Weingast B.R., Wittman D. The Oxford Handbook of Political Economy

Подождите немного. Документ загружается.

chapter 35

..............................................................

COMPARATIVE

PERSPECTIVES ON

THEROLEOFTHE

STATE IN THE

ECONOMY

..............................................................

anne wren

In his inﬂuential work on “Political Control of the Economy,” Tufte (1978, 104)

concludes that:

The single most important determinant of variations in macroeconomic performance from

one industrialized democracy to another is the location on the left-right spectrum of the

governing party...the governing political party is very much responsible for major macro-

economic outcomes—unemployment rates, inﬂation rates, income equalization, and the size

and rate of expansion of the government budget.

This statement is signiﬁcant in two respects. First, in its assignment of a high level of

control over important economic outcomes to the state. Second, in its identiﬁcation

of governing parties as the most powerful political actors in the economic policy-

making process. Reviewed nearly three decades later, the strength of these conclusions

appears surprising. In the period since Tufte wrote, both the instrumental capacity of

the state to shape economic activity and outcomes, and the extent of the inﬂuence

of elected governments in the policy-making process, have been the subject of debate

in the comparative political economy literature. In this chapter we will review these

debates, outline how they have inﬂuenced current thinking about the role of the state

in the economy, and suggest several important avenues for future research.

anne wren 643

We will begin by describing the challenges posed to the thesis that state actors

possess the instrumental capacity to engage in macroeconomic demand management

by the rational expectations revolution in economics; by arguments about the con-

tingency of the eﬀectiveness of government interventions in this area on the structure

of socioeconomic institutions; and, more recently, by considerations of the lim-

its placed on state capacities by the international integration of trade and capital

markets. We will then discuss the increasingly important literature analyzing the

capacity of state actors to intervene eﬀectively on the supply side of the economy

to create the conditions for growth, expansion, and stability. We will consider the

extent to which interventions in these areas also may be constrained by the structure

of existing socioeconomic institutions, and by processes of international economic

integration.

Finally we will outline how these debates lead to important questions about the

relative explanatory power of arguments, emphasizing the long-term economic im-

pact of states’ socioeconomic institutional structures, and those focusing on the more

immediate inﬂuence of governing political parties. We will suggest that the assess-

ment of the relative economic importance of state structures and government policy

has important implications for the future direction of the ﬁeld and, in particular, for

the choices which we make in modeling the impact of politics on the economy.

1 States’Instrumental Capacities

.............................................................................

1.1 The State’s Role in Macroeconomic

Demand Management

The ﬁrst critique of governments’ abilities to engage in successful demand manage-

ment emerged from the rational expectations revolution in economics. Building on

the work of Friedman (1968) it was argued that, in a monetary policy game between

governments and rational economic agents, governments could no longer be assumed

to control more than the nominal value of monetary instruments. Any anticipated

change in nominal values would be compensated for by instantaneously adjusting

wages and prices, so that real economic outcomes could not be diverted from their

long-run equilibrium values. This argument challenged claims of the existence of a

¹ Obvious reasons of space have compelled us to limit the number of research questions which we

consider. Among the literatures which we have been forced to exclude, but which clearly also bear

relevance to the issue of the state’s economic role, we would include the new and exciting literature

relating economic policy and performance directly to the structure of the state’s political institutions

(e.g. the electoral system) rather than the socioeconomic institutions considered here (some of this

material is covered in Iversen, this volume, and Persson and Tabellini, this volume); as well as the

literature on the constraints placed on politicians’ policy choices by the actions of autonomous, or

semi-autonomous, bureacratic agents (some of this material is covered in Huber and Shipan, this

volume).

644 the role of the state in the economy

stable negative relationship between unemployment and inﬂation rates—the Phillips

curve (Phillips 1958)—and as a result undermined the inﬂuential contributions of

Hibbs (1977)andTufte(1978) to the literature on the economic signiﬁcance of gov-

ernment partisanship. For an exploitable Phillips curve to exist, governments would

have to be capable of systematically “fooling” economic agents—a violation of the

rationality assumption (Lucas 1972; Sargent and Wallace 1975; and Franzese, this

volume, for a further discussion). Thus it was impossible for left- and right-wing

governments to simply “select” diﬀerent Phillips-curve locations in response to the

distributional preferences of their electoral constituencies, as postulated by Tufte and

Hibbs.

Later Alesina’s “rational partisan” model reintroduced the possibility of observing

real eﬀects of partisanship, as a result of institutional rigidities in wage and price-

setting, and of uncertainty regarding electoral outcomes (see Alesina, Roubini, and

Cohen 1997). Still, Alesina’s model predicted that the real economic eﬀects (in terms

of output and employment) of partisanship would be temporary (disappearing once

wages and prices adapt to the post-electoral environment). Only their eﬀects on

inﬂation were expected to persist.

Adiﬀerent line of argument, meanwhile, considered how the process of macroeco-

nomic demand management is inﬂuenced by the broader institutional environment

in which it takes place. Alvarez, Garrett, and Lange (1991), for example, identify a

non-neutral role for government in inﬂuencing the wage-setting behavior of politi-

cally powerful union movements. Centralized union movements are more likely to

internalize the eﬀects of their wage demands on the economy as a whole (Calm-

fors and Driﬃll 1988). They also have considerable institutional capacity to deliver

economy-wide wage restraint. However, the real beneﬁts to labor of delivering this

restraint are uncertain, unless credible assurances can be obtained that public policy

will encourage the productive reinvestment of proﬁts in the domestic economy,

and will safeguard the welfare of workers (Przeworski and Wallerstein 1982;Lange

1984).

This uncertainty implies that governments can potentially play an important role

in inﬂuencing wage behavior. Common knowledge about parties’ ideological and

electoral commitments, renders left governments more credible as guarantors of the

protection of labor interests. Thus, the argument goes, in a centralized environment,

left governments are more capable than governments of the right of encouraging

wage restraint and achieving superior growth and inﬂation performance. Under a

right-wing government, uncertainty about the distribution of the economic beneﬁts

of wage restraint will reduce the incentive for powerful centralized unions to act

cooperatively. In contrast, in a more decentralized environment, common knowledge

about the ideological and electoral commitments of left governments renders them

more likely to be taken advantage of by opportunistic fragmented unions, so that

their economic performance will tend to be worse than that of governments of the

right. Thus the appropriate role for the government, and the economic eﬀectiveness

of governments of diﬀerent partisan hues, is expected to vary depending on the

institutional context.

anne wren 645

Lange and Garrett’s “institutional partisan” model is quite successful in explaining

variations in cross-national growth performance in response to the oil shocks of

the 1970s. In recent years, however, in response to the increased trend towards the

delegation of monetary policy to independent central banks, a new set of arguments

has been developed which model inﬂation and unemployment rates as emerging from

the interaction between independent monetary authorities and the parties to the wage

negotiation process (unions and employers’ organizations) (Hall and Franzese 1998;

Iversen 1999).

Again these models are designed to show how the impact of policy intervention

can vary depending on the institutional context, but in this instance the government’s

role is more restricted. While it bears responsibility for the selection of the monetary

regime, it may have limited powers to alter it once it is in place. If its selection involves

the delegation of monetary authority to an independent central bank, for example,

this implies that in future iterations of the game, the principal political actors engaged

in the macroeconomic management of the business cycle will be central bankers, and

the representatives of unions and employers’ organizations. Their behavior, in turn,

is dictated by the system of incentives emerging from the institutional environments

in which they operate. In these arguments, macroeconomic outcomes are principally

determined by the interaction between the socioeconomic institutions of the state,

and the impact of their structures on the behavior of economic actors, rather than by

the direct actions of democratic governments.

Additional constraints on the ability of governments eﬀectively to pursue macro-

economic demand management policies are identiﬁed in the growing literature on

the domestic economic impact of trade and capital market liberalization. On the

monetary policy side, for example, it is argued that the ability of large ﬁnancial

institutions to transfer funds rapidly across borders creates a strong impetus for cross-

national convergence in policy outcomes (see, for example, Kurzer 1993;Moses1994;

Keohane and Milner 1996; Simmons 1998). Where exchange rates are ﬁxed, cross-

national convergence in interest rates will be induced by the free movement of capital

in search of the highest returns (Mundell 1961).Whereexchangeratesareallowedto

ﬂoat, states retain some capacity to engineer monetary expansions via currency de-

preciations, but such expansions are rendered less eﬀective by associated increases in

the cost of imports, and increase the prospect of economically damaging speculative

currency attacks (McKinnon 1988). Similarly, on the ﬁscal side, increased trade and

capital market openness creates a number of problems with the pursuit of counter-

cyclical deﬁcit-ﬁnanced expansions. First, trade openness reduces the eﬀectiveness

of such policies, since the multiplier eﬀect associated with domestic expansions is

diminished where large components of domestic demand are international. Second,

capital market openness increases their cost, since higher levels of debt lead to higher

expected rates of inﬂation and are thus likely to be associated with capital ﬂight unless

nominal interest rates are allowed to rise.

These arguments have given rise to an extensive empirical literature which attempts

to estimate the extent of these constraints in practice, and in particular, their impact

on patterns of monetary and ﬁscal policy-making across parties. The ﬁndings of this

646 the role of the state in the economy

literature, while mixed, tend to support the hypothesis that the capacities of govern-

ments in this area have been reduced by economic openness. While the evidence does

not indicate that the constraints on the ability of governments to pursue independent

monetary and ﬁscal policies are overwhelming, it does appear as though the size of

partisan eﬀects in these areas has declined in recent decades, and is associated in

particular with the increased liberalization of capital markets and the introduction of

ﬁxed exchange rate regimes (Oatley 1999;Boix2000). These ﬁndings provide some

support for the hypothesis that the feasibility of Keynesian demand management

policies favored traditionally by left-wing governments has been compromised by the

globalization process.

1.2 The State’s Role in Managing the Supply Side

of the Economy

In light of critiques of their capacity to engage in macroeconomic demand man-

agement, students of the politics of state intervention have turned their attention

to the range of supply-side strategies available to governments with which they can

inﬂuence economic outcomes. Boix (1998), for example, focuses on policies aimed

at providing the investment in physical and human capital needed for growth. He

points out that in this area social democratic governments have traditionally favored

interventionist strategies based on public sector-led investment in both physical and

human capital, placing particular emphasis on strategies aimed at upgrading the

skills of previously low-skilled workers in order to reduce market-based inequities.

Conservative governments, on the other hand, have tended to rely on market-based

strategies to maximize the incentives for private investment in both physical and

human capital, with less regard for their impact in terms of equality.

Iversen and Wren (1998), meanwhile, investigate the eﬀects of labor market policies

on the development of services sectors. They point out that processes of market

saturation and technical change, combined with increased exposure to trade with the

developing world, have signiﬁcantly reduced the relative size and economic signiﬁ-

cance of industrial sectors in the advanced economies.

Since this process of deindus-

trialization tends to be associated with signiﬁcant job losses in traditional industrial

sectors, they argue, a primary goal of governments in all advanced industrial democ-

racies has now become the expansion of service sector employment. Low capacities

for productivity growth in many areas of service production, however, imply that

employment expansion in these sectors is hard to achieve when these countries tie the

wages of sevice sector workers to those in more productive sectors. A trade-oﬀ exists,

therefore, between the potential policy goals of equality and private service sector

employment creation. The alternative, public sector, route to service employment

expansion comes at its own cost in terms of government spending, which must be

ﬁnanced by taxes or debt. This results in a “trilemma,” or three-way choice, between

² See Rowthorn and Rawaswamy 1997, 1998, for useful analyses of this process.

anne wren 647

the policy goals of service employment creation, equality, and ﬁscal restraint (low

levels of spending), whose structure implies that at most two of these goals can be

pursued successfully at the one time.

Iversen and Wren argue that social democratic, neoliberal, and Christian demo-

cratic parties have tended to select divergent responses to the distributional trade-

oﬀs contained in the trilemma. Social democratic governments, for example, tend

to prioritize the goals of employment creation and equality, at a cost of relatively

high levels of state spending (generally ﬁnanced by taxation). Neoliberal governments

have tended to embrace a more market-driven approach, prioritizing the goals of

private sector employment creation and ﬁscal restraint, at a cost of higher levels

of inequality. Finally, European Christian democratic governments have typically

combined the pursuit of the goal of equality with an emphasis on ﬁscal restraint,

leading to poor performance in the area of service employment creation.

Finally, an extensive literature exists on the role of governments in the development

of welfare states. Through the creation of regulatory welfare systems aimed at insuring

citizens against social risk, and, in particular, at guaranteeing their compensation for

the dislocations associated with economic change, governments (in conjunction with

employers and union organizations) have sought to engineer the economic and social

stability required for investment in physical and human capital, and for growth (Sinn

1995;IversenandSoskice2001; Moene and Wallerstein 2001;Mares2003). In doing so,

however, governments have also pursued their own redistributive goals, and this has

led to the observation of signiﬁcant variations in the types of welfare state policies

pursued by governing parties of diverging ideological backgrounds. In this regard,

the path-breaking work of Esping-Andersen (1990) identiﬁed three distinct “worlds”

of welfare capitalism emerging in the process of welfare state development: a social

democratic one emphasizing equality and market-replacing public service provision,

a Christian democratic one emphasizing occupational and income-graduated public

transfer payments, and a liberal one emphasizing market-based self-reliance supple-

mented by a limited needs-based transfer system.

State actors also face constraints in the pursuit of supply-side economic strategies,

however. Here again, the structure of socioeconomic institutions impacts the range

of eﬀective strategies from which they may choose. Consider, for example, govern-

ment policies aimed at reducing wage inequality. Recent research indicates that the

ability of governments to inﬂuence levels of equality is constrained by the structure

of wage-bargaining institutions. In particular, negotiated outcomes emerging from

centralized wage negotiations, as opposed to decentralized or sector-by-sector nego-

tiations, tend to result in more compressed wage structures (Iversen 1999;Wallerstein

1999).

Of course, governments also possess a considerable range of policy instruments

with which they can inﬂuence wage structures directly—for example, by setting

minimum wage and beneﬁt levels, or by imposing legal extensions on collectively

bargained agreements. However, the eﬀect of these policies will always be constrained

by the existing institutional environment. For example, in highly centralized environ-

ments, even neoliberal governments can be expected to preside over relatively high

648 the role of the state in the economy

levels of wage equality, unless they engage in a full frontal assault on the institutional

structure of wage bargaining (a challenging political task).

In addition to these institutional constraints, governments also face new challenges

in designing supply side strategies for an increasingly open economic environment.

The mechanisms through which the constraints of openness are felt on the supply

side are slightly diﬀerent from those aﬀecting governments’ capacities to engage

in demand management. Consider, for example, the supply-side strategies pursued

by social democratic governments in the areas of capital investment, service sector

development, and welfare state protection described earlier (see Boix 1998;Iversen

and Wren 1998; Esping-Andersen 1990). All of these strategies involve the pursuit

of public sector expansions, and the maintenance of relatively egalitarian incomes

structures. Critically, however, they have traditionally been ﬁnanced largely through

taxation rather than deﬁcits and, as such, have been less directly impacted by the

constraints on deﬁcit spending associated with the liberalization of capital markets.

This does not mean that the constraints of economic openness are removed in this

case; rather it means that they center around the eﬀects of taxation and labor market

regulation on economic competitiveness, and on the threat of investment capital to

“exit” the domestic economy in the face of policies which reduce the eﬃciency of their

investments (Kurzer 1993;Rodrik1997; Alesina and Perotti 1997).

Several scholars have pointed out that the grounds for arguing (as conventional

wisdom sometimes assumes) that these eﬀects will lead to a cross-national conver-

gence in favor of neoliberal outcomes are not particularly strong. First, as pointed

out by Garrett (1998), it cannot be assumed that the cost–beneﬁt analyses of capital

investors will lead them to reject all forms of public investment and social pol-

icy. The observation by endogenous growth theorists that the market is likely to

under-provide some of the public goods required for growth—such as investment in

infrastructure, and in the skills and health of the workforce—suggests positive

beneﬁts to capital of their public provision (Romer 1994;Barro1997). Given the

right institutional environment, therefore, social policy can generate positive

externalities—such as higher rates of productivity growth, and political, economic,

and social stability—which the owners of capital ﬁnd it cost eﬀective to fund.

Second,

several scholars have argued that the economic dislocation which openness creates

simultaneously increases citizen demand for welfare state protection from economic

uncertainty and risk. The development and expansion of welfare states in advanced

industrial economies in the 1960sand1970s can be interpreted at least partly as a

result of the responsiveness of state actors to these demands (Cameron 1978)and

there is ample evidence of the generation of similar political pressures in the current

era of globalization (Rodrik 1997).

The globalization process, therefore, does not rob state actors of the policy instru-

ments necessary for the provision of public investment or welfare. Rather it alters

the calculus of political decision-making over economic policy. State actors must

³ Indeed, new research has explicated the positive role played by employers in the development of the

welfare state (Swenson 2002;Mares2003).

anne wren 649

now weigh demands for redistribution and public spending against potential new

economic costs associated with the pursuit of interventionary policies—namely the

risks of reduced competitiveness and an increased risk of capital ﬂight associated with

high levels of taxation and regulation. The outcome of this process clearly depends

critically on the structure of political incentives facing state actors.

Empirical examinations of the extent of the constraints under which governments

operate in the pursuit of supply-side strategies have produced mixed results. The

most consistent empirical result in this area is the observation of a large degree of

structural continuity over time. In the areas of welfare and labor market policy, for

example, several authors provide convincing evidence of the long-term persistence

of social democratic, Christian democratic, and neoliberal “regime types” and their

continued inﬂuence on economic outcomes, in the face of increased economic open-

ness (Esping-Andersen 1990;Pierson1994, 2001;IversenandWren1998;Huberand

Stephens 2001).

The task of establishing the independent economic eﬀects of current government

partisanship and inherited institutional structures has proved more diﬃcult for

methodological reasons. High levels of correlation between many key variables—

such as, for example, the degree of centralization of wage-bargaining institutions, the

strength of social democratic welfare state regime “characteristics,” and the current

and historical electoral strength of social democratic parties—enhanced by the small

size of the sample on which most empirical investigations have focused (eighteen

or fewer OECD nations over a period of thirty or forty years), create problems of

multicollinearity in the conduct of cross-national regression analyses.

As a result,

measures of the independent eﬀects of theoretically relevant institutional variables are

omitted from many analyses. Unfortunately these problems have tended to restrict

our ability to separate the inﬂuence of democratically elected governments on eco-

nomic activity from that of the socioeconomic institutional environment in which

they operate.

2 State Structures and State Actors:

The Role of Governments

.............................................................................

In spite of challenges to Tufte’s original characterization of the nature of political

control of the economy, therefore, developments in the comparative political econ-

omy literature over the last three decades have served to reaﬃrm the instrumental

capacity of the state to inﬂuence economic outcomes. At the same time, as described

in the last section, analyses of government intervention on both the demand and the

supply sides of the economy have increasingly emphasized the contingency of their

eﬀectiveness on the structure of the socioeconomic institutions of the state. These

⁴ See Huber and Stephens 2001 for a careful discussion of these issues.

650 the role of the state in the economy

arguments follow a common logic which identiﬁes the inﬂuence of the structure of

socioeconomic institutions on the behavior of important economic actors (such as

employers and labor unions), and its resulting impact on their behavioral responses

to diﬀerent kinds of policies. Reﬂecting this emphasis on the signiﬁcance of socioeco-

nomic institutional conﬁgurations, several authors have identiﬁed powerful long-run

eﬀects of institutional structures on patterns of economic activity, which persist even

in an increasingly economically integrated international environment.

This emphasis on the persistence of socioeconomic “regime” types, and their

importance in accounting for variations in patterns of economic development and

performance, reaﬃrms the signiﬁcance of the state’s economic role. At the same

time, it calls into question the extent of the capacity of governing political parties

to inﬂuence economic outcomes independently. Perhaps the most comprehensive

set of arguments emphasizing the long-term signiﬁcance of states’ socioeconomic

institutional structures is contained in the new, and inﬂuential, “varieties of cap-

italism” approach developed by Hall and Soskice (2001).

Scholars adopting this

approach have identiﬁed the existence of complementarities between a broad range of

socioeconomic institutions concerned with both supply- and demand-side economic

strategies— including, for example, the institutions of union and employer represen-

tation; of education and training; of the ﬁnancial system; and of the welfare state. The

existence of these complementarities implies that some institutional conﬁgurations

are likely to be more economically successful than others.

Hall and Soskice identify two speciﬁc “varieties of capitalism,” both of which have

the capacity for strong economic growth, but which rely on very diﬀerent models

of socioeconomic organization. The liberal market economy (LME) relies on arm’s-

length exchange and formal contracting—in short, on the institutions of competitive

markets—to coordinate the activities of ﬁrms, and other socioeconomic actors. In

the coordinated market economy (CME), in contrast, economic actors rely more

heavily on non-market relationships to coordinate their activities. Cooperative equi-

librium outcomes are reached through repetitive interactions between economic

actors involved in long-term relationships with each other in a context of closely

monitored networks. An important diﬀerence between the CME model and the more

traditional competitive market model embodied in LME, is the CME’s reliance on a

set of appropriately designed and stable intermediate institutions—such as employ-

ers’ organizations, strong unions, and cross-shareholding networks—which allow

both the exchange of information, and the monitoring, and if necessary sanctioning,

of behavior, which are necessary to enable actors to make credible commitments to

cooperative outcomes.

The authors argue that comparative institutional advantages in diﬀerent types

of production ﬂow from these diﬀerences in organization, rendering LMEs more

competitive in the production of goods requiring ﬂexibility and radical innovation

(for example, in high-tech sectors such as biotechnology or semiconductors), while

CMEscompetemoreeﬀectively in the production of goods requiring incremental

⁵ Iversen (this volume) covers some related issues.

anne wren 651

innovation and the maintenance of high levels of quality in production over long

periods of time (for example, in many areas of consumer durables). They argue

that, given the apparently successful integration of both of these models into the

international economy, there are no grounds to expect cross-national institutional

convergence as a result of increasing international economic integration.

Within the “varieties of capitalism” approach the capacity of government actors

independently to inﬂuence economic outcomes, or the direction of economic de-

velopment, seems particularly tightly constrained. The argument suggests that the

economic eﬀectiveness of a given policy or institution is heavily contingent on the

structure of the broader institutional conﬁguration in which it is embedded. This

implies that government actors face a restricted menu of potential policy choices.

Attempts to engage in reform in one area of policy, without simultaneously reforming

the many other socioeconomic institutions with which it interacts, are likely to lead

to disimproved economic outcomes. Within this framework, there appears to be little

scope for governments to exert a signiﬁcant independent inﬂuence on economic

outcomes. Rather, states are expected to continue to diverge along their existing

equilibrium paths unless they experience a signiﬁcant external economic shock.

Attempts by governments to engage in more wide-ranging reforms in the in-

stitutional environment aimed at shifting a political economy entirely from one

institutional equilibrium to another, meanwhile, are likely to be politically infeasi-

ble given the vested interests of citizens and socioeconomic actors in the existing

system. Firms and workers base a range of investment decisions (for example, in

the development of their production proﬁles, or in education and training) on the

institutional environments in which they are operating. As a result, they develop

policy preferences which favor the support and improvement of existing institutions

(a strategy which promises to protect the value of the returns to existing assets) over

radical institutional change (a strategy which increases the uncertainty associated

with these returns).

For example, Estevez-Abe, Iversen, and Soskice (2001) argue that workers who

have invested in ﬁrm- and sector-speciﬁc skills through participation in the highly

developed systems of on-the-job and vocational training present in CMEs are more

likely to support high levels of employment and unemployment protection. This is

because they require insurance against the relatively high levels of risk associated with

speciﬁc asset investments of this type. (Workers with speciﬁc skills face greater threats

to their income from, for example, a sectoral shock, than those who have invested

in more general skills which can be easily transported across ﬁrms and sectors).

Meanwhile ﬁrms operating in CMEs, whose production proﬁles rely on workers

⁶ Note that, within this framework, it is possible for an economy to be moved away from its existing

equilibrium path by a signiﬁcant external shock. Under these circumstances the returns to existing

investments (or at least the uncertainty associated with them) may be suﬃciently altered as to allow the

implementation of more radical reforms. It is important to note, however, that the authors of the

varieties of capitalism volume do not consider that the increased economic openness of the current

period has created a shock of this type for either CMEs or LMEs. This is because, according to their

argument, both models have shown themselves equally capable of adapting to the new international

economic environment.