positive and negative freedom. UK local government may have only limited
freedom from central control but it has, because of its capacity and size, consid-
erable freedom to do things and undertake initiatives. Indeed one of the great
conundrums of local governance comparison is that you get some local authorities
that have seized an agenda and run with it and done much to transform their
locality and others who have failed to make any impact. Looking at formal
structural diVerences only reveals part of the picture; there has also to be a focus
on how practices are put into place.
Given a concern with informal practice as well as formal structure, the most
fruitful search for a criterion to distinguish systems of local government would
appear to involve a focus on present-day characteristics rather than historical
legacies. The next issue that needs to be confronted is whether to focus on a single
factor or multiple factors in drawing up divisions. Single criteria do not seem
completely convincing and are more prone to shifts in patterns of behavior; that is,
to deterioration over time as eVective criteria. Thus, for example, some studies
have looked at how local governments in diVerent countries responded toWscal
crises (Pickvance and Pretceceille 1991 ) and produced useful insights, but, as time
andWnancial circumstances have changed, the distinctions are not sustainable.
Goldsmith ( 1990 , 1992 ) suggests that you could focus on the underlying ethos of
local government systems. Thus it could be that local government is understood as
part of a clientelistic or patronage system in which local leaders are seen as
defenders of their localities. Such a model might apply to southern Europe.
Alternatively local government might see itself as a promoter of economic devel-
opment and such an ethos is strongest in the United States, Canada, and Australia.
Finally local government might see itself as a welfare provider, and the British,
German, and Nordic systems would all follow that ethos. The trouble is that,
although there is some value in such a classiWcation, it is diYcult to sustain
given the breakdown of the more clientelistic model in southern Europe and the
mixing of welfare and economic development foci in other countries.
The most dominant form of classiWcation in comparative local governance looks
at local government systems as a whole and links together a range of factors.
According to Lidstrom ( 1999 , 103 ), ‘‘the most widely accepted and frequently
cited’’ is that provided by Hesse and Sharpe ( 1991 ). There are three main groups
according to this categorization: A Franco group that would include many of the
countries of southern Europe, an Anglo group based around the UK and Ireland
and to some extent the United States and New Zealand, andWnally a north
and middle European variant including the Nordic countries, Germany, and the
Netherlands. But it is diYcult to be entirely convinced by this classiWcation since
there are such big diVerences within each of the groups.
Page and Goldsmith ( 1987 ) and John ( 2001 ), where the focus is more narrowly
on Europe, adopt a similar classiWcation with a strong division between northern
and southern countries. Denters and Rose ( 2005 , 10 – 11 ), with a wider world focus,
comparative local governance 499