control of the public service became the order of the day in the 1990 s. The language
of reform called for ‘‘responsiveness’’ by public servants to the needs and wishes
of ministers andWve-year contracts for top public servants were instituted to
reinforce the message. Pollitt and Bouckaert ( 2000 , 155 ) identify similar trends in
Canada, France, and Sweden.
The accountability of public servants for their management work is scarcely any
better. In theory, responsibility (for management) can be delegated to agency chief
executives, while accountability (for policy) remains with the minister. But this
distinction hinges on clear deWnitions of both policy and management and of the
respective roles and responsibilities of ministers, senior civil servants, and chief
executives. As the British Cabinet OYce ( 1994 , 24 ) observes, ‘‘it is not always
possible to clearly separate policy and management issues.’’ It also comments
that ‘‘some Chief Executives, especially the ones from the private sector, are very
conscious of being in what they consider to be a fairly precarious position.’’ Again,
similar problems occur in Australia and Canada (see Weller 2001 and Aucoin 1995 ).
Pollitt and Bouckaert ( 2000 , 157 ) dryly observe that ‘‘politicians have not been
spectacularly willing to relinquish their former habits of detailed intervention.’’
Allied to ministerial intervention, public management reforms have created an
‘‘anarchy of aggressive competitive accountability’’ that undercuts performance
(Behn 2001 , 216 ).
- 6 Holding Networks to Account
To the ambiguity of management reforms, we can also add the institutional
complexity of networks, which obscures who is accountable to whom for what.
For many governments, outside police, defense, and social security, there are
few policy areas where the centre has hands-on control and where a command
operating code might work. Governments work with and through many other
agencies; they manage networks, commonly referred to nowadays as partnerships.
As Mulgan ( 2003 , 211 – 14 ) argues, buck-passing is much more likely in networks
because responsibility is divided and the reach of political leaders is much reduced.
Agencies and special purpose bodies have multiple constituencies, each seeking to
hold them to account, and there is no system, just disparate, overlapping demands.
As Peters ( 1998 , 302 ) argues, ‘‘strong vertical linkages between social groups and
public organizations makes eVective coordination and horizontal linkages within
government more diYcult.’’ Once agreement is reached in the network, ‘‘the
latitude for negotiation by public organizations at the top of the network is
limited.’’ The brute fact is that multiple accountabilities weaken central control
(Mulgan 2003 , 225 ).
332 r. a. w. rhodes