Government Policy and Female Labour Market Participation in Kenya: Implications for Poverty Reduction

Government Policy and Female Labour Market Participation in Kenya: Implications for Poverty Reduction By Rosemary Atieno Institute for Development Studies University of Nairobi, P.O. Box 30197, Nairobi Kenya Email: ratieno@uonbi.ac.ke A Draft Paper Submitted to the 2009 Annual IAFFE Conference, Boston, USA June 26-28 2009

Abstract Participation in employment is important for poverty reduction because of its direct effect on incomes and therefore household welfare. Female participation in employment is crucial for poverty alleviation because of the over representation of women among the poor. Women also constitute a significant share of household heads. In Kenya, the problems of persistent and increasing poverty and unemployment are major challenges facing the country. A number of initiatives have been undertaken aimed at reducing poverty. Most of these have focused on accelerating economic growth, which was expected to provide employment opportunities and raise incomes. However, the impressive performance achieved in the first decade of independence was not sustained, with poverty and unemployment continuing. Since the late 1970s, the country has experienced steady declines in economic performance culminating in the negative growth rate of -0.3% recorded in 2000. The declining economic performance meant declining ability of the economy particularly the formal sector to generate employment opportunities. As the formal sector employment has declined, the informal sector employment has recorded sustained increases in employment. The result of the declining performance has been persistent and increasing poverty and unemployment. Despite the increasing importance of the informal sector as a source of employment, some of its characteristics raise questions about its potential for raising welfare and alleviating poverty. The sector is increasingly becoming a sponge for easing out open unemployment into underemployment. An important aspect of the informal sector is that it has become a major employer for the female labour force. This is because of the labour market characteristics, which restrict women from certain occupations like formal sector wage employment. Women are more disproportionately concentrated in certain sectors of the labour market, with limited access to the more remunerative activities. The relationship between the income disparity and the gender gap is further demonstrated by the fact that the proportion of female headed households at higher income levels is significantly lower than that of men. There are hence serious gender inequalities in employment, with implications for poverty alleviation in the country. Policy interventions to improve the access of women to employment opportunities are therefore necessary in efforts aimed at reducing poverty. Using secondary data collected in 1997 through a national survey, (the welfare monitoring survey), this paper analyses the factors determining the participation of females in the labour market in Kenya. The paper discusses the implications of women s over representation in the informal sector and other less remunerative activities for poverty alleviation in the country. In light of the factors that are important in determining access to different categories of the labour market for women, the paper draws conclusions on policies relevant for addressing inequality in access to employment opportunities as an important means to alleviating poverty in the country. 1. Introduction Participation in employment is important because of its direct impact on incomes and therefore livelihoods. In Kenya, female labour force participation is especially 1

important because women constitute a significant share of household heads in Kenya, and they are also over represented among the poor. Although women constitute 53% of the labour force, their participation in wage employment in the modern sector has remained low. Women have access to less than 30% of wage employment (Republic of Kenya 2002)). A number of factors contribute to restrict women s access to formal employment. These include traditional roles, occupational segregation by gender, lack of access to technology, productive assets and credit among others. Women are therefore affected more by poverty and underemployment than their male counterparts. The government policy of focusing on economic growth as a means of realising rapid development failed to ensure sustained increase in economic performance. The declining performance of the economy has resulted in declining ability of the economy to generate employment especially in the formal sector. Due to the inability of the formal sector to generate additional employment, the informal sector has emerged and continued to increase in importance as a source of employment and livelihoods. As the informal sector has grown, it has also become an important employer of the female labour force. This is mainly due to labour market characteristics, that restrict women from certain occupations. Women are more disproportionately concentrated in certain activities. Surveys of the sector in the country show that women dominate most informal sector activities. The 1999 Baseline survey of the sector shows that the ownership of the informal sector enterprises is almost equally divided between male and females. While men owned 52% of the enterprises, women owned a significant 48%. However, of importance for poverty reduction is the fact that in the informal sector, female owned enterprises have been found to employ fewer workers, and less capital compared to the male owned ones (CBS, KREP and ICEG 1999). These factors are likely to restrict women's benefits from the sector. Certain characteristics of the sector also raise questions regarding its potential in generating income and employment and hence poverty reduction. Studies on the informal sector have shown that despite the proliferation of informal sector activities, many of them do not grow (McCormick 1992). Informal sector activities are 2

characterised by small size of activities, few workers often less than six, little growth and limited access to services. A significant proportion of those counted as employed are also underemployed in the informal sector. The sector is therefore increasingly operating as a sponge for easing open unemployment and transforming it into underemployment (UNDP 2002). Generally, there are gender inequalities in employment, with implications for poverty reduction. This paper discusses the factors determining participation in the labour market by women in Kenya, and the implications of women s over representation in the informal sector for poverty reduction in the country. The remaining sections of the paper are organized as follows. Section two traces the economic performance of the country over time, focusing on government policy and its effect on employment and poverty reduction. Section three presents an overview of female employment in the country. Section four is based on the factors that theoretically explain participation of women in the labour market. Section five discusses the methods of analysis used and the types and sources of data used. Section six discusses the results from the empirical analysis. Finally section seven gives the summary and main conclusions and the main policy issues arising from the discussions in the paper. 2. Economic Performance and Employment in Kenya Since independence, the reduction of poverty and unemployment have been major government objectives. A number of policy initiatives have therefore been undertaken to address these vices. Most of these initiatives emphasised achieving rapid economic growth as a means to reducing poverty and improving the standard of living. This position is reflected in various policy documents, from the Sessional Paper Number 10 of 1965 on African Socialism and its Application to Kenya, and subsequent policy pronouncements. However, after decades of intervention, poverty and unemployment have not only increased, but have become more entrenched. The level of poverty currently stands at 56%. Poverty is highest in the rural areas where the poverty level stands at 59.56%, compared to 51.48% for the urban areas (Mwabu et.al, 2001; Republic of Kenya 2001). This implies the need for a new approach, towards poverty reduction and hence the need to understand the role of government policy. 3

The approach of emphasising economic growth succeeded for a while. The country experienced rapid economic growth in the first decade of independence. During the first decade of independence, the economy grew by 6.6% per annum over the period 1964-1973. This impressive performance was however not sustained in the second decade as the growth rate declined in the seventies to 5.2% during the 1974-79 period. There was a further decline in the 1990s to 2.5% during the 1990-1995. This decline has continued with the economy recording a negative growth of -0.3 in the year 2000 1. Since the early 1980s, the country has experienced steady decline in economic performance. For most of 1980s, the economy has been under stress, with the major macro economic indicators like investments and savings as well as capital formation in major sectors experiencing declines. The Public sector employment declined from the mid 1980s mainly due to the retrenchment programmes, privatisation of public enterprises and shedding off of labour in private companies undergoing restructuring. The impact of reforms on employment has therefore been erratic, failing to deliver sustained growth and employment generation. It is observed that paradoxically, the distortions that the structural adjustment programmes were intended to remove further reinforced during the first decade of the implementation of the programmes (ILO 1999). Lack of structural transformation that would lead to employment generation was also a problem (ILO/EAMAT 1999). The formal sector employment has therefore steadily declined in its share of employment. While the formal sector wage employment grew by 2.1% in 1998, this declined to 0.6% in 1999, 0.4% in 2000 and recorded a negative growth rate of -1.1% in 2001 (Republic of Kenya 2002). As the formal sector has declined, the informal sector has become increasingly important in the Kenyan economy as a source of employment and incomes. The share of employment between the formal and informal sectors has changed drastically since then, with the latter overtaking the former in employment absorption. During the last decade, the growth rate in the sector s employment has remained above that of the 1 See National Poverty Eradication Plan 1999 p. 23, and Economic Surveys, Various Issues. 4

formal sector, which has experienced a decline in its growth rate. This has seen the sector s share in total employment rise from 16% in 1980, to 63.6% in 1997 and 70% in 2000. Currently, informal sector s share in total employment stands at 75% (Kenya 2004). Between 2000 and 2001, employment in the sector rose by 11.4%. Its share in GDP has also recorded increases, rising from 13% in 1993 to 18% in 1999 (Republic of Kenya, 2002). Sectorally, the informal sector is the second largest source of employment after small-scale agriculture (Ministry of Finance and Planning, 2000). The 1999 national survey of small and micro-enterprises (SMEs) found that about 26% of the total households in the country are engaged in some form of SME activity (CBS, KREP and ICEG, 1999). The sector is therefore an important source of livelihood for a majority of the country s population. However, although the informal sector has increased its share in employment, looking at its characteristics reveals that its growth may not reflect any aspect of dynamism, but has largely acted as a cushion for the unemployed. The increasing contributions of the informal sector both to GDP and employment sharply contrasts with the decelerating performance of the formal sector, which has resulted from a number of factors. These include the current economic recession occasioned by adverse weather conditions, and reduced economic activity in the main sectors of agriculture, and manufacturing. The on-going reforms in the public sector, namely retrenchment and restrictive government employment policy have further reduced the sector s potential for employment generation (ILO/EAMAT, 1999). The small size of the formal labour market also reflects the constraints facing the sector like high risks, poor infrastructure and lack of social capital (Bigsten and Horton, 1997). The formal sector has therefore faced increasing inability to generate employment, making the informal sector an easy outlet for the unemployed labour force. The declining share of formal sector employment may help explain the rising unemployment in the country in the face of increasing labour supply. The declining economic performance was accompanied with persistent and increasing poverty, unemployment and inequality. This is also reflected in the low human development index for the country further emphasising the declining welfare of the 5

population (UNDP 2002). The poor economic performance could ensure neither an increase in employment generation, nor structural and economic transformation of the economy necessary for the diversification which could spread the gains from growth and hence reduce poverty. There was little real transformation of the economy, with agriculture continuing to support the livelihoods of more than two thirds of the labour force, without any increase in productivity. What is uncertain is the extent to which policy efforts were made to curtail the uncertainties in growth and build strong foundations for economic growth that can generate employment and incomes for a growing population. 6

3. Overview of Female Employment in Kenya There is a general inequality in access to opportunities between men and women spilling over to female employment. Social cultural and economic factors in Kenya have combined to disadvantage women with respect to employment opportunities. Inequalities between men and women in assets, earnings, education and employment still dominate the work place in Kenya. Men mostly control productive assets, like land and real estate, which prevent women from seeking bank loans for investments. Men largely control decision making on household expenditure, thus constraining women s ability to make strategic investments. These affect women s ability to improve their human capital status and hence their advancement (ILO 2004). Women are more likely to be unemployed than men, with the average income being lower for women than men, resulting in more women being poor than men. Although participation rate for women in the labour force has increased over the years, there is still gender inequality in employment (UNDP 2002). Female employment in the modern wage sector accounts for only 30% of the total wage employment (Republic of Kenya 2002). In the rural areas, labour force participation for women is much higher than for men. It is estimated that women form a strong 70% of the total labour force in agriculture. Because of limited opportunities and access to formal employment, the avenue open to them other than agriculture and domestic work is self employment in the informal sector. This results in extreme disparity in incomes and hence the incidence of poverty between men and women. The 1997 WMS II data shows that for both rural and urban areas, the proportion of female headed households at higher incomes is lower than men. This reflects the prevalence of poverty among female headed households (UNDP 2002). A number of factors also make women s participation in the informal sector important. Women s share in formal sector employment is proportionately less than that of their male counterparts, while participation of females in wage employment has remained low compared to men s. The share of women in the labour force also shows that they are disproportionately concentrated in community, social and personal services. Although the women s share in total wage employment has increased to 29.5% in 2001, their share in the traditionally male dominated industries 7

still remains low, while their share in community, social and personal services stands at 42.8% (Republic of Kenya, 1998, 1999, 2002). There are fewer women than men among the regular employees and skilled workers. Women on the other hand outnumber men in the categories of unskilled workers and dominate among unpaid family workers. Regionally, unemployment levels are higher in both rural and urban areas for women than for men. The main characteristics of women owned enterprises in the informal sector are also important for poverty reduction. They start small, grow slowly, and end smaller than the men owned enterprises (McCormick and Mitullah, 1995). They locate more in the home, rely more on less skilled and unpaid workers, and are less likely to diversify into other activities. In addition, women s activities tend to be less remunerative than the men s. Women face a number of obstacles in entering business, which mostly condemn them to low income occupations. Surveys on the labour force participation in both the formal and informal sectors show that women are disproportionately concentrated in certain activities, with limited access to more remunerative enterprises. This is mainly because the choice of occupation depends on factors like education, training, capital requirements, premises and expected earnings. These factors combine to restrict women to trade and other service activities in the informal sector and bar them from other activities. This exclusion from certain activities limits their incomes since male dominated activities are better paid than the female dominated ones. The 1999 ILO report also notes that women s participation in specific sub sectors and activities of the informal sector is quite low (ILO/EAMAT 1999). Household budget surveys have shown that female headed households are more vulnerable to falling into poverty than the male headed ones because of women s limited access to productive assets and education. 8

4. Factors Explaining Participation in the Labour Force Theoretically, a number of explanations have been advanced for female participation in the labour market. The processes that generate the observed distribution of people across occupational categories are important both to households and policymakers. Different approaches exist that can be used to explain the participation in the labour market and hence access to different occupations. The theory of human capital can be seen as a starting point for the analysis of occupational choice and access to different opportunities in the labour market. Occupational differences in this context can therefore be seen as reflecting differences in human capital endowments Mwabu and Evenson (1997). The Human capital hypothesis further argues that differences in wages and segregation in work is largely due to differences in the human capital content of male and female reflected through differences in productivity. The importance of human capital has been used to explain the trend of increasing participation of females in the labour market (Maglad 1998). Investments in human capital like education and health contributes to development by raising labour productivity. Education also reduces fertility and increases child survival, all which make it positively affect female participation in the labour market. The neoclassical explanation of female labour force on the other hand explains it in terms of household characteristics. In this context, the decision to participate in the labour market is an outcome of the income and substitution effects of market work versus leisure as well as that of the market versus unpaid family work. Household characteristics also influence female labour force participation. Education attainment increases women s willingness to work, improves their employment opportunities, and raises their earning potential. Empirically, a number of factors have been identified as determining the involvement of women in the labour markets. Among these are the traditional factors like access to factors of production, credit, information technology and training, the international economic environment and introduction of new technologies as well as changes in the 9

political and social landscape (ILO 1994). Empirical evidence shows that women, especially the heads of households will utilise all opportunities for employment or income. Women are less likely to discriminate in their choice of activities due to the need to cater for their families. This may partly explain the fact that they are found in all types of employment, ranging from permanent salaried employment to temporary wage employment and self-employment (M Bett et. al, 1998). Maglad (1998) identifies a number of factors responsible for women entering the labour market. Looking at female labour supply, Maglad (1998) emphasises the importance of human capital in increasing female labour force participation and shows that expected own wage, spouse s earnings, the number of children and age were important in determining participation in the labour market. Assets however affect work decisions and hours negatively. Spouse s expected wages affected both participation and labour supply negatively. The presence of pre-school children also has a negative effect on participation. Kevane and Wydick (2001) bring another dimension to labour market by looking at women s participation in entrepreneurial activities. They observe that there is an increasing proportion of women involved in entrepreneurial activities, and argue that the share of women in informal employment has increased mainly due to factors like the limited absorptive capacity of the formal sector, difficulty in entry to the formal sector by women, changes in household gender norms and macro-economic dislocations and adjustments. A number of factors in Kenya have been given for this gender disparity in employment. These include occupational segregation in the labour market, social attitudes towards women, inadequate skills and lack of gender sensitive employment policies. The labour market policies of the colonial period, which have been carried over by many African countries like Kenya were aimed at generating cheap labour. Until recently, the Kenyan labour market was highly regulated. The factors that can be seen as explaining public sector employment have therefore included shocks to the system like ad hoc employment policies in the public sector. An important observation in the trends in formal sector employment is that economic variables have 10

had little to explain in employment growth, but rather the shocks, reflecting the ad hoc employment system. Economic factors have not been important criteria for defining employment trends. 5. Methodology 5.1 Methods of Analysis Access to the labour market or job attainment can be seen as an outcome of the interaction between demand and supply. In analysing the choice by individuals to participate in the labour market, we assume the existence of multiple activity choices, including non-participation (Krishsnan et.al, 1998; Glick and Sahn 1997; Mwabu and Evenson 1997). As identified from the literature review, a number of factors determine labour demand and supply. The empirical problem in this study can be described as determining the probability of an individual female choosing to participate in a given sector of the labour market. Given that there are several possibilities of alternative activity choices, the study uses the multinomial logit model, which allows for the identification of factors determining the participation in various sectors. The dependent variable in the logit model is grouped into six categories based on the realities of the Kenyan Labour market. These are public sector, private sector, informal sector, agricultural sector, unemployed and unpaid family worker. 5.2 Types and Sources of Data The study uses secondary data obtained from the 1997 Welfare Monitoring Survey (WMS) III. This data set was collected through a survey conducted by the Central Bureau of Statistics (CBS), using the National Sample Survey Evaluation Programme (NASSEP). In this section, we present the description of the data used. The 1997 WMS III data consisted of 50,713 individuals, from 10,874 households. Out of the total sample, there were 24,910 males (49.1 %) and 25,803 females (50.9 %). From the total sample, adults of the working age were selected consisting of those individuals of 15 years of age and above (CBS 2002). The total adult population 11

is therefore 27,767 out of which 13,277 were males and 14,490 females. The results discussed in this paper are based on the female adult sample. The dependent variable was grouped into six categories based on the realities of the Kenyan labour market (Wambugu 2002; Mwabu and Evenson 1997), namely; public sector, private formal sector, informal sector, agricultural sector, unpaid family worker and the unemployed. The distribution of the total adult sample by the different categories of the labour market participation for the 1997 data set is presented in table 1. A total of 3,933 cases could not be identified by employment category, of which 2,158 were males and 1,775 were females. Table 1: Distribution of the 1997 Sample by Employment Categories Employment Category Total Sample Males Females Public Sector 1,425 1,013 412 Private Formal Sector 1,258 982 276 Informal Sector 3,628 2,324 1,304 Agriculture 6,623 3,604 3,019 Unpaid Family Worker 9,239 2,433 6,806 Unemployed 1,661 763 898 Missing 3,933 2,158 1,775 Total 27,767 13,277 14,490 The female adult sample used in the analysis is therefore 14,490. From the table we see that unpaid family worker constituted the majority of the sample, followed by agricultural sector and informal sector. Most males were employed in the agricultural sector, while most females were in the category of unpaid family worker, followed by agriculture and the informal sector. The females were fewest in the private sector followed by the public sector. This shows that the data reflects the reality of the Kenyan labour market and is consistent with other surveys in the country, which show that women dominate in the category of unpaid family worker (CBS 1996). The 1999 census analytical report on labour force Volume IX shows that most of the employed economically active population were self employed, working for family 12

gain and receiving no salary or wages in either the family business or agricultural holdings (CBS 2002). 6. Results and Discussions 6.1 Summary Statistics Table 2 gives the summary statistics of the variables used by gender for the 1997 data set. We observe that males are on average older than their female counterparts with an average age of 34 compared to 33 for females. The mean age for the sample is however 34 years. The mean number of years spent in education is higher for males at nine compared to females at eight. Given the Kenyan education system, this can be interpreted to mean that while the school drop out level for boys is secondary school, for girls it appears to be primary school. Girls are therefore less exposed to education beyond primary level of schooling. The household size did not differ between males and females and so was the number of infants. Among the females, only 21% were household heads compared to males where 59% were household heads. 61% of the males reported being in gainful employment, compared to 45% for females. Table 3 gives the summary statistics for explanatory variables across the categories of the dependent variable. The table shows that the unemployed category consists of relatively younger population compared to the other categories. At a mean age of 32 years, unemployed females are older than their male counterparts in the same category whose. Agricultural sector has the oldest workers, with the mean age for males being 39 years while that for females is 38 years. The public sector has the highest age gap between males and females with 38 years for males and 34 for females. The household size did not vary much between males and females, but in all the categories except the public sector, males show on average larger household sizes than their female counterparts. In terms of education, the public sector workers, both males and females show higher levels of education than the other categories in terms of years of schooling, followed by the private sector. Agriculture shows the least level of education for both males and females. In terms of household headship, more males were household heads, compared to their female counterparts for all the employment categories. This can be attributed to the nature of the population, which appears to be 13

young, with a mean age of 34 years. The focus of this paper is females. However, in this section, we have presented summary statistics for both males and females in order to give the structure of the population from which the sample has been drawn. In the subsequent sections, the analysis focussed only on the female population. 14

Table 2: Summary Statistics for the Variables used in the Study by gender- 1997 Total Sample Male Female Variable N Mean Std dev. N Mean Std dev. N Mean Std d Age 27,767 34 16.2 13,277 34.25 16.208 14,490 33.9 16.04 Years in school 27,767 9 3.02 13,277 9 3.15 14,490 8.7 2.82 Household 27,767.39.488 13,277.59.492 14,490.21.408 Headship (1=yes) Household size 27,767 4.6 2.716 13,277 4.6 2.695 14,490 4.6 2.735 Land size owned 27,767 27.2 152.17 13,277 27.45 152.1 14,490 26.58 152.3 Number of Infants 27,767 1.43.628 13,277 1.43.623 14,490 1.3.632 In gainful employment (1=yes) 27,767.53.499 13,277.61.487 14,490.45.497 15

Table 3: Summary statistics for explanatory variables by employment category and Gender -1997 Public sector Private Informal Sector Unemployed Agriculture Un Variable Formal Sector Fam Male Female Male Female Males Females Males Female Male Female Ma Age 38 34 35 31 35 33 31 32 39 38 36 Years in School 12 12 10 11 9 9 9 9 8 8 9 Household.91.39.84.43.76.34.31.15.76.37.51 headship (1=yes) Household Size 4.3 4.9 4.6 4.1 4.5 4.5 4.9 4.5 4.5 4.8 4.5 Land owned 32 26 28 34 29 29 41 45 26 18 20 Gainful employme.97.97.96.92.82.74.09.05.85.80.41 nt (1=yes) Number of infants 1.4 1.4 1.3 1.75 1.4 1.4 1.4 1.4 1.5 1.4 1.4 16

6.2 Participation in the Labour Market In this section, we present results of the regression analysis for participation in the labour market for the females. The reference category for the labour market participation is taken as the unemployed category (Liao 1994; Mwabu and Evenson 1997). This category consists of economically active population, without regular employment and actively seeking work (CBC 2002). This section presents the multinomial logit regression results for the 1997 data. Table 4 shows the coefficients of the multinomial logit regression. Table 4: Multinomial Logit Parameter Estimates for Labour Force Participation for the Female Sample 1997 Variable Public sector Private sector 0.795*** 0.765*** Age (3.04) (3.08) Informal Sector 0.718*** (3.54) Unpaid family worker 0.495*** (3.27) Agriculture 0.380*** (2.65) Age 2-0.0086*** (-2.64) 0.680*** Years of (3.83) schooling Marital status -0.270 (0.91) -.0083*** (.2.76) 0.533*** (3.01) -0.124 (0.41) -0.0083*** (3.47) 0.332** (2.01) -.150 (0.54) -.0048*** (3.12) 0.261 (1.56) -.280 (1.02) -0.0037*** (2.60) 0.237 (1.43) 0.197 (.73) Land owned 0.252 (0.88) 0.252 (0.88) 0.216 (0.74) 0.252 (0.88) 0.251 (0.88) Household head 22.182*** (4.01) 24.305*** (4.46) 25.758*** (5.69) 30.896*** (8.29) 30.637 Household size Rural/urban cluster Constant 0.075 (0.48) 1.884* (1.72) -45.615** (8.79) -0.024 (0.15) 2.117* (1.92) -46.03** (9.23) 0.027 (0.18) 1.355 (1.33) -42.069** (10.25) -0.0056 (0.04) -0.410 (0.39) -40.511** (7.89) 0.075 (0.51) -2.074* (1.74) -35.965*** (9.95) 17

Model χ 2 (40) 199.77 N 14,490 Log likelihood -385.234 Note: *** Significant at 1%; ** Significant at 5%; * Significant at 10%. Figures in parentheses are the Z-statistic From table 4, we see that female participation in the public and private sectors is likely to increase with age, years of schooling, being a household head and being in the urban areas. Age squared, has a negative and significant coefficient for all the categories, showing that beyond a certain age, older women have lower chances of participation in the labour force. The land owned and household size do not appear to have any effect on participation in both the public and private sectors. For the informal sector, the same effects are seen. However, being in the rural or urban areas does not seem to affect the participation of females in the informal sector. Whereas participation in the unpaid family labour and agriculture is likely to increase with age, the years of schooling and land owned do not have any effect on female participation in the two sectors. Household headship is likely to increase participation in unpaid family worker, but it has no effect on participation in the agricultural sector. Being in the urban area reduces the chances of participation in agriculture significantly. From the results the following observations can be made. Education as represented by the years of schooling increases women s chances of being employed in the public and private sectors. That education is not significant for agriculture and unpaid family work is important since these are the sectors where women are most employed. Increasing women s access to education would therefore increase their chances of accessing public and private sector employments which also have higher remunerations. Although the years of schooling is significant for informal sector participation, more women are still employed there. This may be attributed to the occupational discrimination against women in the labour market or the limited job opportunities in the formal sector. 18

7 Summary, Conclusions and Policy Issues 7.1 Summary and Conclusions This paper has traced the economic performance and government policy in Kenya over time, highlighting its contribution to the persistent and increasing poverty and unemployment. The paper uses empirical analysis to show factors that are significant for participation in the labour market. The paper shows that government policy did not contribute to a dynamic and sustained economic performance. It did not also generate the necessary structural transformation that is necessary for economic development. This negatively impacted on the economy s ability to generate employment especially in the formal sector. Trends in employment have reflected ad hoc measures rather than clear government policies. The result has been increasing unemployment and poverty, which have become more entrenched. The declining ability of the formal sector to generate additional employment has seen the rising importance of the informal sector as a source of employment and livelihoods. However, despite its rising importance the informal sector s characteristics compromise its potential to generate incomes and reduce poverty. The growth of the informal sector does not therefore reflect any aspect of dynamism, but it is largely acting as a cushion for unemployed labour. This does not present reflect the potential to effectively address the problem of poverty. Unemployment and poverty has affected women most, due to their over representation in unpaid family work, agriculture and the informal sector. These sectors have lower remuneration and therefore lower incomes. Reducing poverty therefore requires measure that target increasing women s productivity by strengthening their human capital. The regression results show that education is significant in women s access to formal sector employment. Investments in women s human capital through education is therefore critical for reducing unemployment and poverty. The over concentration of women in the sectors with low productivity and hence returns negates efforts to reduce poverty. 19

7.2 Policy Issues Since women are mostly concentrated in those sectors with low incomes, poverty reduction and employment creation among women may require pro-poor growth policies, targeting those sectors with highest potential returns for women. Agriculture and the informal sector are the major employers of female labour force. Reducing poverty and creating decent work opportunities for women will however involve more than just the commercialisation of agriculture. It will require increasing the opportunities for women to become more productive and earn more through their labour. Therefore policies to improve the economic performance of agriculture, which is considered the backbone of the economy will need to be accompanied by policies to improve women s access to assets, education, health and other social services that enable them to improve their human capital. These include improving girls access to higher education and increasing women s access to productive assets. Improving the access to financial services is also important if women are to exploit their human capital potential like entrepreneurial potential. The informal sector is an important source of employment for the female labour force. In the informal economy, measures to improve women s employment in the sector need to focus on services that enable them to spend more time on income generating activities. This will also require the need to build the capacity of local authorities to provide social services and infrastructure that facilitates profitable income earning opportunities. There is also need for deliberate policy to increase the private sector potential to generate additional employment. This will require the promotion of labour intensive private sector-led growth by facilitating the expansion and job creation capacities of the private sector. It will also require the implementation of macro-economic policies that lead to pro-poor growth and reduces inequalities. 20

References: Bigsten A. and Horton S. (1997). Labour Markets in Sub-Saharan Africa. University of Gothernberg and University of Toronto. Mimeo Central Bureau of Statistics (2002). Kenya 1999 Population and Housing Census Analytical Report on Labour force. Vol. IX. Central Bureau of Statistics, Nairobi Central Bureau of Statistics, K-REP and ICEG (1999). A national Survey of Small and Micro enterprises in Kenya. Central Bureau of Statistics (1996). Welfare Monitoring Survey II, 1994 Basic Report. Central Bureau of Statistics, Nairobi. ILO/EAMAT (1999), Kenya: Meeting the Employment challenges of the 21 st Century, ILO Geneva. ILO (2004). Gender and Employment Dimensions of Poverty: Policy Issues, Challenges, and Response. ILO, Geneva. Kevane M. and Wydick B. (2001). Microenterprise Lending to female Entrepreneurs: Sacrificing Economic Growth for Poverty Alleviation? World Development, Vol. 29. number 7, p.1225-1236. Krishnan P., Selassie T. and S. Dercon (1998). The Urban Labour Market during structural Adjustment: Ethiopia 1990-1997. Centre for the Study of African Economies, WPS/98-9, Oxford University. Maglad N. (1998). Female Labour Supply in Sudan. African Economic Research Consortium, Special Paper 30. M Bett A., Akaffou A. and Alphonse A.A. (1998). The determinants of the supply of women s labour in Francophone Africa. Research proposal submitted to the African Economic Research Consortium, Nairobi. McCormick D. and Mitullah W. (1995). Policy experiences of women in Kenyan small enterprises. Paper presented at UNESCO Meeting on Women in Informal Sector, Nairobi. McCormick D. (1992). Why Small Stay Small: Risk and Growth in Nairobi s Small- Scale manufacturing. IDS Working Paper number 483. Mwabu G. and Evenson R. E (1997). A Model of Occupational Choice Applied to Rural Kenya. African Development Review, 9(1-14). 21

Mwabu G., A. Geda, N. Jong, and M. Kimenyi (2001). Determinants of poverty in Kenya: Household Level Analysis. Discussion paper, no. 9, Kenya Institute of Public Policy Research and Analysis. Republic of Kenya (2002). Economic Survey, Government Printers Nairobi. Republic of Kenya (2001). Economic Survey, Government Printers Nairobi. UNDP (2002). Kenya Human Development Report 2001. UNDP Kenya. 22