Presentation on theme: "Institutional Determinants of Poverty: The Case of Kenya By Jane Mariara Godfrey Ndeng’e Domisiano Mwabu."— Presentation transcript:
Institutional Determinants of Poverty: The Case of Kenya By Jane Mariara Godfrey Ndeng’e Domisiano Mwabu
Introduction Kenyan regarded as an African success story early into the post-independence years of many African countries impressive growth between 1960 & 1970 (6.1%). But declining growth rates thereafter lowest achieved between (1.5%) Gradual recovery ( ) current is 5.8% Due to poor econ. performance, increased income inequality and access to basic services, 13.6 million Kenyans in 2000 lived under the poverty line. A high of 17 million or 56% of the population in 2005 under the poverty line.
Introduction cont’d Government’s commitment to fight poverty dates back to independence 1st Sess. Paper focused on the elimination of poverty, disease and ignorance. Various development plans and sectoral plans thereafter targeted poverty reduction and growth. Recent initiatives A number of policy and strategy papers geared towards achieving broad-based sustainable improvement in the welfare of all Kenyans National Poverty Eradication Plan (NPEP) Poverty Reduction Strategy Paper (PRSP). The Economic Recovery Strategy Paper (ERS). The Millennium Development Goals (MDGs)
Motivation Achieving the ERS and MDGs in Kenya is an uphill task & performance towards realizing the goals is still low. The failure to drastically improve the country’s investment and savings record threatens the recovery effort. Increasing studies on poverty in Kenya but a dearth of empirical studies on institutional determinants of poverty. Objectives Paper aims at investigating institutional determinants of poverty in the context of growing inequalities, increasing absolute poverty, and challenges in achievement of the ERS and MDGs targets. Also aims at suggesting a pattern of investment portfolio that is likely to have the greatest impact on poverty reduction.
Methodology and Conceptual Issues Poverty analyzed in the context of institutional structures prevailing in different geographical regions Institutions defined as formal and informal rules that govern behavior of economic agents (North, 1991). –Formal institutions by laws governing licensing of businesses; inter-regional movement of commodities; property ownership and sale, particularly land; establishment of order and peace in a region… –Informal institutions customs and social beliefs and norms in a region.
Methodology and Conceptual Issues (cont..) Institutions can also be broadly conceived as encompassing organizations (Putnam, 1993; Platteau, 1994) Institutions may also include public utilities and social capital. Following Putman and Platteau, we focus on organizations such as cooperatives, marketing boards, schools, health facilities, courts, and police stations (Table 1) No distinction is made between social capital and institutions (Putnam, 1993) because of data limitations.
Table : Type of Institutions & associated data sources InstitutionsProxy VariablesData Source Land tenure system Proportions of land under private and public /government ownership Statistical Abstracts, Economic surveys Social infrastructure Roads, electricity, health, education and water facilities Statistical Abstracts, Relevant Ministries CooperativesNumber of active cooperatives including coffee and tea cooperative societies; marketing boards e.t.c Relevant Ministries Law order & governance Courts, police posts, prisons, provincial Administration Relevant Ministries MarketsMarket centers: towns and other urban Centers Statistical Abstracts Legislative Affairs Number of parliamentary constituencies Constitution of Kenya Review Commission
Methodology and Conceptual Issues (cont..) There are generally two approaches to the analysis of poverty determinants. Probabilities of being poor estimated using logit or probit procedures, based on the FGT measures of poverty. Household welfare functions (expenditure functions) estimated using least squares methods. There are issues with both approaches but the two methods can equally well explain poverty (see main paper) Both approaches are used in this study to model poverty Poverty measures are based on CBN absolute poverty lines (see paper for choice of poverty lines) Our innovation is to introduce a vector of district level institutional variables as determinants of welfare.
Data I Distribution of Institutions Market Institutions: Proxied by number of municipalities, county councils and all towns. –Cooperatives also included as a proxy because most cooperatives deal with marketing and processing of goods and services Land Ownership: measured by trust land e.g: national parks and other reserves. Road Infrastructure Per capita roads network, measured by different road types. Law, Order and Governance System Proxied through the number of constituencies, (parliamentary representation), the number of administrative divisions, number of prisons, courts and police stations. Social Services: education inputs, health institutions & public expenditures
Data I Distribution of Institutions (Cont..) Analysis of the regional distribution of institutions reveals wide disparities in institutional endowments. The analysis implies that some regions less endowed with key institutions per capita have relatively lower welfare. Coast province is at a relative advantage in endowment of all institutions except education services. Western province is clearly at a relative disadvantage. Nyanza (poorest province) , no evidence that it is the worst in terms of institutional endowments
Data II Survey Data Welfare Monitoring Survey III (1997) data. Contains data on socio economic characteristics of the household, economic activities & time use, household asset endowments, consumption, income e.t.c. To this data, we map in district level data on institutions. Data suggests marked differences between characteristics of the poor and non-poor –The poor had significantly lower levels of post primary education than the non-poor and had larger families. –Poor households had higher dependency ratios than non poor households. –The poor were more concentrated in activities of lower economic status
Results and Discussion 3 different series of regressions: –District level model of institutional determinants of poverty –Household level determinants of poverty –Household and institutional determinants of poverty. Poverty measured through FGT measures, adult equivalent monthly expenditure and food poverty status All the district level models were estimated using ordinary least square regressions with district poverty means estimated from binary variables as dependent variables
District level Institutional correlates of poverty The number of constituencies per capita unexpected sign implying that parliamentary representation may not be an important correlate of poverty. The ratio of public to private trained secondary school teachers is welfare improving Health institutions are important correlates of welfare. Total area under water strong positive impact on the poverty rates, implying that districts with a lot of water experienced higher levels of poverty than their counterparts with less water. This could be explained by the fact that such districts have lots of waste water (such as large flood plains/wet grounds) which adversely affect productivity and welfare.
Household level determinants of poverty Age and age squared have an inverted U shaped relationship with household welfare, households become worse off with age of household head. Female headed households were poorer than male headed households Larger households are likely to be poorer than smaller households. Education was a significant correlate of household poverty and welfare was an increasing function of education attainment. Households headed by formal sector workers were better off than households in lower status employment.
Household level determinants of poverty (cont…) Distance to source of water associated with higher levels of poverty, implying that households that had to spend a lot of time collecting water had less time for productive activities and thus lower welfare Number of rooms in a house and number of total livestock units owned were negatively correlated with household poverty rates, implying that assets are important correlates of poverty. Other than for Central province, all other regions were likely to have higher poverty rates than the Rift Valley province.
Household & Institutional Determinants of Poverty Combining institutional and household level determinants significantly improved the fit and stability of the models. The results for household characteristics were consistent with those of regressions with household characteristics alone. The impact of land holding was reversed in the expenditure function while livestock ownership had the expected and significant impact. Signs and magnitudes for coefficients of regional dummies were significantly different from those in the household characteristics model. Household characteristic results for rural and urban areas were also consistent with the household variable only model.
Household & Instit. Correlates of Poverty (cont..) Per capita endowment of active cooperatives, health centers and ratio of public to private school teachers was found to be negatively correlated with absolute and food poverty rates. Government land and total length of earth roads were positively correlated with expenditures and therefore lowered the likelihood of households falling into absolute poverty. Overall the results support the expectation that institutions are important correlates of poverty, but parliamentary representation did not seem to matter
Policy Recommendations Specific interventions targeting poor households: Employment creation and agricultural sector growth should also be prioritized as outlined in the ERS. Targeting livestock producers and pastoralists through extension services, veterinary services, markets and reduction of input prices. Support rural water development programmes targeting areas where access to water is a major problem. Considerations for universal secondary education given the huge gap between primary and secondary school enrollments. Government also needs to strengthen and promote post secondary institutions of learning, focusing on quality and relevance. Other institutions that indirectly impact on human capital development (for example, health sector initiatives) need to be targeted to complement education sector policies.
Policy Recommendations (cont..) Institutional Level There is need to design pro-poor and targeted policies to provide the additional impetus needed to build institutions and investments that are welfare improving. The specific policy recommendations include: (i) Co-operatives and marketing boards: Improved governance and management, infrastructure and widening the resource base for cooperatives to increase their efficiency. Cooperatives targeting poor households, small holders and small businesses should be targeted to facilitate access to extension services, credit and inputs. (ii) Social Infrastructure: Provision of quality care in available units is essential water resource management (reduced pollution and increase access); improvement of roads to enhance marketing of products... Rural feeder roads need to be incorporated in the current road maintenance projects.
Policy Recommendations (cont..) (iii) Local and Legislative Affairs Need to strengthen and closely monitor the recently established constituency and district based development funds: the Constituency Development Fund (CDF), the Community Development Trust Fund (CDTF), the District Roads Fund, the AIDS Fund, the Local Authority Transfer Fund (LATF), and the Constituency Education Bursary Fund to ensure that they benefit the poor. (iv) Law Order and Governance systems The critical policy areas that require attention include: restoration of the rule of law, maintenance of an efficient and motivated police force; development of strong coordinated administration and governance systems; elimination of corruption; strengthening capacity for crime management including investigation and prosecution.
Policy Recommendations (cont..) (v) Role of Private Sector: Private sector, NGOs, civil society and CBOs need to be encouraged to partner with the government and other stakeholders in provision of basic services and also in monitoring and evaluation of existing institutions.
Further Research The WMS data used in this survey is quite dated (almost 10 years old), but it was the only national survey data available at that time. The analysis needs to be updated as new survey data becomes available. Highly aggregated secondary data on institutions was used. Further research should be undertaken to disentangle the relative importance of various correlates of poverty incorporating disaggregated data and also focusing on quality of institutions. Further research should also consider the role of socio- cultural and agro-climatic factors and other institutions such as non-governmental, civil society and community based organizations in poverty alleviation.