
promised
industry; the anticommunists
on
sive restructuring
of the economy
based
liberalized
private property,
prices, and curtailed state subsidies for industry.12Not only do these
11
Giovanni
Sartori defines a polarized party system by the ideological distance between parties, but
are unavailable in the postcommunist
world. Others measure polarization
such precise measures
using
indices assume that cleavages are po
indices of social cleavages, such as ethnic fractionalization. These
The salience of the ex
litically salient, but many cleavages do not translate into political movements.
cases seems to be quite
communist/anticommunist
high. Other types of
cleavage in the postcommunist
as ethnic divisions, also do not translate readily into economic policy. For example, na
cleavages, such
in
of a market economy
continued
the construction
tionalist rhetoric aside, Prime Minister Meciar
Slovakia. Sartori, Parties and Party Systems (New York: Cambridge University
Press, 1976); William
Quarterly Journal
Easterly and Ross Levine, "Africa's Growth Tragedy: Policies and Ethnic Divisions,"
and Prop
ofEconomics 82 (November 1997); Philip Keefer and Stephen Knack, "Polarization, Politics,
Bank, Washington,
D.C., 2000). For a critique of these indices, see
erty Rights" (Manuscript, World
for Constructing
Ethnic Frac
of Constructivism
David Laitin and Daniel Posner, "The Implications
tionalization
Indices," APSA-CPNewsletter for the Organized Section in Comparative Politics oftheAPSA 12
(Winter2001).
12
Zyuganov, My Russia: The Political Autobiography
ofGennady
Gennady
Medish
(fn. 7).
(Armonk, N.Y.: M. E. Sharpe, 1997), pt. 4; Aslund
Zyuganov,
ed. Vadim
PERILSOF POLARIZATION
313
competing platforms have little in common, but they also produce dif
ferent winners
Not
and losers.
all ex-communist
parties
fall
into
the "traditional"
category.
John Ishiyama and others have argued that the Hungarian,
Polish,
are far more
ex-communist
and Slovenian
Lithuanian,
parties
are
ted to market-oriented
reforms than
other ex-communist
the region.13 These
parties
social-democratic
European
more
left-wing
adopted
parties in
to
roughly akin
economic
parties and often
factions. Given
ex-communist
commit
platforms
faced competition
their histories,
from
however,
these parties lacked credibility, particularly with private business. Here
I treat
these
My
in two ways.
Ini
parties
ex-communist
parties.
ex-communist
parties.
ex-communist
reform-oriented
I treat them
tially,
I then place them
as distinct
from
in one group with
"traditional"
other
treatment of political polarization is similar to that of Stephan
Haggard
and Robert
Kaufman,
who
use
antisystem
parties, which
they
describe as "left and populist parties that have historically mobilized
as indicators
or
of po
anti-oligarchic
protests,"
anti-capitalist
true for traditional
ex-communist
This depiction
par
rings
as used here may
in
ties.14 Political
bring
sharp
swings
polarization
around
larization.
policy. In addition, however, it raises the possibility of change in the
underlying
The War
structural
features
of the economy.
of Attrition
tends to undermine
polarization
Alesina
and Allen
least two ways. Alberto
Political
economic
Drazen
in at
performance
in
that
argue
politi
cally polarized settings it is difficult for politicians to agree on economic
costs on
but impose distributional
social welfare
that promote
a
costs of transformation
in
be
The
may
variety
paid
specific groups.15
or
of ways: by levying taxes on capital, by sacking state workers,
by end
sectors. The
to
key is that each group prefers
loss-making
ing subsidies
these costs are to be
that some other group pay these costs. Where
policies
divided among competing political groups that have very different pol
icy preferences
and where
these groups
do not know who
will
concede
the political battle by agreeing to pay the costs of transforming the
13
successor
creates three categories of communist
liberal, and dem
parties?standpatter,
Ishiyama
on their policy positions
in three areas: the economy, the communist past, and
ocratic reformist?based
the value of democratic
Structures, Leaders,
Ishiyama, "Communist Parties inTransition:
competition.
in Eastern Europe," Comparative Politics 27 (January 1995); idem,
and Processes of Democratization
of Ex-Communist
"The Sickle or the Rose: Previous Regime Types and the Evolution
Parties," Com
Political Studies 30 (June 1997).
parative
14
and Kaufman, The Political Economy
Haggard
versity Press, 1995), 167.
15
and Drazen
Alesina
(fn. 4).
ofDemocratic
Transitions
(Princeton: Princeton
Uni
314
WORLD
economy,
(awar
stalemate
political
POLITICS
of attrition)
is
likely
to ensue. Dur
ing this stalemate both parties seek to shift the costs of the new policy
to their
opponents,
political
economic
policies
a situation
a
Consider
country
that inhibits
the introduction
to promote
necessary
growth.16
two groups: one earns revenue from
coherent
with
of
capital
and the other from labor.17Although both will benefit by transforming
the economy,
each prefers that the other group bear the distributional
associated
with
The
transformation.
faction
capital-oriented
to end the
liberalize prices, and open in
constraint,
prefers
soft-budget
costs
to
labor-oriented
faction prefers a gradual re
dustry
foreign trade. The
taxes on the
form with elements
of protection,
significant
private sector,
to workers
to cushion
and redistribution
the costs of transformation.
Where
of these
over the distribution
prevents
agreement
political
polarization
to introduce
difficult
coherent
costs, it is exceedingly
policies
that promise to improve the economy. In a highly polarized setting
each group
tion ensues.
expects
the other
to concede
first and a costly war
of attri
one faction wins the
Only when
political
struggle?for
example after
an election
we
the
loser?should
expect coherent govern
marginalizes
a
ment
the
sector, and improved
response by
policy,
productive
private
economic
The
consolidation
of
forces around a
performance.18
political
war of attrition
similar
ends
the
and
allows
the winners
roughly
policy
to shift the costs of transformation
onto the losers.
the war of
Ending
attrition
is
essential
for
the
of
coherent
and
quickly
adoption
policies
the resumption
of growth.
awar of attrition
The
of
logic
set
in a postcommunist
special bite
the scope of economic
the potential
ting. Given
change,
gains for win
ners and costs for losers are
In
because
addition,
especially high.
political
are often in flux, the
in transition
institutions
economies
early winners
has
16
Alberto Alesina
and Howard Rosenthal,
Partisan Politics, Divided Government and the Economy
(New York: Cambridge University
Press, 1995); Morris Fiorina, Divided Government
(Boston: Alwyn
Bacon, 1996).
17
over an asset and choose to
or not
More
formally, two players bargain
fight
fight. If neither fights,
each receives 0. Fighting
w.
is costly so each player must pay a cost, c, and the winner receives a
prize,
Each is uncertain about the costs the other side can bear. As long as w > 0 > c, each party
prefers to
rather than not fight. Given
these incentives, each chooses to fight in hopes of
fight and receive w-c,
to their opponents. The result is awar of attrition.
shifting the costs of transformation
Not Fight
Not
18
Alesina
^
Fight
0,0
S
FiSht
w-c,
and Drazen
(fn. 4).
Player
1
Fight
-c, w-c
-c
-c, -c
PERILS OF POLARIZATION
315
have an incentive to delay political consolidation around one political
faction
they
can
shape political
Credible
thereby
credible
amassed
Once
great wealth.
having
to lock in their economic
institutions
done
so,
gains.19
Commitment
to
In addition
increases
they have
until
the war
of attrition,
polarization
political
in economic
the probability
of sharp changes
and
policy
to make
a
in the government's
confidence
undermines
ability
over
to
commitment
time.20 Potential
transfers
property
rights
exacerbating
in polarized
of political power are often decisive moments
sys
political
tems because
all groups expect great swings in economic
policy. Losers
can
winners
and vice versa. At a minimum,
these
become
quickly
a maximum,
in
they
changes may bring about rapid changes
policy. At
In either case, political
in the economy.
may lead to structural changes
increases uncertainty
and leads firms to discount
their ex
polarization
pected
returns
on investment
and to lobby
the state as an insurance
pol
icy against political change.
incur
institutions
that offer few protections
against
judicial
a
state
in
environment
increase
sions by the
polarized
postcommunist
can
in
the size of potential
governments
swings
Incoming
policy.
Weak
manipula
rights using creeping ^nationalizations,
change property
to
tions of tax policy,
of property
sup
reassignments
rights
political
so.
have
their rivals.21 And
done
porters, and threats to incarcerate
they
to respect the
cannot
Because
governments
credibly commit
incoming
property rights of their rivals in a polarized setting, the latter have little
incentive
to make
long-term
ers of the current
vorable
even the
Moreover,
support
for fa
that their opportunities
last only so long as their allies
government
recognize
state officials will
by
so
to invest. Think
incentives
of a
they, too, have weak
an
in
is considering
in polarized
Russia who
making
treatment
are in power,
investments.
businessperson
vestment
after the introduction
of an economic
reform
program.
Facing
19
and Democracy:
Hellman
(fn. 3); Timothy
Insights from the Post
Frye, "Presidents, Parliaments
in Andrew Reynolds,
communist World,"
ed., The Architecture
ofDemocracy: Constitutional Design,
and Democracy (New York: Oxford University
Press, 2002).
Management,
Conflict
20
and Government
"A Positive Theory
and Guido Tabellini,
of Fiscal Deficits
Alberto Alesina
Debt," Review ofEconomic Studies 57 (July 1990); Jakub Svensson, "Investment, Property Rights, and
and Evidence," European Economic Review 42, no. 7 (1998).
Political Instability: Theory
21
this dynamic. A Russian oligarch recounted to a journalist that in early
One anecdote highlights
at the Davos meetings
Economic
Soros told him: "'Boys,
of theWorld
Forum, George
February 1996
[Soros's] argument was
your time is over. You've had a few good years but now your time is up.' His
he said, should be careful
that the communists were definitely going to win. We Russian businessmen,
to get to our jets in time and not lose our lives." Chrystia Freeland, The Sale of the
that we managed
to
(New York: Random House,
2000), 192.
Capitalism
Century: Russia's Wild Ride from Communism
316
WORLD
that the current
the probability
POLITICS
may
government
give way
to a govern
ment with far different preferences thatwill put any such investment at
to invest.22
risk, he or she will be reluctant
in
In contrast,
Slovenia who
think of a businessperson
nonpolarized
can invest with
not
that economic
will
the confidence
change
policy
in
This
of
the
environment.23
dramatically,
regardless
changes
political
is central
confidence
to investment
reform more
for economic
decisions
in particular
and support
generally.24
The effects of political polarization are likely to be especially strong
prior
to elections.
As
elections
the probability
approach,
of a large
swing in policy increases, making it especially difficult for the govern
ment
sector to invest in
activities.
the private
productive
over future economic
lead
busi
policy may
uncertainty
are
in. Politi
results
their funds abroad until the election
to convince
Heightened
nesses to
cians may
to
actually
park
in an election
seek to promote
growth
a
from
response
generate
progrowth
year, but their ability
sector in a
the private
setting will be limited. In polarized political
polarized
therefore
economic
from
and lowest
expect
an election
systems, we
to be
in the years farthest
highest
growth
in election
years.25
leads to slower
that political
polarization
In sum, we should find
rates of economic
in
that growth
systems is
polarized
political
growth,
to the time to the next election,
and
that
related
po
political
inversely
larization
leads to more volatile policy.
cases
Particular
suggest
the ex-communist
Bulgaria
munist United
Democratic
the plausibility
In
of the argument.
polarized
Socialist Party and the anticom
Bulgarian
in the coun
Front, the two leading factions
economic
different
reforms.
have proposed
very
try over the last decade,
to sustain an
has gained
sufficient
Neither,
however,
power
political
over time such that it could
costs of trans
the
economic
impose
policy
formation
on
the other,
so each
faction
has
instead
used
its time
in
22
On Russia, see Aslund
(fh. 7); Vladimir Mau, Russian Economic Reforms as Seen by an Insider: Suc
cess orFailure? (London: Chatham House, 2000); Jerry Hough,
The Logic ofEconomic Reform in Russia
Institution, 2001).
(Washington, D.C.: Brookings
23
Studies 45,
On Slovenia, see Sabrina Petra Ramet, "Slovenia's Road to Democracy," Europe-Asia
no. 5 (1993); Joze Mencinger,
"The Slovene Economy," Nationalities
Papers 21 (Spring 1993); Jeffrey
in Slovenia," inOlivier
and Economic
Reform
"Political Independence
Sachs and Boris Pleskovic,
in Eastern Europe, vol. 1,
Kenneth A. Froot, and Jeffrey D. Sachs, eds., The Transition
Blanchard,
of Chicago Press, 1994).
Studies (Chicago: University
Country
24
Przeworski
(fn. 1).
25
and Vladimir Gim
For a treatment of the political business cycle in Russia, see Daniel Treisman
or the
of Chudar," British
Manipulations
pelson, "Political Business Cycles and Russian Elections,
Journal of Political Science 31 (April 2001). See also Timothy
Frye and Edward Mansfield,
"Timing is
in the Post-Communist
World"
and Trade Liberalization
Elections
(Manuscript, Ohio
Everything:
of Pennsylvania, November
and University
State University
2001).
D.
PERILSOF POLARIZATION 317
The
Countries*
Polarized
Average
Growth
TABLE 1
at a Glance
Cases
Countries
Nonpolarized
Rate
-7.9%
(1.1)
Average
Albania
Growth
Rate
-1.2%
(. 71)
Armenia
Belarus
Azerbaijan
Croatia
Bulgaria
KyrgysstanCzech Republic
Moldova
Estonia
Romania
Russia
Ukraine
Hungary
Kazakhstan
Latvia
Lithuania
Macedonia
Poland
Slovakia
Slovenia
Tajikistan
Turkmenistan
Uzbekistan
N = 249; T= 4.65; standard deviation in parentheses
countries
aThese
office
have had polarized
to redistribute
political
economic
systems
resources
than to promote
growth. Consequendy,
a few
industrial
large
conglomerates,
for at least half
the years
in the
sample.
to its own
rather
supporters
has
been
weak.
Indeed,
growth
such asMultiGroup,
have weak
in
the state.26 By contrast,
the economy
by looting
nonpolarized
as
anticommunist
such
Estonia
and
and ex
countries,
Uzbekistan,
communist
have dominated
and quickly
factions,
politics
respectively,
ened
imposed
Political
the economic
costs
about
uncertainty
and economic
countries,
Some quantitative
evidence
these
of transformation
economic
policy
on their
opponents.27
has been much
lower in
much better.
performance
is also consistent with
the argument.
In
Table 1,1 divide the cases into countries that had polarized political sys
26
and Bruce Parrot,
On Bulgaria, see John D. Bell, "Post-Communist
Bulgaria?" in Karen Dawisha
in South-Eastern
York: Cambridge
eds., Politics, Power, and the Struggle for Democracy
Europe (New
(Cheltenham, U.K.: Ed
Press, 1997); John Bristow, The Bulgarian Economy in Transition
University
as State Breakers in Postcom
ward Elgar, 1996); Venelin Ganev, "The Dorian Gray Effect: Winners
Studies 34 (January 2001).
munism," Communist andPost-Communist
27
see Gerald M. Easter, "Preference for Presidentialism:
Postcommunist
On Uzbekistan,
Regime
in Russia and the NIS," World Politics 49 (January 1997); Asad Alam and Arup Bannerji,
Change
and Kazakhstan: A Tale of Two Transition
Paths" (Manuscript, World
Bank, Washing
see Ole
and Lars Johannsen, The Baltic States after Independ
ton, D.C., 2000); on Estonia,
Norgaard
ence
Mass.: Edward Elgar, 1999); Terry Cox and Bob Mason,
Social and Economic
(Northampton,
inEast Central Europe (Cheltenham, U.K.: Edward Elgar, 1999).
Transformation
"Uzbekistan
WORLD POLITICS
318
terns for at least half the period under study and those that did not. I
eraging
because
are
countries
many
polarized
we count
figures
for the
rates
growth
Nonetheless,
two
growth for these
groups.28 Av
over
time
information
forfeits
and polarization
the rates of economic
then compare
annual
average
countries
and -1.2
countries
were
nonpolarized
percent
reform communist
at the
tistically
significant
the
gate
relationship
.05 level. Thus,
between
assess
cludes
factions
as
are
there
GDPGrowit
I estimate
the arguments,
the following
model,
and factors often
for political
polarization
of growth.
=
?0
+
+
+
+
^Polarization?
+
^GDPGrowiM)
+
+
in
which
cited
as de
^^Democracy^
$3GovernmentSpendingit
$sGDPperCapitai(t_1}
closely.
Analysis
variables
terminants
to investi
grounds
and growth more
polarization
II.Quantitative
To
for only part of the decade.
was -7.8
in the
percent
growth
in the
If
countries.
nonpolarized
these
polarized,
-2.2
for
countries
and
the
percent
percent
polarized
are sta
In either case, these differences
countries.
with
-5.9
polarized
economic
+
fy?penness^)
$7LogInfiation
^6Warit+
?{$9Country?
i(t_1}
+ Z($10Yearit)+ eit (1)
is the real rate of year-to-year
variable, GDPGrow.f,
dependent
/
GDP in country
in 1998 U.S. dollars. The
/
in
in
measured
year
change
rate
in GDP in the countries
of annual economic
average real
growth
economic
Scholars have measured
under study is -3.8 percent.
perfor
The
Jose Maria
ways, but as Luiz Carlos Bresser-Perreira,
note:
economic
crite
"The
ultimate
Przeworski
and
Adam
Maravall,
a
coun
success
can
rion for evaluating
of reforms
the
only be whether
at
of inflation.,,29
levels
stable
and
moderate
resumed
try
growth
mance
in many
The main independent variable of interest is political polarization.
the seat share of the largest ex-communist
(an
an
anticommunist
holds
(ex-communist)
ticommunist)
in Bulgaria
in 1994 the anticommunist
the executive.
For example,
won
Forces
29
of the seats and formed
Union
Democratic
of
percent
Polarization.
measures
faction
when
to the
the largest party opposed
prime minister
its polarization
Socialist
Party. Thus,
Bulgarian
from
the ex-communist
score was
29.
28
Here I treat a polarized country as having at least 20 percent of the seats held by a traditional
the executive is held by an anticommunist
ex-communist
(ex-commu
(anticommunist)
party when
nist) in a given year.
29
inNew Democracies: A Social
Economic Reforms
and Przeworski,
Maravall,
Bresser-Perreira,
Democratic Approach (New York: Cambridge University
Press, 1993).
PERILSOF POLARIZATION
319
on
little consensus
have produced
is
there
but
economic
strong
type
regime
growth,
has gone hand in hand with de
sentiment
that economic
performance
in the
is consistent with the par
world.30 This
mocracy
postcommunist
better
than other
should perform
tial reform view that democracies
a value of 1 for each year that Freedom
takes
countries.
Democracy\
rates a country as "free," that is, a lor 2 on its
scale of
House
7-point
from
Studies
other
whether
rights;
political
it takes a value
otherwise
of 0.31
levels of government spending may cushion the distributional
High
effects
regions
influences
of free markets
and
theorists
thereby
en
promote
Similarly,
growth.32
that targeted government
spending
argue
However,
may generate
spending
high levels of government
growth.33
more
resources
at the ex
to pursue
also
rulers
may
give
personal wealth
is the ratio of
pense of economic
performance.
GovernmentSpendingit
as a
of gross domestic
government
percentage
product.34
spending
to the world
is
often
cited as critical to economic
economy
Openness
includes an an
the statistical model
growth.35 To capture this notion,
dogenous
growth
nual index of external liberalization compiled byWorld
mists
and country
autarkic economic
experts.
system
Bank econo
.
ranges from 0 for a completely
Openness
to 100 for an economic
system fully open to
trade.36
foreign
I include
country
including
a
dummy variable, War^
in awar. I control
is involved
the gross
domestic
product
that equals
for the wealth
per
capita,
1 for each year that a
of each country by
GDPper
Capita.,..
30
Crisis: The Latin American
Karen L. Remmer,
and Economic
Experience," World
"Democracy
Is the Third
"The Politics of Economic Transformation:
Politics 42 (April 1990); Joan M. Nelson,
in Eastern Europe?" World Politics 45 (April 1993); Aslund, Boone, and
World
Relevant
Experience
(fn. 3).
Johnson (fn. 6); Hellman
31
is nonetheless
somewhat arbitrary. I estimate model 1 after
This threshold, although a convention,
as (a) 3 and lower, (b) 4 and lower.
so does not alter the
Doing
redefining the threshold for democracy
are available
at Freedom
Annual
results. Data
Scores,
House,
Country
Survey of Freedom
III
Similar results are obtained using updated POLITY
1972/73-1998/99
(www.Freedomhouse.org).
are then treated as a dummy variable with the technique advocated by
scores for
democracy, which
Third Wave with the
Keith Jaggers and Ted Robert Gurr; Jaggers and Gurr, "Tracking Democracy's
1995).
Polity III Data," Journal ofPeace Research 32 (November
32
as a
and Przeworski
(fn. 29). That government
Bresser-Pereira, Maravall,
portion of
spending
ofthat spending.
GDP is relatively high indicates nothing about the content or beneficiaries
33
and Peter Howitt, Endogenous Growth Theory (Cambridge: MIT Press, 1998).
Phillipe
34EBRD Aghion
(fn. 9).
35
"Economic Reform and the Process of Global Integration,"
Jeffrey D. Sachs and Andrew Warner,
onEconomic
1, no. 1 (1995); Robert Barro, Determinants
ofEconomic Growth
Activity
Brookings Papers
David Romer, "Does Trade Cause Growth?"
(Cambridge: MIT Press, 1997); Jeffrey A. Frankel and
89 (June 1999).
American Economic Review
36
in the former Soviet re
(fn. 6) argue that economic performance
Aslund, Boone, and Johnson
states in the region due to "different underlying
structural factors,
publics may differ from that in other
a
such as the greater reliance on military-industrial
greater
longer history of communism,
production,
in the ruble zone when control over
the communist
reliance on trade within
bloc, and membership
money creation disintegrated."
320
WORLD POLITICS
to exhibit
economies
stronger
expect wealthier
might
perfor
a
to
would
mance.37 Alternatively,
neoclassical
growth
approach
predict
that poorer countries would
exhibit higher rates of growth.381
also in
to ac
cluded a lagged value of the dependent
variable, GDPGrow.,
_1}
count for
in
the
data.39
temporal dependence
One
The
includes
sample
twenty-five
countries
postcommunist
during
the period 1990-98.40 After pooling these data, I report an ordinary
least squares (OLS) model
for the annual rate of economic
growth.41 Be
the sample has fewer than 250 observations
and the data are
cause
in a
arranged
consistent
Long
To
panel,
covariance
and Laurie
tests
statistical
matrix
on a
are based
(HCCM) as suggested
heteroskedasticity
by J. Scott
recently
Ervin.42
I also in
economic
conditions,
exogenous
as
in
countries
This
is
for
each
year.
dummy
important,
to similar exogenous
the region were
from the interna
shocks
exposed
tional economy. To control for unmeasured
factors specific to individual
of the econ
such as institutional
countries,
legacies and the composition
control
clude
for unmeasured
variables
omy, I also add a dummy variable for each country. Including fixed-effect
variables
dummy
gives
ported
model
reduces
confidence
greater
but are available
are
lagged
concerns
for omitted
variable
in the results. These
from
the author. The
a year to reduce
by
fixed
economic
the likelihood
bias and thereby
are not re
effects
variables
of reverse
in the
causation.
Results
Results
Controlling
from model
1 in Table
2 are consistent
for a range of factors, Polarization.
the argument.
is negatively
and signif
with
37
and Howitt
(fn. 33).
38Aghion
Barro (fn. 35).
39
on the
Achen
grounds
Christopher
critiques this strategy for addressing temporal dependence,
that the lagged endogenous
variable may substantially deflate the impact of other independent vari
ables. The results of the argument are stronger when the lagged dependent variable is dropped. Achen,
Variables Can Suppress the Explanatory Power of Other Independent Vari
"Why Lagged Dependent
ables" (Paper presented at the annual meeting
of the Political Methodology
Section of the American
Los Angeles,
Political Science Association,
July 2000).
40
The former Soviet countries enter the data set in 1992 after the fall of the Soviet Union.
41
the years in the data by more than 2:1,1 do not employ the correc
Because the units outnumber
tion suggested by Nathaniel
Beck and Jonathan Katz; Beck and Katz, "What to Do (and Not to Do)
with Time-Series-Cross-Section
Data in Comparative
Politics," American Political Science Review 89
1995).
(September
42
Long and Ervin show that in samples of fewer than 250 observations Ordinary Least Squares
standard errors. They propose a correction that ismore ap
(OLS) regression may produce inconsistent
are
more
this
for
small
Results
from
propriate
samples.
analysis
slighdy stronger when opting for the
traditional approach of using robust standard errors with clustering on country. I am thankful to an
reviewer for alerting me to this article. Long and Ervin, "Using Heteroscedasticity
Con
anonymous
sistent Standard Errors in the Linear Regression Model," American Statistician 54 (August 2000).
Table
Polarization
2
and Growth1
Model
Polarization
-.26**
continuous
(.12)
Polarization
continuous
-.23**
AllCP
(.10)
Polarization
(2.73)
(3.07)
-4.31**
Polarization
dummy all CP
(2.17)
1.98
Democracy
(2.28)
spending
1.98
(2.25)
GDP per
capita
(2.32)
-.03
-.06
.03
.07
(.11)
.52
2.61
(.10)
(.10)
3.06
1.83
(5.75)
(5.70)
(5.12)
(5.32)
(5.06)
-.0047**
-.0046**
-.0038**
-.0041**
-.0040**
-6.86*
(4.11)
Inflation (log)
2.69
(2.29)
(.11)
(.0017)
War
2.22
(2.33)
.03
1.48
Openness
2.47
(.11)
Government
-1.41**
(.60)
GDPGrow Lag
(.0017)
-6.85*
(4.73)
-1.59**
(.64)
.22**
(.09)
Time
-9.66**
-7.36**
dummy
(.0015)
-7.86**
(3.69)
-1.28**
(S5)
-7.52*
(.10)
(.0015)
-7.81**
(4.1)
-1.52**
(3.54)
-1.28**
(.62)
.15
.23**
.17*
.22**
(.10)
(.0016)
(.11)
(.10)
-.31
to election
(.69)
Time
to election*
2.11*
polarization
(1.21)
18.31**
Constant
(7.94)
.71
R2
*p<.10;**p<.05;***p<.01
aFixed effects for countries
Dependent
variable
(7.57)
184
183
N
16.96**
20.08**
(7.79)
and years are included
is the real rate of annual change
(7.34)
.72
.72
in the analysis
in GDP in U.S.
but
20.01**
(7.60)
184
184
184
.71
15.76**
are not
$1998.
.73
reported.
322
WORLD POLITICS
related
icantly
to economic
growth.43
A
ten-point
rise in the polariza
tion index yields a 2.6 percent decrease in growth, indicating that the
are
as well as
fairly large,
statistically
significant.
as
wealth
of a country,
is nega
proxied by GDPperCapita.(tl)y
a
result that supports neoclassical
models
tively associated with growth,
rate
in
of growth. A10
increase
wealth
decreases
the
of
percent
growth
and high inflation are also negatively
related to eco
by 2 percent. War
results
The
once
to economic
is unrelated
proper
growth. Democracy.t
growth
are introduced.
to
controls
is
Economic
related
openness
positively
as indicated
but
the
is
coefficient
growth,
insignificant,
by Openness.,
1}.
nomic
not
GovernmentSpendingitis positively associated with growth but is
on the
the coefficient
vari
significant. As expected,
lagged endogenous
not
is
and
but
able, GDPGrow.{tljy
very large, indi
significant
positive
cating that growth is fairly volatile in this period.
In the preceding
I have
analysis,
measure
ex-communist
in
the
parties
included
only the "traditional"
In model
of polarization.44
2 I
adopt amore expansive definition of polarization by including the four
"reformed"
nition
has
ex-communist
as well.
parties
on the substantive
Adopting
this broader
defi
little impact
of polarization.
significance
3 presents
the results using a dummy variable for polarization
on the
once an
movement
reaches a cer
relying
opposition
logic that
a
to the
tain threshold,
it becomes
viable alternative
ruling party, while
Model
3
those below this threshold do not. In model
takes a value
polarization
of 1, when
either an anticommunist
controls
the executive
and the
at
ex-communist
least 20 percent of seats
party holds
largest traditional
or where
a traditional
in
ex-communist
the execu
controls
parliament;
at least 20 percent of
tive and the largest anticommunist
control
parties
the seats in parliament,
and 0 otherwise.
Such parties are almost always
the largest opposition
the help of other par
party and, particularly with
are
to
to
costs of trans
well
frustrate
ties,
attempts
impose the
placed
formation
on their
primarily
that the average annual
from model
Results
3
supporters.45
rate
in
countries
indicate
of growth
is
polarized
7.3 percent
countries.
lower than in nonpolarized
I also included
To test the robustness
of this indicator,
all ex-com
more
seats
munist
in parliament
than 20 percent of the
parties that held
or held control of the executive. This measure
does not distinguish
be
43
this threshold to 15 percent does not alter the results presented
in columns 3 and 4 in
Dropping
Table 2. The polarization measure retains its sign and significance with this new coding.
44
Ishiyama (fn. 13).
45
if an antisystem or anticapital
and Kaufman
(fn. 14) consider a party system polarized
Haggard
ist party has 15 percent of the seats in parliament.
PERILSOF POLARIZATION
tween
reformed
and
2 illustrates
Table
323
4 in
parties. Model
to a range of
cod
plausible
of
the
variable
impact
dummy
traditional
ex-communist
are robust
that the results
the
As expected,
ings of polarization.
more
treatment
the
of polarization
using
expansive
in
is 4.3 percent
countries
weaker?growth
polarized
Polarization.
retains its statistical
other countries?but
this analysis the inclusion
amild ameliorative
effect
The Electoral
and Growth
lower
than
in
In
significance.
communist
had
parties
to
previous mod
compared
of the four reformed
on
growth when
this recoding does not dramatically
els. Nonetheless,
is somewhat
Calendar,
Political
change
the results.
Polarization,
5 examines the impact of the electoral calendar in polarized and
Model
settings
nonpolarized
tions?Polarization.t,
by including
that multiplies
an interaction
the number
term, TimeToElec
of years until the next
national election by 1 if the political system is polarized and by 0 oth
erwise.46 As
increases
in a
setting and uncertainty
approach
polarized
the course of future policy, one would
expect politicians
to be more
to engage
in
economic
opportunism
likely
elections
about
and businesses
in strategies
than
The
in model
significant
5 suggests
to promote
growth.
and negative
coefficient
that in election
on the variable
years (when
low in polarized
time
Polarization.
to elections
is 0),
countries. The posi
is particularly
growth
on the interaction
term TimeToElection*
tive and significant
coefficient
a country
Polarization.
indicates
that in a polarized
setting the farther
an
a normal
is from
the growth
the higher
rate.47 Consider
election,
economic
economic
de
electoral
countries,
growth
four-year
cycle. In polarized
clines on average by 3.3 percent
three years prior to an election, by 5.4
to an election,
two years
in the year pre
percent
prior
by 7.5 percent
an election,
in polar
in election years. Thus,
and by 9.6 percent
ceding
in
is
farthest
from an
economic
ized countries
years
growth
highest
as the election
not
It
is
that
election and declines
approaches.
surprising
on economic
in the non
growth
cycle has little effect
as indicated
coeffi
countries,
insignificant
by the statistically
polarized
cient on TimeToElection..
it
calendar
that the electoral
The
results of this analysis demonstrate
the electoral
has
a
large
effect
on
growth
in polarized
countries;
they
also
suggest
46
elections in a parliamentary
elections here include parliamentary
National
system and presiden
tial elections in a presidential
system.
47
systems may hasten calls for an election, thus suggesting
growth rates in parliamentary
Declining
it is notable that declining growth seems
of growth and elections. However,
the potential endogeneity
to have no effect on elections in nonpolarized
countries.
WORLD
324
POLITICS
that growth is reduced because of the potential swings in policy that
Potential
Since
variables
idential
Most
important,
they provide
fur
of the argument.
Variables
Omitted
consensus
is little
there
models,
economic
in such countries.
elections
surround
ther confirmation
concerns
I also address
on
for growth
the proper
specification
that may influence
for omitted variables
the J-curve view, I include
with
Consistent
performance.48
To measure
and
elite
for presidential
pres
power
partisanship.
use
scale of presi
I
and
Matthew
John Carey's
power,
Shugart
dential power as adapted slightly for the postcommunist cases by Frye,
elite parti
of measuring
the difficulty
of
the percentage
first measures
The
sanship,
the
seats held
The
second
ex-communist
the
probes
largest
by
party.50
in a presidential
head
of
of
the
government?a
president
partisanship
a
in a
minister
system and
creating
system?by
parliamentary
prime
career
is based on the executive's
which
the variable
ElitePartisanship^
to 1989 and his
the ex-communist
with
party.51
relationship
prior
I added
the partial reform argument,
with
Consistent
Fragmenta
cases Noriel
Roubini
and
for the postcommunist
which
adapts
tion.^
can
actors
measure
extent
to
which
of
the
partisan
Jeffrey D. Sachs's
Hellman,
and Tucker.49
I created
Given
two variables.
block policy change.52This variable has been used by Hellman
EBRD and measures
players whose
48
As Ross Levine
assent
the number
is needed
and the
factions
and institutional
of political
a change in policy.53
to make
and David Renelt note: "There does not exist a consensus theoretical framework
to guide empirical work on growth, and existing models do not completely
specify the variables that
statistical inference on the relationship between growth and
should be held constant while conducting
of Cross-Country
"A Sensitivity Analysis
the primary variables of interest." See Levine and Renelt,
82 (September
Growth Regressions," American Economic Review
1992), 943; Nazrul Islam, "Growth
A Panel Data Approach,"
1995).
Quarterly Journal ofEconomics 110 (November
Empirics:
49
Electoral Dynamics
(New
Shugart and Carey, Presidents and Assemblies: Constitutional Design and
"Data-Base
York: Cambridge University
Press, 1992); Timothy
Frye, Joshua Tucker, and Joel Hellman,
on Political Institutions
in the Post-Communist
World"
Ohio, 2001). Herbert
(Data set, Columbus,
on eco
offer a conditional
and Edward Malesky
Kitschelt
theory of the effects of presidential power
in the postcommunist
world. They
nomic performance
argue that countries with weak economic
and thus the effect of presidential power is conditional upon initial
prospects choose strong presidencies,
one
seem to
a
the argument would
proposed here.
require model different from the
prospects. Testing
Economic Reform" (Paper pre
"Constitutional Design and Post-Communist
See Kitschelt and Malesky,
Political Science Association,
sented at the annual meeting of theMidwest
Chicago, April 2000).
50
based on their political base, see Barbara Geddes,
For a discussion
of elite policy preferences
in James E.
in Contemporary
and Institutional Change
Countries,"
Developing
"Douglass C. North
eds., Competition and Cooperation: Conversations with Nobelists
Levi, and Elinor Ostrom,
Alt, Margaret
about Economics and Political Science (New York: Russell Sage Foundation,
1999).
51
2 and 3.
For more detail on the definition of these and other variables, see Appendixes
52
in the Industrial Countries,"
and Budget Deficits
Roubini
and Sachs, "Government
Spending
in
of Budget Deficits
Economic Policy 8 (April 1989); and idem, "Political and Economic Determinants
the Industrial Democracies,"
European Economic Review 33 (1989).
53
Hellman
(fn. 3); EBRD (fn. 9).
PERILSOF POLARIZATION
325
I included variables that account for the influence of the Interna
tionalMonetary Fund, the European Union, and geographic proximity
toWestern
also included other variables
that scholars have
Europe.541
as the percentage
to
economic
such
of the
long-run
growth,
GDP
in
levels
of
size
of
the
the
education,
popu
produced
agriculture,
attributed
lation, life expectancy, and initial GDP per capita circa 1989.55Adding
not
in any substantive
fashion:
or
the
significance
changed
measure.
are
es
level
of
the
the
results
Indeed,
significance
polarization
from the base model.
sentially unchanged
these variables
none
Robustness
I assessed
country
determine
the
does
of these variables
statistical
Checks
the robustness
of the results
the analysis one
the results
whether
from
sample
the results
change
achieved
into
two
five-year
at a time
depended
periods
in several ways.
I
each
dropped
to
and reestimated
the model
on a
single
to see whether
case.
I then
split
the results were
driven by rapid declines early in the decade.56Neither of these changes
affected
the sign or significance
of the variables
of interest.57
I also ran the results dropping the fixed effects for countries, for
then for both countries
and years, and report the results on lines
so
of
and
Table
the
3.58 Doing
1, 2,
3, respectively,
only strengthened
measure.
In each of these specifications,
the
results for the polarization
years,
or
that changes
sign
only variable
significance
comes
in each model.
and significant
positive
is openness,
which
be
the
Thus,
impact of
openness is sensitive to the inclusion of fixed effects.When
the country-specific
fixed
effects,
I added
a
dummy
variable
I dropped
for mem
bership in the former Soviet Union.59 The coefficient on FSU{ was
54
The
Fund 2000 (www. Imf.org/external/np/tre/tad);
Randall
International Monetary
Stone,
Transition
(Princeton:
Lending Credibility: The International Monetary Fund and the Post-Communist
Princeton University
Press, 2002).
55
Martha DeMelo,
Cevdet Denizer,
and Alan Gelb, From Plan toMarket: Patterns of Transition,
World
Bank Policy Research Paper (Washington, D.C.,
1996); Vladimir Popov, "Shock Therapy ver
sus Gradualism: The End of the Debate
theMagnitude
of The Transformational
Reces
(Explaining
sion)," Comparative Economic Studies 42, no. 1 (2000).
561 had to drop the fixed effects for countries and years due to the small sample sizes in these esti
mations.
57
Measures
of corruption and institutional quality are not directly included in the model but are
often correlated with the wealth of a country. In addition, polarization may foster corruption and weak
institutions by heightening
incentives to lobby to prevent changes in policy from affecting your firm.
58
There
is great debate about the inclusion of fixed effects?dummy
variables for countries or
models
such as this. Some argue that including fixed effects reduces the potential for omit
years?in
ted variable bias, while others argue that there is little theoretical basis to include fixed effects and that
on Research
and Methods
in Inter
the cure is usually worse than the disease. See Symposium
Design
national Relations,
International Organization
55 (Spring 2001).
59
a
is a rather crude indicator of the institutional
The dummy variable
legacy of Soviet polity.
FSUit
aspects that set the former Soviet republics apart are captured in other variables, for example,
Many
326 WORLD
POLITICS
3
Table
Robustness
1. Polarization,
-.
continuous
16***
(without country-specific fixed effects)
2. Polarization,
continuous
(.06)
-.28**
(without year-specific fixed effects)
3. Polarization,
Checks51
continuous
(.12)
-.18***
(without either country or year fixed effects) (.05)
4. Polarization,
continuous
-.23*
lagged
(excluding reformed communist party (.12)
lagged by a year)
all CP lagged
5. Polarization,
-.15*
(including reformed communist party (.09)
lagged by one year)
6. Polarization,
-.57***
average
for three previous
years
communist
party)
moving
excluding
polarization
reformed
7. Effect
of
(average
lagged
growth
on
(.19)
-.004
polarization
_(.088)_
*p<.10;**p<.05;***p<.01
a
1-6 are OLS estimates
Entries
ticity
consistent
standard
year-on-year
change
2 are included
Table
not
presented.
Entry
errors
measures
for various
in
parentheses.
in 1998 $US. Note
in GDP
in these
7 reports
analyses,
the effect
but
The
that
of
with
heteroskedas
polarization
is the real rate of
variable
dependent
1 of
the remaining
variables
in model
to conserve
of lagged
growth
space
on
their parameter
polarization
estimates
including
the
are
in
dependent variables inmodel 1 of Table 2.
negative and lay just beyond the bounds of significance (p=.105). In
cluding
FSU.t
had no
impact
on the
significance
of polarization.60
I then examined the possibility that slow growth leads to political
rather than vice versa.
polarization
measures
in equation
of polarization
I lagged both of the continuous
1 by one year and report the re
so on. The effects of
GDP per
in the
capita, miles from Vienna, government
spending, and
membership
former Soviet Union
should also be captured by the country-specific
fixed effects. One problem with
the
variable is that it is constant over time. A somewhat more refined measure
includes a dummy
FSU.t
variable for each year that countries are in the ruble zone. This variable is not significant when added
to the base model.
60
a correction for
The size and significance of the results are essentially unchanged
using
growth
rates that takes into account the size of the informal economy. Marcelo
Selowsky and Ricardo Martin
calculate growth rates after increasing the GDP by the fraction
FSU countries) or
X./10 (for
xt/3 (for
other countries), where xit = the share of private sector output in GDP. They take their estimates for the
size of the informal economy from the EBRD's Transition Report (fn. 9). See Selowsky andMartin,
"Pol
and Output Growth
in the Transition Economies," American Economic Review Papers
icy Performance
and Proceedings 87 (May 1997).
PERILS OF POLARIZATION
suits on lines 4 and 5 inTable 3. Doing
as the coefficients
327
so did not materially affect the
on the
measures
lagged
I also measured
polarization
results,
their significance.
score over the three
polarization
is significant,
this
variable
Again,
retain
of polarization
by taking the average
years (year /, /- 1, /- 2).
previous
as is
on line 6 of Table
3.
reported
That the lagged value of polarization is a fairly good predictor of growth
rates
is evidence
that causation
flows
to
from polarization
growth.
In addition, I lagged the value of growth and placed it on the right
hand
side of equation
1 and
placed
the polarization
measures
on the
left hand-side of equation 1 and report the results on line 7 of Table 3.
The insignificant coefficient on the lagged value of growth also sug
that growth
is not causing polarization.61
there
is
that political
evidence
Thus,
strong empirical
polarization
is quite plausible. The polar
of causation
drives growth. This direction
in years preceding
ization measure
the measures
is taken from elections
gests
that bad
the argument
Moreover,
to
formance
leads
may
necessarily
political
polarization
cases. Susan Stokes argues that if voters
for
these
priate
of economic
growth.
economic
per
less appro
in a transition
be
believe
that the economy must get worse before it gets better,
economy
in the transformation
that
indicate
then economic
decline
may
early
encour
current
need not
is on track.62 Thus,
economic
decline
policy
voters
age
to
some
support
extremist
candidates.
Beyond
transition
such as Samuel
scholars,
argue that it is
Huntington,
rather than vice versa.63 In sum,
that leads to polarization,
rapid growth
are
and are not readily re
the roots of political
complex
polarization
duced to economic
performance.
economies,
61
To
the possibility of reverse causation further, I conducted an instrumental variable/two
the following
system of equations:
regression analysis by estimating
=
+
Polarization
$parliamentary
it ?0 ?^Populations
System it+ eit. (1)
=
+
+
GDPGrow.t
?0 fifiovernmentSpending.^
$2GDPperCapitai{t_v
+ e.. (2)
+
+
+
+
$AGDPGr
i(t-i)^Country.)
Wjear.)
?3Wari(t-i)
address
stage least-squares
In the first stage I use a continuous variable for the size of the population
and a dummy variable for
measure of
In
of a parliamentary
polarization.
regime as instruments for the continuous
the second stage, I include variables commonly found inmodels of economic growth. The effect of po
on
in this analysis at the .10 level. Because
litical polarization
growth remains statistically significant
are
of growth and polarization
(1) the number of cases is fairly small, (2) the determinants
poorly
and (3) it is difficult to find proper instrumental variables, there is reason to be skeptical of
understood,
Vari
in this setting. Larry Bartels, "Instrumental Variables and 'Quasi-Instrumental'
this technique
ables," American Journal of Political Science 35 (August 1991); Peter Kennedy, A Guide to Econometrics
(Cambridge: MIT Press, 1993).
62
and Market Reforms: The Limits of Economic Voting," Compar
Susan Stokes, "Public Opinion
the existence
ative Political Studies 29 (October 1996).
63
Samuel Huntington,
Political Order in Changing
Societies (New Haven: Yale University
Press, 1968).
WORLD POLITICS
328
III. Political
and Policy
Polarization
that political polarization hin
The preceding analysis demonstrates
dered
in the cases
economic
countries
growth
economic
growth
at hand.
declined
Volatility
In addition,
the
with
approach
in polarized
a
of elections,
finding that is consistent with the view that political polarization af
fects growth by increasing the possibility of a change in policy. In this
some
I examine
section,
with
the argument.
This
analysis
that
should
be consistent
another
means
of testing
evidence
ancillary
provides
the
implications of the theory and offers an additional opportunity to iden
tify the mechanisms
that drive
the argument.
Thus far, I have emphasized how political polarization intensifies the
credible commitment problem by making large swings in policy more
to this view,
increases economic
likely. According
polarization
political
to
in
economic
incentives
and reduces
engage
uncertainty
productive
If this argument
is correct, we
such as saving or investing.
behavior,
should expect countries with more polarized
systems to have
political
extent
in
liberalization.
the
economic
of
greater year-on-year
volatility
To
assess
this argument,
=
Policy Volatilityit
I estimate
the following
+
equation:
+
?0 ^Polarization^
?2Democracyit
+
+
?40/7lV
$3GovernmentSpendingit
+
+
+
+
$sGDPperCapitai(t_1} $6Warit ^Loglnflation^^
?sOpennesSi(Hlj
+
?9Fragmentationit
+
li?wCountryit)
+ Z(?uY^)
+ <>.,(2)
The dependent variable in equation 2, Policy Volatility i?tis the absolute
value of the percentage change in theWorld Bank Liberalization Index
country
i in year /.The
index consists
of three elements:
the extent
of lib
eralization in foreign trade, the extent of liberalization in internal prices,
Bank
and the size of the private sector, as calculated
annually byWorld
and
It takes a value of 0 for an ideal-type command
economists.
economy
the dependent
100 for an ideal-type
Because
free-market
economy.64
variable measures
the absolute value of the percent change in the index,
in economic
it includes both increases and decreases
liberalization.65
64
in foreign trade and
The Liberalization
Index includes two clear policy elements,
liberalization
is harder to classify as a policy ele
domestic prices. It also includes the size of the private sector, which
ment. Nonetheless,
the index does include progress in privatization, which is an important policy mea
sure. In addition, reassertions of state control over
would affect the size of the private
property?which
it is
transitions of political power in polarized countries. Thus,
sector?have
been common following
to include the size of the
sector in this index.
important
private
65
Index.
1 for data on annual average changes in theWorld
Bank Liberalization
See Appendix
PERILS OF POLARIZATION
329
The main independent variable of interest is political polarization,
which ismeasured with a dummy variable including the reformed com
munist
countries
Polarized
parties.
should
greater
experience
year-on
year volatility in policy.
Countries
with
robust
because
democracies
may
democracies
experience
institutional
less policy
constraints
volatility
place stronger
on
than do autocracies. Autocrats
policymakers
likely have greater op
to introduce
in
outside
of formal legal mech
portunities
changes
policy
are marked
a
anisms.
countries
whose
systems
Similarly,
political
by
large number of veto points may find itmore difficult to make rapid
in policy.66
Accordingly,
changes
I include
Fragmentation.
which
is a
version of the scale of political fragmentation developed by Roubini
and Sachs
that has been modified
In addition,
at war
countries
for the postcommunist
world.67
high levels of inflation may
or with
ex
to these conditions.
in response
I
volatility
in
2.
The
and Loginflation
wealth
.(t_1} equation
Warit
also influence
of policy. On
of a country may
the volatility
the one
GDP may have stronger
institu
hand, countries with
high per capita
as
to
such
check
the
and
courts,
tions,
bureaucracies,
parties,
political
perience
therefore
greater
include
policy
ability of politicians
wealthier
countries
greater capacity
a
country's wealth
Countries
with
to change policy quickly. On
have
the other hand,
resources
and have
greater control
course of
the
the
effect of
To
capture
change
policy.
on
I
include
policy volatility,
GDPperCapita.(t_1}.
high levels of government
spending may experience
may
to
over
to
less policy
because
the government
has the resources
volatility
smooth
the effects of the business
that
governments
cycle. Conversely,
more
to
tools and greater capacity
spend freely may have
change policy
measures
I thus include a variable
size of government
that
the
rapidly.
as a percentage
of GDP.
spending
I also add a dummy variable for the four countries that rely heavily
on
economies
whose
upon oil as a source of revenue. Countries
depend
in the interna
oil and gas may experience
fluctuations
due to changes
tional market
for crude oil. Alternatively,
the oil and gas sector may
exert
inordinate
preferred
influence
also
economy may
eign trade regimes
portant
to include
the
state, making
changes
from
their
to occur.
to the world
openness
Finally,
as countries with
liberal for
shape policy volatility
are more
to external
It is also im
shocks.
exposed
less
policy
over
likely
Openness.^
because
it is highly
66
in Political Systems: Veto Players
George Tsebelis,
"Decision-Making
British Journal ofPolitical
and Multipartvism,"
mentarism, Multicameralism,
67
Roubini
and Sachs (fn. 52); Frye, Hellman,
and Tucker (fn. 49).
collinear
with
the
in Presidentialism,
Parlia
Science 25 (January 1995).
330
WORLD POLITICS
Table
of economic
volatility
4
liberalization4
Model
Polarization
.62*
(.35)
-.40
Democracy
(.40)
Government
.02
(.01)
spending
1.58
Oil
(.99)
GDP per
capita
.0042**
(.0021)
-.04
War
(.41)
.07
Inflation (log)
(.10)
.99
Openness
(2.07)
.08
Fagmentation
(.17)
.53
Constant
(1.79)
N 167
R2_.53_
*p<.10;**p<.05;***p<.01
variable
"Dependent
Index.
Fixed
effects
index of economic
existing
use OLS
is absolute
for countries
and therefore
liberalization
level of economic
inWorld
change
are included
in the analysis
of annual
value
and years
liberalization
within
serves
regression with
errors
suggested by Long
fixed effects.
country-specific
but
Liberalization
are not
to control
a country. As
the correction
standard
Bank
for heteroskedasticity
and Ervine.681
also employ
reported.
for the
before, I
consistent
year and
The results from Table 4 suggest that countries with polarized polit
ical systems
experience
greater
year-on-year
variation
in the extent
of
economic liberalization. Polarizationit is both significant and positively
related to the volatility of theWorld Bank Liberalization Index. On av
erage, the policy volatility index is 62 percent higher in polarized coun
68
Long
and Ervin
(fn. 42).
PERILSOF POLARIZATION
tries. These
the more
are
at the .10 level with
significant
as well.69
of polarization
countries
somewhat
experience
higher
results
sive measure
Wealthier
expan
of policy
related
economy
significantly
a war,
is government
in
involvement
spending,
on oil.70
on Democ
the
coefficients
Similarly,
to the world
but openness
volatility. Nor
volatility,
to
policy
inflation, or dependence
racyit and Fragmentation.
on
policy volatility.71
indicate
levels
is not
that neither
has a significant
impact
importantly, political polarization is associated with increased
Most
policy
331
volatility
in the cases
under
study. These
are consistent
results
with the argument that policy volatility is amechanism bywhich polit
ical polarization
tional wisdom
affects
economic
on
the dynamics
In nonpolarized
countries
growth. They
of economic
also fit the conven
in the
liberalization
dominated
such
region.
by anticommunists,
as Estonia,
an initial
in
have
been
minimal
after
swings
policy
jump
turnover has been
in
Government
early in the transformation.
frequent
are committed
to
but as all the major political
Estonia,
creating
parties
some form of a market
for
with
considerable
economy
scope
private
in
been minimal.
property,
changes
subsequent
policy have
In
countries
such as
dominated
by the ex-communists,
nonpolarized
in theWorld
Bank Index have been min
annual changes
Turkmenistan,
imal throughout the 1990s, as liberalization has proceeded fairly slowly.
Similarly,
has made
according
slow but
to theWorld
Bank
Kazakhstan
Index, nonpolarized
toward a liberal economy.
steady progress
in
such as Ukraine,
countries,
By contrast,
polarized
in
has
fits and starts.
liberalization
Russia,
proceeded
Bulgaria,
In Ukraine
and
the
World Bank Liberalization Index fell sharply in 1993 after initial halt
ing steps toward amore liberal economy. In Bulgaria the index fell in
1994 and 1995, after the election of the Bulgarian Socialist Party. In
Russia
the economic
environment
far less liberal
became
in 1998.
In
deed, the only cases inwhich theWorld Bank Liberalization Index fell
significantly
in a given year have occurred
in
polarized
political
settings.72
69
on country rather than
are stronger
using robust standard errors with clustering
Again the results
the HCCM correction suggested by Long and Ervin (fn. 42).
70
The relationship between oil and policy volatility is sensitive to the coding of polarization.
71
are correlated at .45. Dropping
either from the analysis individu
and Fragmentation.
Democracyit
results are, however, sensitive to the coding of polarization.
ally does not change the results. These
72
these cases of back
The Liberalization
Index experiences
relatively few reversals. Nonetheless,
seem to be important.
the average year-on-year
increase in the Liberalization
Index?
Using
sliding
variable
in the regression
rather than the absolute value year-on-year
the dependent
change?as
occur ten times and take
place
produces different results. For individual elements of the index, reversals
the political system
in years inwhich
in Russia, Ukraine, Romania, Bulgaria, Belarus, and Azerbaijan
is polarized.
WORLD POLITICS
332
The
of the policy
uncertainty
in
environment
polarized
countries
has
also been reflected in high levels of market volatility. In highly polar
for example,
ized Russia,
more
than
25 percent
by
ber 1993, inwhich
the market
following
for privatization
vouchers
fell
in Decem
elections
parliamentary
price
reformist parties fared poorly.73 In addition, the
main
the Russian Trading
index for the Russian
equities market,
Sys
in the month
than one-third
tem, increased by more
following Yeltsin's
In contrast, par
first-round
elections.74
victory in the 1996 presidential
in 1996 and 1998 caused
liamentary elections in the Czech Republic
on the main
only small ripples
equities market
countries
with
systems
polarized
political
economic
environments.
volatile
IV. Implications
index, the PX-50. Thus,
seem to have
especially
and Conclusion
casts doubt on the two dominant
to the
evidence
My
approaches
poli
tics of economic
I find that
reform. In contrast to the J-curve argument,
and relations with
elite partisanship,
the strength of the executive,
the
European
and the IMF are not
Union
In contrast
to the
growth.
prescriptions
view, regime type and broad governing
In addition,
economic
growth.
growth
significandy
motivated
coalitions
related
to economic
by the partial reform
are also unrelated
to
is unrelated
to
geography.75
Observers of the postcommunist world have debated at length the
roots
of economic
in the region.
performance
Poor
economic
perfor
mance is typically attributed either to rent seekers supported politically
by large ex-communist
overly ambitious
posed
such analyses
are useful
but
or to liberal
who have im
politicians
that
suggests
policies.76 My work
is
It
the
combination
of
both
incomplete.
parties
economic
and anticommunist
each with
factions,
to
its version of
the other from implementing
power
prevent
rules of the game that has undermined
economic
the economic
growth.
or ex-communist
countries with dominant
anticommunist
Conversely,
relatively
sufficient
strong
ex-communist
factions
have produced
political
more
their
counterparts.
polarized
better
Indeed,
economic
performance
this argument
suggests
than
that
73
and the Limits
in David
of Credible
Commitment,"
Timothy
Frye, "Russian Privatization
Weimer,
ed., The Political Economy ofProperty Rights: Institutional Change and Credibility in the Reform
Planned Economies (New York: Cambridge University
Press, 1997).
of Centrally
74
in Russia (Ann Arbor: Uni
Timothy
Frye, Brokers and Bureaucrats: Building Market Institutions
and Mikhail Myagkov,
of the
"The Emergence
Press, 2000); Roderick Kiewiet
versity of Michigan
Private Sector in Russia: A Financial Market Perspective," Post-Soviet Affairs 14 (January-April
1998).
75
and Reilly (fh. 5).
Fish (fn. 5); Kopstein
76
Aslund
(fh. 7); Stiglitz (fn. 2); Mau
(fn. 22).
PERILSOF POLARIZATION
is a need
there
for more
clines
into how
research
have managed
that have marked more
to avoid
ex-communists
333
the countries
the prolonged
countries.
polarized
dominated
economic
by
de
This article contributes to the ongoing analysis of the relationship
between
munist
in the postcom
and economic
performance
the commonly
held view that eco
into question
reform go hand in hand in these cases.77 It is not
democratization
world.
It calls
nomic
and political
or autocracy
democracy
per se that drives
economic
growth,
but
rather
it is the balance of political power between the largest anticommunist
parties and ex-communist
It also
complements
parties that shapes economic
performance.
re
two influential
about economic
arguments
form in the region. First, it builds on thework of Hellman by identify
ing the political conditions that may underpin the "partial reform
equilibrium."78
economic
reform
In polarized
countries
the short-term
winners
from
to
incentives
the
have particularly
state,
strong
lobby
if only tominimize the possibility of a large swing in policy. Indeed, the
evidence presented here finds that countries with polarized political
systems have been likely to fall into the "slow growth equilibrium" de
scribed by Hellman,
while
anticommunists
dominate
in which
countries
have
largely
either
avoided
or
ex-communists
this path.79
In addition, it identifies the political conditions that lead reform
oriented
politicians
to
adopt
the
strategies
and
tactics
advocated
by
Shleifer andTreisman.80 Reformist politicians facing strong opposition
as Shleifer
and
parties may favor co-optation,
in
In
their
in
occurred
Russia.
Treisman
contrast,
argue
counterparts
will
these
eschew
tactical
because
settings
compromises
nonpolarized
can
on their
the costs of transformation
with
opponents
they
impose
from
ex-communist
little resistance.
Finally,
political
business
this article
business
also contributes
cycle.81
incumbent
cycle,
to elections
to
improve
In the
to
politicians
their chances
long-standing
treatment
debates
on the
of the political
rates
seek to inflate growth
prior
office. Despite
of retaining
the
standard
77
Nelson
(fn. 30); Aslund, Boone, and Johnson (fn. 6).
78
Hellman
(fn. 3).
79
The polarized countries also experience higher increases in income inequality than do nonpolar
ized countries. Data from seventeen countries collected by Branko Milanovich
reveal that the average
increase in income inequality between
1988 and 1994 was 46 percent in polarized countries and 31
countries. See Milanovich,
Income, Inequality, and Poverty during the Tran
percent in the nonpolarized
sition (Washington, D.C.: World
Bank, 1998).
80
Shleifer andTreisman
(fh. 7).
81
E. B. Tufte, Political Control of theEconomy (Princeton: Princeton University
Press, 1978); Doug
lass Hibbs, Jr., The American Political Economy: Macroeconomics
and Electoral Politics (Cambridge: Har
vard University
and Rosenthal
Press, 1987); Alesina
(fn. 16). The literature on the political business
cycle is vast and has evolved significantly over the past twenty years. The discussion here is abridged.
WORLD POLITICS
334
theoretical
advances
in this literature,
the empirical
record of the polit
ical business cycle has been mixed.82 Indeed, I find that in the post
rates in
fall as elections
systems
growth
polarized
in
these electoral
effects are
approach, whereas
settings
nonpolarized
on the incentives
absent. By focusing
of firms, the argument
advanced
here can account for this outcome:
the heightened
that sur
uncertainty
a
to
in
rounds elections
incentives
invest as
setting weakens
polarized
communist
elections
world
approach,
Political
munist
leading
has
polarization
world.
been
highest
as elections
to lower
in polarized
growth
settings.
in the postcom
slowed economic
growth
rates have
in polarized
countries
growth
In addition,
in years that are most
distant
from
an election
and decline
is a key mechanism
approach. Thus,
political
polarization
in the
that shapes economic
behavior
world. Whether
postcommunist
cases
is an open ques
these insights travel beyond
the postcommunist
old and new regime factions gen
political
struggles between
in democratizing
in Latin
similar levels of polarization
countries
or Africa? Has
America
similar eco
generated
political
polarization
tion. Have
erated
nomic
outcomes?
82
James A. Alt
1983).
These
and Alec
are
Chrystal,
questions
Political
for future
Economics
inquiry.
(Berkeley: University
of California
Press,
APPENDIX
Years
of
Polarization
(TraditionalExCom Parties:
20% Threshold)
Country
Albania
Years
Belarus
Bulgaria
Abs. Value
Polarization
Average
Polarization
(AllExChange in
Com Parties:
Continuous Liberalization
Index
20% Threshold) Measure
1991-95
1997-98
1991-95
1997-98
1991-92
1990-96
1990-98
1991-92
1990-96
1990-98
Croatia
Czech
Rep.
1990-92
1990-92
Estonia
Georgia
Hungary
Average
Annual
of
Armenia
Azerbaijan
1
1990-92
1990-92
1994^98
20.3
Average
Annual
GDP
Growth,
1989-98
.42
.29
-4.93
4.6
.59
8.1
.72
-7.68
18.0
.47
-1.49
30.9
.76
-3.95
11.8
.13
-2.16
4.4
.57
-.22
.48
-2.94
7.8
.73
-9.26
2.4
.17
-.31
-4.67
12
11.4a
Kazakhstan
Kyrgyzstan
Latvia
1990-95
1990-93
Lithuania
1990-95
1990-93
1993-96
0
.59
35
.94
-3.91
14.6
.58
-3.72
3.1
.59
-3.8
11.9a
Macedonia
1990-93
1990-93
15.7
Moldova
1990-94,1998
1990-94,
20.9
.60
-9.12
0
.29
1.8
20.8
.69
-3.04
25,5
.70
-6.26
4.9
.58
.34
0
.15
.24
-5.11
1998
1989
Poland
1990-98
1990-98
Romania
Russia
Slovakia
Slovenia
1989,1993
98
1990-98
1990-98
1992-98
31.8a
19.3a
1995-98
Tajikistan
1995-98
Turkmenistan
Ukraine
Uzbekistan
1990-98
1990-98
Mean
0
.44
-8.17
0
.43
-8.17
19.8
.69
-8.9
0
.75
-.77
10.8
.54
-3.80
(13.0)a
a
This
includes
all ex-communist
parties
336
WORLD POLITICS
2: Variables from Robustness
inText
Defined
Appendix
Checks Not
1. Presidential power. Based on Shugart and Carey (fh. 49), as adapted by
and Tucker (fn. 49).
Frye, Hellman,
2.
0
Fragmentation.
if country
a
/ has
noncompetitive
system
of govern
ment, allowing elites to make policy with few institutional or partisan con
straints (Azerbaijan, Kazakhstan, Turkmenistan, Uzbekistan, and post-1996
Belarus); 1 if i has a single-party parliamentary government or a presidential
government with majority support in the assembly (Lithuania in themid-1990s;
Ukraine under President Kravchuk; andMoldova under the Agrarian Demo
cratic Party); 2 if/ has a two-party or a divided presidential government (Russia,
Kyrgyzstan, and Poland during the coalition government of the Democratic
Left Alliance and the Peasant Party); 3 if / has a three-party government; and 4
a
if/ has
government
composed
of four
or more
parties.
3. Elite partisanship. It ismeasured based on the occupational history and
partisan base of the executive?the
president in a presidential system and the
a
minister
in
prime
parliamentary system. It takes a value of 1 for leaders who
held a top position in the communist party prior to 1989 and retained that po
sition after 1989 (for example, Karminov inUzbekistan); 2 for a leaderwho held
a lower-level party post in the communist era or held high office in the commu
nist
party
but
became
the
executive
running
as head
of
a reformed
communist
party (for example, Kwasniewski in Poland) or against an unreformed commu
nist party (for example, Yeltsin inRussia); 3 is for executives who did not hold a
party post in the communist era (for example, Klaus in the Czech Republic).
4. % GDP
from agriculture. EBRD (fn. 9).
5. Life expectancy. EBRD (fn. 9).
6. Size of the economy. EBRD (fn. 9).
7. Initial GDP circa 1989. EBRD (fn. 9).
8. Geographic proximity. Miles from Vienna
9. IMF. 1 for
ment
agreement,
each year under
and 0 otherwise.
an IMF
poverty
reduction
or structural
adjust
10. EU. 0 if no formal relationship with the EU; 1 if applied for member
ship in the EU; 2 if it has signed an interim agreement; 3 if it has signed an as
sociation
agreement
with
the EU
in a
given
year.
Appendix
Government
spending
3
Descriptive
Statistics
Mean (SD)
N
38.9
Minimum
Maximum
188
10
206
0
1
250
0
100
250
0
1
250
1
3
249
1099
12906
246
0
100
250
0
3
250
0
1
250
0
1
182
3.7
67.7
180
0
21
249
1152
495263
205
0
4
250
0
1
83
(11.4)
.47
Democracy
(.50)
.56
Openness
(.38)
.13
War
(.33)
Elite partisanship
2.1
(.92)
GDP per
capita
4267
(2246)
Communist
Party
43.45
(33.7)
European
Union
.84
(.99)
FSU
.60
(.49)
Oil
.14
(35)
% GNP from
agriculture
18.8
(13.7)
Presidential
GDP
power
8.9
(5.46)
35204
(74159)
Fragmentation
1.86
(1.17)
IMF
.38
(.48)
GDPGrow
-3.82
(9.89)
249
-52.6
14.2