Democratic welfare states the most consistent configurations of local government
institutions. Yet these classifications have typically relied on historical
and cultural classifications rather than on consistent analytical
criteria and have arrived at varying conclusions about how distinctive
local government in these countries is. For Michael Goldsmith and
Edward Page (1987) and Jens-Joachim Hesse and L.J. Sharpe (1991), the
four Nordic countries share a “Northern European” model of local government
with such other countries as Austria, Germany, the Netherlands,
and Switzerland. This model differs from Napoleonic systems that rely on
administrative centralization but are politically decentralized. Other classifications
derived by Anders Lidström (2003) and Robert Bennett (1993)
from historical differences treat the local government systems of Nordic
countries as a type distinct from other Northern European systems. In a
rare deductive classification, Goldsmith (1992) goes so far as to identify
the type of local government in the Scandinavian countries by the delivery
of welfare state services. However, he classifies local government under
very different welfare states in this category as well and offers no explanation
of the relation between welfare states and local government in other
countries.
These existing typologies suggest a more consistent relation between
the Social Democratic welfare states and a distinctive type of local government
than under other types of welfare state types.1 Yet these typologies
of local government ultimately require recasting in more analytical,
transparent metrics. A more precise comparison will also enable a closer
analysis of local government systems and their relation to welfare states.
Local Capacity and Supervision in the Infrastructure of Local
Governance: An Indicator-Based Comparison
A comparative classification of local government can start from the many
specific indicators that have increasingly become available in parallel form
for all advanced industrial countries. The following comparison will build
both on quantitative indicators and on qualitative ones coded in quantitative
terms. These indicators, encompassing fiscal as well as political and
administrative dimensions of empowerment and supervision, enable a
more systematic view of the ways that these characteristics of local government
in Social Democratic welfare states compare to those of other
Organisation for Economic Co-operation and Development (OECD)
countries.
Distinct from federalism or such efforts to measure decentralization as
Lijphart’s, these indicators focused specifically on the local level of government
in relation to those at higher levels. Relations of localities to the
federal and central units within federal states could thus be classified in
terms of supervision and capacities in the same way as local relations with
central governments in unitary states. Along with municipalities, the units
classified as local encompassed the somewhat wider scales of government
at the county level in such settings as the Nordic countries, the United
Kingdom, the United States, and Germany.2
Quantitative data made it a straightforward matter to derive comparative
indicators for several dimensions of these concepts. An array of qualitative
indicators captured national institutional variations along principal
dimensions of local capacities and local supervision. Where coded qualitative
assessments were largely derived from existing secondary literature
or potentially subjective assessments, an online working paper explained
individual results and gave detailed citations by country (see Sellers 2006).
Where possible, these results were checked and verified through a blind
duplicate coding procedure.3 For a number of other indicators, including
institutional ones such as the legislative constraints on local taxation, the
indicators had to assimilate results for federal units into a parallel indicator
to those for countries with a single, unitary higher-level government.
For this purpose, results for federal units were generally averaged, with
equal weights given to the result for each federal unit.4
Each aggregated index can be considered a formative measure of a
common concept. Unlike a reflective index, which presumes that differences
among individual indicators might be because of measurement
error, a formative index of this kind takes each indicator to capture a
distinct dimension of a general property. Such an index depends for
validity on component indicators that capture all the relevant dimensions
of the concept being measured (Adcock and Collier 2001, 538; Edwards
and Bagozzi 2000). Care was thus taken to encompass as many dimensions
as possible of the politico-administrative and the fiscal dimensions generally
considered crucial to both local capacity and supervision.
To assure “construct validity” (Adcock and Collier 2001, 537), individual
measures must be commensurable and receive proper weights in
relation to each other. Each quantitative indicator was standardized on a
scale from 0 to 2, where 2 measured the highest level of local capacity or
supervision. Qualitative indicators were standardized to a parallel 0–2
scale. Aggregation proceeded by averaging the indicators for a given
category. To avoid privileging any single dimension, each individual indicator
received equal weight. For similar reasons, the aggregated politicoadministrative
and fiscal dimensions of empowerment and supervision
were also weighted equally in the overall index.
Local Capacities
The politico-administrative and fiscal dimensions of local capacity vary in
largely parallel ways. Most notably, these indicators point consistently to
stronger local government capacities in the Social Democratic welfare
states than in practically any other OECD countries (Table 3).
As the clearest measure of formal institutional guarantees for
local authority, an indicator classifies the many constitutional textual