An international programme of research funded by the UK Department for International Development is currently posing such questions in Argentina, Côte d'Ivoire, Ghana, Kenya, India, Sri Lanka, Venezuela and Zimbabwe. All eight States are learning to act as regulators and enablers rather than direct providers. How do they score?

The baseline hypothesis of the Role of Government in Adjusting Economies programme is that - far from making government easier - these new roles may often require additional skills and capacity on the part of officialdom. Governments face technical, political and administrative difficulties never before encountered, as they strive to support a wide variety of new forms of service provision in which they now hold only a tenuous stake.

Research findings indicate some of the lessons already starting to emerge from the multi-country studies. Research teams from five UK universities are examining new forms of government intervention in four sectors; urban water supply, clinical health services, agricultural marketing and upgrading textiles enterprises in all focus countries.

On the question of how far government is prepared to change the way it organises itself in response to the new demands, researchers encountered mixed responses. In the economic sectors under study (agricultural markets and industrial development) liberalisation was much in evidence, yet the organisation of government action rarely turned out to have changed as radically as might be expected. Obstacles were:

• widespread persistence of practices of central control even where there has been a nominal shift to decentralised management and contracting-out of services
• poor understanding of (or unwillingness to accept) attempts to put public or private provision of services onto a performance-oriented contractual basis.

Proposed reforms seem to be most radical in countries where there is least capacity to block them. Thus proposals for reform often appear more radical in Africa, where bureaucracies tend to be less elaborate, than they might seem in South Asia. But though arguments for "de-bureaucratising" the State are stronger in South Asia, that region's bureacracies may - perversely - offer more stubborn resistance to role reform, even where the proposed change of role appears only moderately radical.

Successful examples of new arms-length relationships between sector and State include agricultural marketing in Sri Lanka and industrial advice and promotion in Ghana and Zimbabwe. There are also cases where unreformed public sector organisations score high levels of performance, such as Ghana's Cocoa Board. Most cases examined, however, exemplified weak performance with slow and uncertain adoption of reforms, the latter often existing more on paper than in practice. Some of our research, notably in the social sectors (health and urban water supply) appears to indicate that it can make better sense to put effort into boosting direct provision of services by government agencies than into inventing complex new contractual deals.

Situations that appear most likely to cast doubt on the effectiveness of government are those involving indirect administration (enabling and "market-friendly" regulation), pricing and charging in non-market conditions, and managing contracts for service outputs, such as certain social services, that cannot be quantified. Scenarios more favourable to good performance include rule enforcement by official agencies, the contracting-out of services whose outputs can be quantified, and direct administration. Organisational factors that seem to support good performance in the context of the "new", modes of provision are:

• management autonomy of direct providers
• business orientation on the part of managers
• financial autonomy of subcontracting parties
• reliable and adequate pay and incentives
• clear and foolproof contracts.

Key institutional factors are:

• authentic willingness by politicians and officials to retreat from management
• a background of effective civil service reform
• detachment of direct provider ventures from civil service structures
• ample resourcing, responsive money markets
• clear property rights, consistent legislation.

Richard Batley
Development Administration Group
School of Public Policy
University of Birmingham
Edgbaston
Birmingham B15 2TT
UK

T: +44 (0)121 414 4985
F: +44 (0)121 414 4987

E: R.A.Batley@bham.ac.uk
URL: http://www.bham.ac.uk/DevAdmin