Rural-urban Migration and Poverty in Kenya: is Agriculture the Answer? Maria Sassi Dep. of Business University of Pavia e-mail: msassi@eco.unipv.it The aim of my presentation is the understanding of the impact of rural-urban migration on economic development and poverty in Kenya and the role ot technical efficiency change in agriculture.

Importance of the topic Rural-urban migration Sub-Saharan Africa Social, economic, political problems of major significance Rural-urban migration is of particular concern in Developing countries and particularly in SSA where the migration of hundreds of millions of rural people to cities in these chiefly agrarian countries is creating the social, economic and political problems of major significance. For many years the economic development literature has viewed rural-urban migration favourably, but today, it is also recognised that the phenomenon should go with suitable policy intervention in rural areas. Agrarian countries

Theoretical literature (1/2) Dual economy Models (1950s-1960s) Todarian Frameworks (1970s-1980s) New Economics of Labour Migration (form 1990s) The theoretical debate and its consequence policy recommendations can be articulated in three types of contributions. Those referred to:

New Economics of Labour Migration (2/2) Remittances allocation in rural areas Internal migration and rural development Need for economic analysis and justification Rural development policies for poor hhs without migrants MIGRATION POLICY GOAL Accommodate migration flows while preventing the widening of rural and urban imbalaces Agricultural policies

Focus of the analysis - Rural migrant: 70% of urban labour force; Specific policy interventions; Poverty: 54% & 80% in rural areas. Kenya The “growth sector” (24% GDP, 53% exports; 62% labour force); -Increase in productivity. Agriculture Tech. efficiency ch. My analysis wants to provide a preliminary contribution to this debate providing an empirical analysis focused on: Kenya that provides a natural experiment to study rural-urban migration because rural migrants constitute a significant proportion of urban labour force, about 70%. And over time the phenomenon has been object of specific policy interventions mainly of Todarian nature with the introduction of migration flows restrictions and urban wage subsidies and a very little attention to rural areas or the introduction of policies with very little impact on rural development. More recently, with the Poverty reduction strategy paper and the commitment of the country on the MDGs there has been a change of focus: agriculture is seen as the “growth sector” and it has been underlined the need for specific policies particularly aimed at increasing agricultural productivity. This explains my interest in simulating a technological efficiency change in agriculture In my empirical analysis I have adopted CGE approach because as underlined by the theoretical literature on migration ther is the need for understanding how a set of policies affects income, outuput and employment in both urban and urban areas and sectors simultaneously. I am not entering in to the model details due to time restriction, however the 5 sectors CGE model I have developed for Kenya accommodates the 2003 SAM elaborated in 2006 by Kiringai, Thurlow, Wanjale that I have duly reviewed according to the aim of the analysis; The IAKE-CGE model works simulation the interaction of various economic actors across markets as specified in the neoclasical GE theory. Impact of migration policies on income, output & employment in both urban and rural areas & sectors SIMULTANEOUSLY CGE approach 2003 SAM

SAM revision Basic SAM RevisedSAM 50 activities and commodities Agriculture; Food industry; Industry others; Service sector; Public Administration, Health & Education 50 activities and commodities Now I’m try to disaggragete capital accout. Rural Slkilled, Semi-skilled and Unskilled; Urban Slkilled, Semi-skilled and Unskilled (Labour force survey, Central Bureau of Statistics) Labour factors: Skilled; Semi-skilled Unskilled;

Simulations Rural-urban migration Best scenario for rural migrants 10% rural skilled to urban skilled Rural-urban migration 10% rural semiskilled to urban semiskilled 10% rural unskilled to urban unskilled At the existing rural urban wage differential, there exixts an effective demand for migrant labour service proxied by the unit ptobability of securing an urban job. Within this framework I have simulated two typologies of external shocks: The former refers to rural-urban migration; The latter technical efficiency change in agriculture. The assumption is that at the existing rural urban wage differential, there exists an effective demand for migrant labour service proxied by the unit probability of securing an urban job. There is no unemployment. Rural labour migrates in such a way that its supply exactly matches the quantity of effective demand generated in urban areas. There are no differences in information available to rural labour; nevertheless, there exist differences in preferences for urban employment as embodied in labour skills. Best scenario for rural migrants There is NO unemployment, and differences in information. There exist differences in preferences for urban employment as embodied in labour skills

10% rural skilled to urban skilled Simulations 10% rural skilled to urban skilled Rural-urban migration 10% rural semiskilled to urban semiskilled 10% rural unskilled to urban unskilled 10% increase in efficiency of agricultural production function Within this framework I have simulated two typologies of external shocks: The former refers to rural-urban migration; The latter technical efficiency change in agriculture. The assumption is that at the existing rural urban wage differential, there exists an effective demand for migrant labour service proxied by the unit probability of securing an urban job. There is no unemployment. Rural labour migrates in such a way that its supply exactly matches the quantity of effective demand generated in urban areas. There are no differences in information available to rural labour; nevertheless, there exist differences in preferences for urban employment as embodied in labour skills. Technical efficiency change 50% increase in intermediate input specific efficiency in agricultural production function (agricultural, food industry, other industry and service inputs)

Rural - urban migration No significant impact

Households by expenditure decile 10% rural migration: Impact on income (% change) - Households by expenditure decile (I=low; X=high)

Technical efficiency change 10% increase in efficiency of agricultural production function

Impact on income - current prices (%) 10% increase in efficiency of agricultural production function Impact on income - current prices (%) Increase in rich-poor disparities Reduction in rural-urban disparities Households by expenditure decile - (I=low; X=high)

Impact on consumption – constant prices (%) 10% increase in efficiency of agricultural production function Impact on consumption – constant prices (%)

Share of households consumption by product 3.8 -5.3 1.7 1.6 5.3

Impact on macroeconomic indicators (%) 10% increase in efficiency of agricultural production function Impact on macroeconomic indicators (%) Impact on exports (%)

Impact on gross product at factor costs (%) 10% increase in efficiency of agricultural production function Impact on gross product at factor costs (%)

10% increase in efficiency of agricultural production function Impact on wages (%)

Technical efficiency change 50% increase in intermediate input specific efficiency in agricultural production function Impact on Hhs income (% change) Households by expenditure decile (I=low; X=high)

Agricultural inputs Food industry inputs Industry other inputs Service inputs

Impact on macroeconomic indicators - current prices (% change)

Impact on gross product at factor costs - current prices (% change)

10% increase in intermediate service input efficiency in agricultural production function Impact on wages (%)

Conclusions (1/3) Role of rural-urban No impact on income migration and development Micro analysis Macro analysis integration

Conclusions (2/3) Rich-poor and rural-urban inequalities? Support to overall economc development Key role of thechnical progres Increase in households welfare Rich-poor and rural-urban inequalities?

Conclusions (3/3) Reduction in agricultural prices Increase in Farmers return & livelihoods Increase in labour cost Withdrawl from agriculture Liberalization & Price fluctuations

Theoretical literature (1/3) Dual economy models - Initial stage of development; - Intensive agricultural economy; - Lack of capital & surplus of agricultural labour the needed manpower for industrialization. Todarian frameworks New Economics of labour migration Rural urban migration facilitation to favour economic take-off

Theoretical literature (1/4) - Understanding of the reasons of urban unemployment and its link with rural-urban migration Dual economy models Todarian frameworks - Migration flows restrictions; - Wage subsidies Uniform wage subsidies B-S; B Urban wage subsidies Compensations to rural areas H-T New Economics of Labour Migration Job creation in rural areas & inequalities F.

Theoretical literature (1/5) Support to R-U migration Increasing efficiency in job matches Dual economy models - Who, when and how to migrate Todarian frameworks - Internal migration and rural development Remittancies; Policies for rural development. New Economics of Labour Migration

Migration policies in Kenya a. Colonial period b. Post-colonial period Kenyatta presidency (1963-1979) Moi presidency (1979-2002) Kibaki presidency (from 2002)