Measuring pro-poorness: does decentralisation help the poor?

Decentralisation may currently be popular, but is it as pro-poor as it is made out to be? Does it really bring responsive government closer to the people? Can we measure the differential poverty alleviation impacts of fiscal, political and administrative decentralisation? Should pro-decentralisation advocates guard against the risk that they are inadvertently harming those they seek to assist?

Research from the Institute of Development Studies applies confirmatory factor analysis to decentralisation data collected from 108 countries in 1996 to propose a measurement model for decentralisation. Arguing the need for greater clarity when analysing various aspects of decentralisation, it produces evidence that because the poor lack political and organisational resources at local level, they have a political advantage in the continuing centralisation of political and government institutions.

The researchers defined decentralisation in terms of the extent to which power and authority are devolved from central to any lower level of government. They also defined the pro-poorness of governments as the proportion of public spending devoted to health, education, social security, social welfare, housing and community amenities. Factor analysis confirmed that concepts of fiscal, administrative and political decentralisation are distinct in practice. Some countries are relatively centralised along one dimension and decentralised along another.

Standardising for the effects of other factors such as differences in national income per head, economic openness and demographic structure, the researchers analysed the relationship between each measure of decentralisation and the extent of public spending on social sectors. This indicated that governments of countries that are administratively decentralised tend to direct a higher proportion of their revenues to social spending while politically decentralised states spend little on social sectors.

Regression models were used to show that:

• There is no consistent, significant statistical association between the degree of fiscal decentralisation and pro-poor public spending.
• Administrative decentralisation generally appears to be positively associated with pro-poor public spending.
• Political decentralisation tends to have an erosive effect on the political capacity of the poor to assert their interests in relation to the rich.
• Decentralisation heightens a ‘race to the bottom’ competition for investment and forces local governments to pursue policies that favour the rich who are unlikely to invest in jurisdictions that threaten to introduce pro-poor policies.

The ideal combination to achieve the most pro-poor policy environment is to increase political centralisation while encouraging administrative decentralisation and the pro-poor efficiency effects associated with it. Policy-makers who wish to make decentralisation pro-poor are encouraged to overcome the current conceptual confusion and:

• think more rigorously about the multidimensional nature of decentralisation and in particular, pay attention to the political dimension
• realise that centralisation along one dimension can be combined with decentralisation along another
• strengthen the institutions that centralise political functions
• encourage national political parties, unions and interest groups and adjust electoral institutions to favour issues and contests with national scope
• encourage local autonomy to decide on policy packages, collect information and implement pro-poor efforts
• undertake research to ask why countries decentralise along one dimension rather than another and what impact decentralisation has on such social outcomes as participation, accountability and distribution.

Source(s):
‘Who Gets What From Whom? The Impact of Decentralisation on Tax Capacity and Pro-poor Policy’, IDS Working Paper 179, Institute of Development Studies, by Aaron Schneider, February 2003 Full document.

Funded by: DFID (SSRU R7616)

id21 Research Highlight: 17 February 2003

Further Information:
Aaron Schneider
Institute of Development Studies
University of Sussex
Brighton BN1 9RE
UK

Tel: +44 (0)1273 621202
Contact the contributor: a.schneider@ids.ac.uk

Mick Moore
Institute of Development Studies
University of Sussex
Brighton BNI 9RE
UK

Tel: +44 (0)1273 678689
Contact the contributor: m.p.moore@ids.ac.uk

To obtain copies of publications contact:
Gary Edwards
Publications Office
Institute of Development Studies
University of Sussex
Brighton BN1 9RE, UK

Tel: +44 (0)1273 678269
Fax: +44 (0)1273 621202
Contact the contributor: G.Edwards@ids.ac.uk

Institute of Development Studies (IDS), UK

Other related links:
'The political economy of Latin America: decentralisation the key to growth?'

'Decentralisation: not necessarily always a good thing?'

'Responsibility without power – decentralisation of primary healthcare in Chile'

'Decentralisation in Uganda: is the jury still out?'

'State reforms and the decentralization of the agricultural and rural public sector: Lessons from the Latin American experience'

'Distributive politics and the benefits of decentralisation' from CSGR

'ACTS Realities or Rhetoric? Revisiting the Decentralization of Natural Resources Management in Uganda and Zambia'