Using Aid to Prevent Conflict

Horizontal inequalities and aid policy

Aid can be a powerful instrument for helping to prevent conflict. One key way it can do this is by contributing to a fair distribution of resources across different groups in society. But today, this issue is largely ignored in aid policy. This briefing explains why inequalities between groups, or ‘horizontal inequalities’, need to be addressed in aid policy, and sets out how this can be done.

 

H

orizontal inequalities (HIs) are inequalities between religious, ethnic or cultural groups. Such inequalities may occur in economic, social or political dimensions, or all three. When large, and especially when consistent across dimensions, horizontal inequalities make conflict more likely.

Aid accounts for a very large proportion of government expenditure in poor countries. Potentially, therefore, aid donors have the power to make a major contribution to the development of an inclusive society by reducing horizontal inequalities.

Yet at present these issues are often ignored in aid policy, and the result is that aid sometimes serves to accentuate HIs.

Aid addressed to the reduction of HIs is in line with the Human Rights approach to development as well as the objective of eliminating social exclusion.

Aid policy can contribute to reducing horizontal inequalities by:

n Including the issue of group equity in all policy discussions with government.

n Including horizontal inequalities in the agenda of Poverty Reduction Strategy Papers (PRSPs).

n Promoting a more equal distribution of resources across groups in all public expenditure reviews.

n Using sums spent directly by agencies on social and economic projects which reach deprived groups and deprived regions.

n Reviewing macro-economic policies to consider their implications for horizontal inequalities.

n Increasing the allocation of aid to ethnically heterogeneous societies to help counter HIs.

n Supporting the collection of data on group inequalities and monitoring their evolution.

 

What are
horizontal inequalities?

Horizontal inequalities are inequalities between culturally defined groups, e.g. ethnic, religious, racial or caste-base groups. They encompass economic, social and political dimensions:

 

n Economic HIs include inequalities in asset ownership, employment and other economic opportunities, leading to inequalities in incomes.

 

n Social HIs include inequality in access to services, such as education, health and housing, associated with inequalities in health and educational outcomes.

 

n Political HIs consist in inequalities in the group distribution of political opportunities and power, including control over the presidency, the cabinet, the army, the police, and regional and local governments.

 

How to design aid policy to reduce horizontal inequalities

Aid is a powerful tool for reducing HIs. It accounts for a very high proportion of total government expenditure in poor countries generally, the countries most vulnerable to conflict. For example in Cambodia, aid exceeds total government revenue excluding grants and in Ghana it accounts for nearly 80%. And, of course, the influence of aid donors goes beyond their own expenditure to their role in developing country governments’ policies more generally, through discussions on macro-economic policy, public sector reviews, sectoral plans and via the PRSPs.

Donors are beginning to accept the need to include policies to reduce inequalities and exclusion in conflict-sensitive development (DFID 2007). Once correcting HIs is accepted as an important objective of aid policy, there are many ways that aid donors could contribute towards achieving this.

First, donors should monitor what is happening to HIs in the country. In some countries, there are good data and policies, but in many more information is not collected systematically on this issue. Aid donors should support collection of social and economic data on group lines, by including questions about group identity in censuses and household surveys in a sensitive way.  Imaginative use of existing data can give a good guide to the situation – in particular, regional data can often be used as a proxy for ethnic data.

Secondly, aid donors should include considerations of HI in all aid instruments, including PRSPs, public expenditure reviews (PERs), Poverty Reduction Budget Support (PRBSs), Sector Wide Programmes (SWAPs) and macro-policy design.

Correcting HIs can involve a wide range of policies, many of which are affected by aid instruments, especially where aid donors have broad discussions of policy with national decision-makers. A first priority is to sensitize such decision-makers to the need to consider HIs in policy-making. NGOs should also be aware of these issues. Appropriate policies to reduce HIs differ according to context, but relevant policies include:

n investment, credit and extension policies to help offset regional imbalances;

n education and health policies aimed at ensuring ethnic/religious and regional balance in access;

n policies that promote skills and training of disadvantaged groups;

n policies to outlaw discrimination including fair employment legislation;

n policies to help disadvantaged groups realize their legal rights, e.g. via legal aid;

n policies towards rights to land and natural resources to ensure balance in access and benefits;

n policies towards achieving equality in cultural recognition;

n policies to regulate the media to ensure equal access;

n policies towards civil society to encourage group interactions; and

n consideration of affirmative action policies where other approaches do not contribute enough to reducing inequalities.

This is a long list. In practice, early assessment of the country situation will reveal where the priorities are. For example, in Ghana, where there are large horizontal inequalities, it is apparent that the major problem is the economic development of the North and its integration into the Southern economy. This means infrastructure investment is a priority and assistance in generating productive opportunities in the North.

HIs need to be introduced in PRSPs.  Reviews of existing PRSPs show that ethnic or religious distributional issues are rarely reported upon, though this is beginning to change in conflict-prone areas. While gender equity is considered in the majority of cases, a review showed that ethnic minorities were mentioned in only a quarter of the cases examined. Cases in which ethnic minorities were not mentioned included heterogeneous countries at risk of conflict, such as Azerbaijan, Benin, Burkina Faso, Chad, Guyana, Malawi, Mali, Mauritania, Mozambique, Niger, Rwanda, Tanzania and Uganda (Stewart and Wang 2006). Group equity considerations should also enter into the decisions as to who to include in the participatory exercises.  As noted by Booth and Curran ‘the participation of minorities or indigenous peoples is either often overlooked or simply regarded as impractical due to their marginalization’ (Booth and Curran 2005:11).

HIs need to be included in any donor involvement in PERs and in discussions of PRBSs. This requires an assessment of the distribution of the benefits of public expenditure by group. Given data problems, this can often be approximated by regional data. The equity of group distribution of government expenditure can then be introduced as an explicit consideration along with the normal efficiency and poverty-reduction criteria.

SWAPs also require an assessment of the group equity of programmes, with requirements for improvement when large inequalities occur.

Macro-reform programmes have strong implications for distribution. Programmes generally favor tradables over non-tradables which can have serious implications for group distribution, sometimes so much so that they more than offset any public sector policies designed to correct such inequities. This is the case in Ghana, where the North is benefiting from deliberate attempts to increase social sector investment, but this is more than offset by economic developments favoring the rest of the country, so that North-South inequalities in incomes and poverty are increasing. Less than 1% of foreign direct investment goes to the North. The group distribution of macro-policies needs to be assessed and taken into account in the design of policies.

Finally, HIs need to be considered in country allocation of aid. Countries with heterogonous populations need more aid than homogeneous ones, because of the need to consider equity in resource allocation. Moreover, heterogeneous countries have been shown to have lower growth rates than homogeneous ones. Yet, at present, aid allocation favors homogeneous countries.

In the longer run, the objective of reducing HIs is compatible with other aid objectives, such as promoting growth and reducing poverty:

n It will assist growth, by tapping the potential of the deprived groups.

n Improved political stability will assist both growth and poverty reduction.

n It will contribute to poverty reduction goals by helping reach the most deprived, often difficult to reach with normal poverty reduction programmes.

 

A Human Rights approach

A Human Rights approach to aid and development offers a helpful general approach from the perspective of reducing HIs, while being grounded in natural law and international agreements. In particular, respect for minorities, policies to outlaw discrimination, and assistance for the greater empowerment of weak groups are all central to such an approach, while contributing to reduced group inequalities.

 

How aid can worsen horizontal inequalities

Aid is intended to promote growth and poverty reduction. It is frequently blind to issues of HIs. The impact of aid, consequently, can be to worsen horizontal inequalities where these do not form an explicit consideration in aid policy:

 

n Some programmes designed to favor particular groups contribute to widening inequalities. Transmigration programmes in Indonesia, supported by the World Bank and other donors, privileged Javanese and marginalized some local groups, worsening land and income distribution across groups. 

n Structural adjustment programmes are blind to these issues and can exacerbate horizontal inequalities where more privileged groups are in the favored tradable sectors and worse-off groups are concentrated in the subsistence and non-tradable sectors. In Mozambique, the gainers are the producers of cashews, cotton and sugar, mainly located in the Centre and South, and the losers the much poorer groups in the North.

n Projects and programmes to reduce poverty are often focused on the areas that are easiest to reach – those around the capital city – and remote areas tend to be neglected, worsening horizontal inequalities. This is the case in Bolivia, Peru and Guatemala.

n Processes of implementation often lead to a bias in the benefits of aid distribution, as particular groups control resource flows. A study of a capacity-building programme for Kenyan civil servants found that, in practice, it was biased by senior officials in favor of their own ethnic group (Cohen 1995).

n Especially where class and ethnicity overlap, the channeling of aid through privileged, educated intermediaries can unwittingly compound prevailing discriminatory attitudes and practices.

n The current country distribution of aid favors ethnically homogeneous countries over heterogeneous ones, partly because large countries (which have greater heterogeneity) tend to receive lower aid flows per capita.

 

The link between

horizontal
inequalities
and conflict

 

Horizontal inequalities can be a major source of conflict in both developed and developing countries. Group inequalities provide powerful grievances which may be used to mobilize people politically, especially when a socio-economically deprived group is also without political power. The conflict in Côte d’Ivoire, the long wars in Guatemala and Sudan and the rebellion in Nepal are all examples:

n In Côte d’Ivoire, Northerners had two-thirds of the years of education of Southerners in 1998/9 while their infant mortality rate was 30% higher. Political inequalities were added to these socio-economic inequalities when the Northern candidate was debarred from the presidential candidacy in 1995 and 2000. Northern grievances were articulated in an anonymous document, the ‘Chartre du Nord,’ pointing to the inequities in resource distribution. Both political and socio-economic HIs were thus factors behind the violent rebellion which broke out in 2002.

n In Guatemala, the indigenous peoples have been badly treated and subject to acute deprivation from colonial times.  Even today, their literacy rate is only 70% of that of the non-indigenous and their infant mortality rate is 56% higher. These strong inequalities, accompanied by almost complete political exclusion, underlay the massive conflict which lasted for more than two decades.

The commission which investigated the conflict found that the historical root causes of the conflict related to the ‘exclusionary, racist, authoritarian and centralist’ characteristics of the state, economy and Guatemalan society (Comisión para el Esclarecimiento Histórico (CEH) 1999: 81).

n In Sudan, severe and longstanding HIs were clearly an element in the decades-long North/South conflict and are now relevant to the Darfur situation.

The ‘North’ region, comprising about 5% of the population, has exerted a firm grip on political power since independence. It accounted for between 60% and 89% of ministerial appointments for most of the period from 1956 to 2000, with that figure never falling below 47%. Gross primary enrolment in the South in 2002 was just 11.6% in 2002, compared with 82.4% in the North and 45.5% in Darfur. The infant mortality rate is systematically higher, and literacy lower, than the North throughout the country, with the worst situation in the South. Fiscal transfers, far from offsetting these HIs, actually reinforce them. Development expenditure per capita in Darfur was just 17% of the national average in 1996-2000, while no data at all were published for the South. 

Regional redistribution of revenue and expenditure is an essential aspect of any post-conflict reconstruction, which was recognized in the 2005 peace agreement, and the issue is now one of implementation. The need for redistribution has become more acute with the discovery of oil resources, mainly located in the South.

n In Nepal, sharp caste and ethnic inequalities overlap with severe regional inequalities.  Mean years of schooling among the top Brahmin caste and the ethnic Newar were double the national average in 1996, and their average incomes 30% and 60% higher than the national average, respectively. In the case of the Newar, who are mostly located in and around the Kathmandu valley, this performance largely reflects equally serious regional inequalities. While development policies have attempted to redistribute towards impoverished regions, key political positions remain dominated by the top castes and the Newar ethnic group. Between them, the Brahmin and Chhetri (the second-ranked caste) and the ethnic Newar accounted for over 95% of recruitment into the civil service in 2001, despite constituting only around a third of the population.  Parliament is also dominated by the same three groups – a situation which in fact worsened following democratization in 1990. n

References

Booth, D., and Z. Curran. 2005. Aid instruments and exclusion. Report for the UK Department for International Development. London: Overseas Development Institute.

 

Cohen, J. M. 1995. Ethnicity, foreign aid and economic growth in sub-Saharan Africa: The case of Kenya. Development Discussion Papers No. 520. Cambridge, MA: Harvard Institute for International Development.

 

Comisión para el Esclarecimiento Histórico (CEH). 1999. Guatemala, Memoria del Silencio. Available online at: http://shr.aaas.org/guatemala/ceh/mds/spanish/toc.html.

 

Department for International Development (DFID). 2007. Conflict Policy Paper. London: HMSO.

 

Stewart, F., and M. Wang. 2005. Poverty reduction strategy papers within the human rights perspective. In P. Alston and M. Robinson, eds., Human Rights and Development. Oxford: OUP, pp 447-474.

 

Centre for Research on Inequality, Human Security and Ethnicity (CRISE)

Queen Elizabeth House

University of Oxford

3 Mansfield Road

Oxford OX1 3TB, UK

Tel +44 1865 281810, Fax +44 1865 281801

 

This policy briefing is based on CRISE Working Paper 36: The
Implications of Horizontal
Inequality for Aid
by Graham Brown and Frances Stewart.
Available online at:
http://www.crise.ox.ac.uk/pubs/
workingpaper36.pdf