stant Professor
Department of Political Science
Encina Hall West
Stanford University
awren@stanford.edu











The author gratefully acknowledges helpful comments on earlier versions of this paper from
Peter Hall, Torben Iversen, David Laitin, Jean Oi , Paul Pierson and members of the Stanford
Comparative Politics Workshop. All remaining errors are my own.



Introduction

The most advanced economies of the OECD are currently experiencing significant economic
transformations along two dimensions. First, they are undergoing a process of de-
industrialization: shedding labor in manufacturing sectors, and relying more and more on
services sectors to act as the chief engines of employment growth. Second, in common with
many other nations at different stages of development, they are facing an increasingly integrated
international economic environment, in which cross-national flows of goods and capital are
greater than ever before.

In this paper I explore the implications of these changes for theories about the impact of
government partisanship on economic outcomes. In a globalized post-industrial economy do we
see significant economic effects of government partisanship ? And, if so, can we interpret these
effects in the context of traditional cleavages between left and right over economic outcomes
?

In recent years, hypotheses about the constraints placed on national governments by the
increased globalization of capital and trading markets, have suggested that the power of
governments to independently select policy and to direct patterns of economic performance, may
have been substantially reduced, if not removed. These arguments have provoked an important
empirical literature. The principal focus of this literature so far has been in the areas of
Keynesian-demand management and welfare state policy (See, for example, Garrett 1998; Oatley
1999; Boix 2000; Huber and Stephens 2001; Swank 2002).
2



In this paper, I take as a starting point Iversen and Wrens (1998) argument that the process of
de-industrialization and the transition to a service based economy creates a new set of trade-offs
between the policy goals of employment creation, equality and budgetary restraint (by which is
meant low levels of government spending on goods and services). The structure of these trade-
offs is such that at most two of these important policy goals can be pursued successfully at the
one time, creating what the authors label a trilemma (or three way choice) for governments.

The balance of existing evidence suggests that the de-industrialization process itself does not
stem primarily from economic globalization (although it is undoubtedly a contributory factor
(See Wood 1994). Rather it strongly influenced by internal processes of economic change (in
particular technical change and the saturation markets in many areas of consumer durables)
associated with high levels of economic development (Piore and Sabel, 1984; Appelbaum and
Shettkat 1994, 1995; Rowthorn and Ramaswamy 1997, 1998)

This does not imply, however, that the transition to a post-industrial economy is unaffected by
the globalization process. Rather, there are grounds to expect that, as in many other areas of
policy, governments face constraints in their selection of a response to the service economy
trilemma emerging from the increased globalization of capital and trading markets. In this paper,
I identify the nature of these constraints, and assess the extent of their empirical impact.

Iversen and Wren find evidence of significant variations in economic outcomes associated with
the trilemma across political economic regime types. Thus in countries with a strong political
tradition of social democracy, we tend to observe high levels of equality, coupled with high
3



levels of public service employment creation, at a cost in terms of high levels of public spending.
Strongly neo-Liberal regimes, in contrast, tend to combine high levels of private service sector
employment creation with fiscal restraint, at a cost in terms of low levels of equality. Meanwhile,
traditionally Christian democratic regimes, tend to exhibit high levels of equality and fiscal
restraint, at a cost in terms of low levels of service employment creation. These findings resonate
with observations elsewhere in the literature of the persistent divergence of socio-economic
institutional regimes in the face of economic globalization (Esping-Andersen 1990; Pierson
2001; Hall and Soskice 2001).

The very persistence of patterns of divergence in economic outcomes across socio-economic
regimes, however, indicates that institutional regimes may, in themselves, act as a significant
constraint on the power of current governments to independently direct the economy. An
important empirical task therefore, as highlighted by Huber and Stephens (2001), remains the
identification of the independent economic effects of government partisanship and of the
structure of inherited institutions in a globalized environment.

In this paper, therefore, I investigate the extent to which governments are constrained in their
capacity to respond to the trilemma of the service economy both by economic openness and by
the structure of inherited institutions. I find that while increasing economic openness and
inherited institutional structures both impose constraints on governments policy choices, the
economic significance of the latter is significantly greater. Moreover, neither constraint is
sufficiently strong to overwhelm governments capacities to execute independent choices in this
area. This is indicated by the existence of statistically and substantively significant effects of
4



government partisanship.

The other argument which I advance in the paper is that the distribution of partisan preferences
of the economic trade-offs contained in the trilemma, cannot be understood in traditional left-
right terms. Rather, the analysis of the effects of partisanship in a post-industrial economy
requires us to adopt a two-dimensional approach to the study of economic politics.

In the next section I will begin by briefly reviewing Iversen and Wrens (1998) argument about
the structure of the economic trade-offs faced by governments in a services based economy. I
will then show that an analysis of partisan responses to the trilemma requires us to move beyond
traditional conceptions of left and right and adopt a multidimensional approach to economic
politics. Finally I will identify the constraints under which governments might be expected to
operate in their selection of a policy response to the trilemma, and will argue that we should
expect to see significant economic effects of government partisanship in spite of the these
constraints.

In the rest of the paper I elaborate and test a model based on these hypotheses by means of
multiple regression analysis using a dataset covering 15 OECD countries over a twenty year
period (1975-1997). I find evidence of significant effects of partisanship in influencing levels of
equality, public service sector employment and total service sector employment over the period
studied, in spite of the apparent existence of constraints associated with trade and capital market
openness, and from the structure of inherited socio-economic regimes.

5



A Theory of Partisanship in Globalized, Post-Industrial Economies.

(1) The New Structure of Economic Trade-Offs.

Iversen and Wren (1998) argue that the process of de-industrialization and the transition to a
service based economy creates a new set of trade-offs between the policy goals of employment
creation, equality and budgetary restraint (by which is meant low levels of government
spending on goods and services). The structure of the trade-offs is such that at most two of these
important policy goals can be pursued successfully at the one time. In particular, the article
argues that as a result of the continuing productivity gap between manufacturing and many areas
of service production, solidaristic wage policies have the effect of inhibiting private service
sector employment creation. Thus a trade-off exists between the policy goals of equality and the
expansion of employment in the private services sector
1
. The alternative public sector route to
service sector expansion, meanwhile, is associated with a different kind of distributional cost
that is, the higher levels of public spending, taxation and/or deficits required to finance the public
sector wage bill. The structure of these trade-offs, therefore, suggests that the transition to a
6



1 It is important to note that societies were not faced with a trade-off between employment
creation and equality in the so-called golden age of manufacturing expansion in the 1950s and
1960s. In the golden age high rates of prod