Carolina Mejía-Mantilla1 Understanding employment transitions and their implications in terms of welfare and labor outcomes is key for countries to design labor market policies aimed at inclusive growth, particularly in developing countries. Using panel data, this note characterizes the magnitude and nature of the main employment transitions in Uganda between 2009 and 2016 and explores how these transitions correlate with measures of wellbeing (mainly per capita consumption) and employment outcomes. Understanding employment transitions is an and employment outcomes that accompany these important aspect of designing public policies that transitions hinge critically on gender, age and level of allow people to benefit from economic education. The findings are aligned with the recent development. People progress in the workplace by literature on employment transitions in developing accessing the labor market, by getting more efficient countries and, more specifically, in African countries. at the job they do, or by switching to a better job. All These results suggest that the country should make of these can entail changing occupations or sectors, an effort to improve the productivity of both the changing locations, or shifting from informal agricultural sector and the services sector (which is employment to formal employment. Generating absorbing the majority of workers coming from the quality jobs in a sustainable manner is a priority area primary sector) and create more job opportunities for Uganda. The government has become across all sectors of the economy. These increasingly concerned with the challenge of creating recommendations are in line with those of Beegle jobs and enhancing the quality of the existing ones, and Christensen (2018), which aim to accelerate as evidenced in the National Development Plan II poverty reduction in the Africa region. (2015/16 - 2019/2020). In the case of Uganda, like many countries in Sub- Saharan Africa, there is considerable persistency in terms of employment status, geographical The majority of workers reside in rural areas and location and sector. The large majority of workers work as self-employed in agricultural activities. are likely to remain in the agricultural sector, mainly Four main trends were observed in the country’s as self-employed in rural areas, and thus, in informal labor market over the last ten years. Firstly, after a jobs. For the few that transition, changes in wellbeing sustained increase, labor force participation declined 1 This policy note summarizes research on the topic by Pablo Adrian Garlati (Universidad Javeriana), Carolina Mejia- Mantilla (World Bank), Dino Merotto (World Bank), and Michael Weber (World Bank). from almost 90 percent in 2012 to 79 percent in 2016. and formal employment and (iv) across sectors – This change has been mainly driven by youth and agriculture, services and industry. We use five waves women, for whom the decline has been most of the Uganda National Panel Survey (UNPS), significant. The second defining characteristic of spanning over the period 2009/10 to 2015/16, to employment is the widespread prevalence of analyze the magnitude of the transitions and analyze informality and how little this trait has changed over if there are differences across age, education and time. As in most of Sub-Saharan Africa, self- gender. Then, we correlate the transitions with employment, non-paid employment, and informal changes in consumption per capita and, in some 2 wage employment remain by far the most common cases, wages and hours worked. While the nature of employment types, adding to almost 80 percent of the analysis does not demonstrate any causal total employment (ILO, 2015). Thirdly, while the share relationship, we believe that these correlations are of urban employment increased by ten percentage informative. points in the last decade, reaching 24.6 percent in 2016, the large majority of workers still reside in the Inactivity/employment transitions rural areas of the country. Employment rates are persistently high in Structural transformation over the last decade Uganda. For the first transition analyzed, between has been limited. The fourth stylized fact is that being inactive and employment, we restrict the while agriculture remains the most common type of sample to household heads only. 3 Close to 95 employment (64 percent of all employment in 2016), percent of all household heads remain employed there has been a modest sectorial shift towards the from wave-to-wave and 45 percent go from being services sector, from 20 in 2005 to 29 percent in 2016; inactive to being employed. This is a very similar and a very limited shift towards the industrial sector, finding to that obtained for Mauritius (World Bank from 6.8 to 7.3 percent. Despite the fact that the value 2018), where 93.7 percent of individuals remain in added per worker is higher in the services sector employment over the period 2005-2012. compared to the agricultural sector, by a ratio of 6 to 1, productivity has been declining in the services There seems to be a premium for transitioning sector in the last five years, a concern for the coming into employment. Workers that stay employed years. enjoy on average a 10 percent increase in their consumption per capita, while those that they leave inactivity enjoy an increase that is almost twice as high. The premium for accessing the labor market is particularly high for women as seen in Figure 1. Taking the country context into account, our Surprisingly, we observe that transitions out of the analysis focusses on four employment transitions: employment are also beneficial for consumption. (i) between inactivity and employment, (ii) between However, we suspect that the causality runs in the rural and urban employment, (iii) between informal other direction: individuals transition out of the employment only because they are better off. 2 3 Defined here by whether the worker is under a contract with We include only heads of household, given that a large the employer. share of inactive population has access to higher consumption only through the head of household. Percentage Change in Consumption by Adult Equivalent the Sub-Sharan African experience of low levels of rural-to-urban migration linked to the slow pace of Figure 1: Inactive-Employed Transitions (age 15+ and heads of household) structural transformation in the region (World Bank, 2012). Younger workers are more likely to transition, 50 especially from urban to rural areas, more educated 40 workers are more likely to move from rural to urban 30 areas, and transition rates are very similar for males 20 and females. 10 The changes in wellbeing that accompany these 0 transitions depend greatly on gender, education All Male Female and age. Moving from rural to urban areas benefit Active-to-inactive Stay inactive Inactive-to-active Stay active workers in the 25 to 34 years age bracket and brings Source: Staff calculation based on UNPS microdata women higher consumption growth when compared . to men (albeit not compared to having stayed in rural Figure 2: Rural-Urban Transitions (age 15+) areas), as seen on Figure 2. Moreover, more educated 12 individuals obtain larger gains from a rural to urban transition in terms of wage growth, and those moving 10 from urban to rural areas experience a lower increase 8 in hours worked across all groups analyzed 6 4 Informal-to-formal transitions 2 Informality is the norm and persistent for 0 workers, as in most countries of the region (ILO, All Male Female Urban-to-rural Stay rural 2015). 4 An important finding of the study is that Rural-to-urban Stay urban transitioning out of informality is very unlikely for Source: Staff calculation based on UNPS microdata. workers (we observe this movement for only 2 percent of the workers in our sample). Informality Rural-urban Transitions seems to be a more permanent condition in Uganda than in other countries of the region such as Tanzania As in other parts of Africa, rural-urban transition and Ghana (Falco et al. 2014).5 Transitioning out of is not a common phenomenon. Movements from formality it is slightly more probable for more rural to urban areas were less common than from educated (primary completed and higher) and older urban to rural areas: around 5 percent in the first case (above 35 years of age) workers. This seems to versus 15 percent in the second. This is in line with coincide with a labor market model in which 4 Workers are classified as formal if they are: (i) employees and/or income tax, (iii) unpaid apprentice or (iv) household contributing to a pension fund, (ii) employers, own account, farmer. 5 unpaid family workers registered for VAT and/or income While the authors separate self-employment and unpaid tax or (iii) paid apprentice; and informal if they are: (i) work as separate employment status, persistency in these employee not contributing to a pension fund, (ii) employers, states is high, close to 80 percent for self-employment and own account, unpaid family workers not registered for VAT slightly lower for unpaid work. informality is comparable to queuing for high quality where there is significant sector persistency and jobs, particularly amongst the young (Bosch and workers moving away from the primary sector Maloney, 2010). experience positive changes in earnings (World Bank 2018). Transitioning from formality to informality Percentage Change in Consumption by Adult Equivalent carries a penalty (see Figure 3). While consumption Figure 3: Informal-Forma Transitions (age 15+) growth is about 10 percent for those that remain 18 formal, it is only 3 percent for those that transition into informality and negative for men. It is also 13 negative for the young (15-24 age bracket) and the 8 less educated. Transitioning into formality or informality leads to an increase in the number of 3 hours worked (compared to being static), which may -2 be explained by a learning or adjustment period All Male Female implied by the change of status. Formal-to-informal Stay informal Informal-to-formal Stay formal Sectorial transitions Source: Staff calculation based on UNPS microdata. Figure 4: Sectorial Transitions (age 15+) The more educated are more likely to move out 150 of agriculture. While transitioning out of the agricultural sector is rare, workers are slightly more 100 likely to transition from agriculture to services than from agriculture to the industrial sector; consistent with the aggregate employment trends. Importantly, 50 more educated workers have a higher chance of moving out of agriculture, and of remaining in the 0 industrial and services sectors. Inter-sectorial Services-to-agriculture Industry-to-agriculture transitions do not seem to be correlated with gains Stay at agriculture Agriculture-to-industry Services-to-industry Stay at industry in terms of consumption growth, and the results on Agriculture-to-services Industry-to-services hours worked are mixed (see Figure 4). Source: Staff calculation based on UNPS microdata. However, the outlook changes somehow when only wage workers are considered. Wage workers ABOUT THE AUTHOR gravitate towards the services sector, and those Carolina Mejía-Mantilla is a Senior Economist at transitioning out of agriculture into the industrial sector do exhibit a higher hourly wage growth. This is the World Bank’s Poverty and Equity Global not true for the services sector, a result likely linked Practice (GPV). cmejiamantilla@worldbank.org. to the productivity differences across sectors. These patterns are similar to those observed in Mauritius, This note series is intended to summarize good practices and key policy findings on Poverty-related topics. The views expressed in the notes are those of the authors and do not necessarily reflect those of the World Bank, its board or its member countries. Copies of these notes series are available on www.worldbank.org/poverty