Food Policy 67 (2017) 153–174 Contents lists available at ScienceDirect Food Policy journal homepage: www.elsevier.com/locate/foodpol Are African households (not) leaving agriculture? Patterns of households’ income sources in rural Sub-Saharan Africa q Benjamin Davis a, Stefania Di Giuseppe b, Alberto Zezza c,⇑ a Food and Agriculture Organization of the United Nations, Italy b Universitá di Teramo and Food and Agriculture Organization of the United Nations, Italy c World Bank, United States a r t i c l e i n f o a b s t r a c t Article history: This paper uses comparable income aggregates from 41 national household surveys from 22 countries to Available online 7 November 2016 explore the patterns of income generation among rural households in Sub-Saharan Africa, and to compare household income strategies in Sub-Saharan Africa with those in other regions. The paper seeks to under- JEL classification: stand how geography drives these strategies, focusing on the role of agricultural potential and distance to Q1 urban areas. Specialization in on-farm activities continues to be the norm in rural Africa, practiced by 52 O1 percent of households (as opposed to 21 percent of households in other regions). Regardless of distance R2 and integration in the urban context, when agro-climatic conditions are favorable, farming remains the Keywords: occupation of choice for most households in the African countries for which the study has geographically Income explicit information. However, the paper finds no evidence that African households are on a different tra- Non-farm employment jectory than households in other regions in terms of transitioning to non-agricultural based income Agriculture strategies. Africa Ó 2016 The World Bank. Published by Elsevier Ltd. This is an open access article under the CC BY IGO LSMS license (http://creativecommons.org/licenses/by/3.0/igo/). 1. Introduction improvement of the policy environment and better terms of trade, provide a more conducive environment for higher agricultural Agriculture declines as a share of aggregate output with overall growth and an opportunity for the much awaited structural trans- growth in GDP per capita as countries undergo the structural trans- formation in Africa (Binswanger-Mkhize et al., 2010). formation that accompanies economic development (Chenery and A rather large body of literature has developed over the last Syrquin, 1975). In rural areas of developing countries, the decline 20 years investigating the importance and features of rural non- in the relative importance of agriculture and the expansion of rural farm income and employment in the developing world, the deter- non-farm activities are likely features of the process of economic minants of households’ participation in and returns to different development. Growth in rural non-farm activities cannot be seen income-generating activities, and the extent and determinants of in isolation from agriculture, however, as both are linked through rural household income diversification (FAO, 1998; Barrett et al., investment, production, and consumption throughout the rural 2001; Lanjouw and Lanjouw, 2001; Haggblade et al., 2007; economy, and in relation to urban centers, and both form part of Winters et al., 2009, 2010; Davis et al., 2010). The 2007 World complex livelihood strategies adopted by rural households. Better Development Report on agriculture and the 2011 IFAD Rural Pov- incentives for agriculture during the past decade, via the erty Report also devoted much attention to these themes. A major conclusion of these studies is that rural household income diversi- q fication is the norm rather than the exception, and that while We would like to thank an anonymous reviewer, Raka Banerjee, Chris Barrett, Gero Carletto, Luc Christiaensen, Roberto Esposti, Peter Lanjouw, and participants to endowments (e.g. physical, human, natural capital) and wealth two ‘‘Agriculture in Africa – Telling Facts from Myths” team workshops for play a role in driving engagement in different economic activities, comments on earlier drafts. We are indebted to Amparo Palacios Lopez and Siobhan some degree of diversification off the farm is common at all levels Murray for helping us link the household data with georeferenced information. We of welfare. Due to data limitations, however, the question remains also acknowledge the excellent research assistance of Marco Tiberti. We are solely as to whether this is occurring in Africa, a latecomer to the process responsible for any errors. ⇑ Corresponding author. of structural transformation. Conventional wisdom would have it E-mail addresses: benjamin.davis@fao.org (B. Davis), stefania.digiuseppe@fao. that rural households in Sub-Saharan Africa are primarily org (S. Di Giuseppe), azezza@worldbank.org (A. Zezza). http://dx.doi.org/10.1016/j.foodpol.2016.09.018 0306-9192/Ó 2016 The World Bank. Published by Elsevier Ltd. This is an open access article under the CC BY IGO license (http://creativecommons.org/licenses/by/3.0/igo/). 154 B. Davis et al. / Food Policy 67 (2017) 153–174 employed in agriculture, with relatively little agricultural wage processes, rendering them difficult for modeling. In early stages, labor, and even less non-agricultural wage labor due to limited resource-based industrialization may be geographically scattered, industrialization. but as activities that are not based on natural resources increase, Less discussed in the literature is the role of geography in deter- they tend to be located in large centers. The extent to which mining rural household income strategies. Deichmann et al. (2008) these activities will move to secondary urban centers and/or rural identify two main strands of literature that help frame the argu- areas will depend upon the policy environment (Hamer and Linn, ments around location and income diversification. First, one key 1987). empirical regularity of the rural farm/non-farm employment liter- Bringing these arguments and evidence together, it becomes ature is that at very low levels of development, non-farm activities clear that both exogenous physical location as well as the tend to be closely related to agriculture. Growth in the agricultural interaction between sectors and endogenous policy-related issues sector (e.g. due to technological change) leads to growth in the come into play in complex ways that complicate predictions of non-farm economy, thanks to the backward and forward linkages the spatial location of economic activities in rural areas. from agriculture. Taking advantage of newly available data, this paper seeks Such growth patterns are not likely to be location neutral, as to compare the income strategies of rural households in potential for agricultural growth and agro-industrial demand for Sub-Saharan Africa with those of households in other countries, agricultural products are not randomly allocated across space. Over taking into account different levels of development. Specifically, time endogenous sectoral growth biases may play a role, as infras- this paper seeks to understand the role of agriculture in the rural tructure and other investments may tend to locate where growth is economy, the profiles of households reducing their participation occurring, leading to increased spatial disparities in growth pat- in the agricultural sector, and the degree to which income terns. In Latin America, this has attracted considerable attention portfolio patterns can be linked to geographical features such as in the context of the debate on the ‘territorial approach’ to rural agro-ecological potential and urban access. development (de Ferranti et al., 2005). As sectoral policies are In order to answer these questions, we use comparable income likely to have differential impacts across space, explicitly incorpo- aggregates from 41 national household surveys with high-quality rating spatial issues into policy design can help counter territorial income data conducted across 22 developing countries, con- distortions in development patterns. structed as part of FAO’s Rural Income Generating Activities (RIGA) The second key strand of literature is the new economic geogra- project. The initial exploration of the RIGA database (Winters et al., phy debate, which focuses on the extent to which geography, as 2009, 2010; Davis et al., 2010) highlighted a number of regularities opposed to institutions, explains differential development out- concerning household patterns of income diversification in devel- comes. One main tenet of that debate is that even if soil quality oping countries. The Sub-Saharan African countries included in and climate were the same everywhere, location would still mat- the database stood out as the only countries for which specializa- ter. On the one hand, dispersion of economic activities occurs as tion in farming, as opposed to holding a diversified income portfo- firms tend to locate in areas with lower wages, and the production lio, was the norm. of non-tradable goods and services locates close to demand. Activ- That analysis was however based on data for only four countries ities connected to non-mobile inputs (such as agricultural land) in Sub-Saharan Africa: Madagascar, Malawi, Nigeria, Ghana. This will by definition be spread across space to some extent. On the paper takes advantage of more recent data from some of the same other hand, agglomeration pushes businesses to locate close to countries and additionally includes data on five more countries consumers or to the source of raw material. Businesses depending (Ethiopia, Kenya, Niger, Tanzania, Uganda), collected as part of on mobile inputs but with higher transport costs for their outputs the Living Standard Measurement Study - Integrated Surveys on would tend to have the highest gains from concentrating in partic- Agriculture (LSMS-ISA)1 project. This new set of countries accounts ular locations. for 51 percent of the Sub-Saharan African (SSA) population in 2012, Moreover, the location of economic activities across space may as opposed to 26 percent in the initial RIGA sample. While caution is be nonlinear. Fafchamps and Shilpi (2003) find for instance that in still warranted in treating this sample as representative of SSA as a Nepal, agricultural wage employment is concentrated in rural whole, its coverage is arguably much more complete. Also, we take areas close enough to cities to specialize in high-value horticulture, advantage of the geo-referencing of households and of the focus but not so close as to be taken over by unskilled ‘urban’ wage labor on agricultural activities that are two of the defining features of opportunities. Non-linearities may also be relevant when city size the LSMS-ISA datasets, in order to analyze the role of geography in is found to matter for engagement in non-farm activities shaping rural income strategies. (Fafchamps and Shilpi, 2003) or for poverty reduction The paper continues as follows. In Section 2, we present and (Christiaensen et al., 2013). Also, specialization may be dependent describe the construction of the RIGA database. In Section 3, we upon a particular market size or specific types of markets analyze the participation of rural households in income- (Fafchamps and Shilpi, 2005). generating activities and the share of income from each activity Agricultural potential and distance may interact in determining in household income, across all households and by expenditure locational advantage, occupational choices and returns to eco- quintile. We then move from the level of rural space to that of nomic activities, but relatively few empirical studies have been the rural household, examining patterns of diversification and able to assess these interactions in low-income country settings. specialization in rural income-generating activities, again In Uganda, Yamano and Kijima (2010) show how soil fertility is across all households, and by expenditure quintile. We also use positively associated with crop income, but not with non-farm measures of stochastic dominance to characterize the relationship income, whereas remoteness and poor road infrastructure lead to between types of income-generating strategies and welfare. lower crop income. In Bangladesh, Deichmann et al. (2008) find In Section 4, we examine the role of location in income generation that the higher the distance to an urban ‘growth pole’, the lower strategies in a multivariate framework, and we conclude in the level of employment in high-return non-farm jobs, particularly Section 5. in areas with good agricultural potential. Finally, different patterns of urbanization (megacities versus growth in small towns) may be associated with development outcomes, but the incentives and constraints driving them 1 See www.worldbank.org/lsms for more information on the LSMS-ISA program of change with different stages of industrialization and urbanization the World Bank, and for full access to the data and documentation. B. Davis et al. / Food Policy 67 (2017) 153–174 155 2. The data To analyze the spatial patterns of income generation, a set of geo-referenced variables from external sources are linked to the 2.1. The RIGA database household-level data via their GPS attributes. This can only be done for the 6 LSMS-ISA datasets covering Ethiopia, Malawi, Niger, Nige- The RIGA database is constructed from a pool of several dozen ria, Tanzania, and Uganda. First, we use an aridity index as proxy Living Standards Measurement Study surveys (LSMSs) and from for agricultural potential, which is defined as the ratio between other multi-purpose household surveys made available by the mean annual precipitation and mean annual potential evapo- World Bank through a joint project with the FAO.2 The most recent transpiration (thus, a higher value of the index identifies wetter additions are the LSMS-ISA project countries (see complete list in areas).7 This is a purely physical, exogenous indicator that reflects Appendix Table A1). Each survey is representative for both urban long-term conditions in a locality. We maintain that this indicator and rural areas; only the rural sample was used for this paper.3 is superior to alternatives that embed the profitability or value of While clearly not representative of all developing countries, or all agricultural production in a given area, as those incorporate contin- of Sub-Saharan Africa, the list does cover a significant range of coun- gent factors such as prices and terms of trade. In this application, we tries, regions, and levels of development and has proven useful in value the fact that the aridity index is truly exogenous. providing insight into the income-generating activities of rural Second, we proxy market access, distance and agglomeration households in the developing world.4 effects with variables that measure the Euclidean (‘as the crow Following Davis et al. (2010), income is classified into seven cat- flies’) distance to cities of 20, 100, and 500 thousand inhabitants. egories: (1) crop production; (2) livestock production; (3) agricul- We choose this measure due to a concern with the potential endo- tural wage employment, (4) non-agricultural wage employment; geneity of travel time measures; roads and travel infrastructure (5) non-agricultural self-employment; (6) transfer; and (7) other. may be built in response to agricultural production or potential 5 All income is net of input costs. Non-agricultural wage employment (Fafchamps and Shilpi, 2005; Deichmann et al., 2008). The Eucli- and non-agricultural self-employment income have been further dis- dean distance is independent of travel infrastructure, but provides aggregated by industry using standard industrial codes, although we a reliable measure of the spatial dispersion of households with do not take advantage of this disaggregation in this study. regards to urban populations. The seven income categories are aggregated into higher level groupings depending on the type of analysis. One grouping distin- 3. The diversity of income sources in Sub-Saharan Africa guishes between agricultural (i.e. crop, livestock, and agricultural wage income) and non-agricultural activities (i.e. non-agricultural 3.1. Agriculture is still the main source of livelihoods in rural Sub- wage, non-agricultural self-employment, transfer, and other Saharan Africa income), and in a second, crop and livestock income are referred to as on-farm activities, non-agricultural wage and self- We begin by looking at the prevalence of household participa- employment income as non-farm activities, and agricultural wage tion in different activities (Table 1, Figs. 1–4).8 The discussion in employment, transfer, and other income are left as separate cate- this section is based on an analysis of the basic descriptive statistics, gories. Finally, we also use the concept of off-farm activities, which aided by a visual interpretation of scatterplots including simple includes all non-agricultural activities plus agricultural wage labor. quadratic trend lines fitted to the data.9 Strikingly, the near totality Income shares can be analyzed as the mean of income shares or of rural households in the countries of our sample are engaged in as the share of mean income. In the first instance, income shares own account agriculture. This is true in Africa (92 percent on aver- are calculated for each household, and then the mean of the house- age), but also in other regions (85 percent) (Fig. 1). While for some hold shares of each income category. In the second case, income households the importance of this participation is relatively minor, shares are calculated as the share of a given source of income over since it includes consumption of a few animals or patio crop produc- a given group of households.6 Since the household is our basic unit tion, agriculture continues to play a fundamental role in the rural of analysis, we use the mean of shares throughout this paper. household economic portfolio. It is hard to overemphasize this result, especially given its robustness across countries and income 2 Information on the RIGA database can be found at: http://www.fao.org/economic/ levels: in the vast majority of the surveys we find that more than riga/en/. 8 in 10 rural households depend to some extent on agriculture. 3 Each country has their own definition of rurality, and government definitions not Regardless of the level of GDP, agriculture continues to be the dis- comparable across countries may play some part in explaining cross-country tinctive feature of rural livelihoods. differences. While recognizing that variation in country-specific definitions of rural may explain observed differences in income composition, the available survey data do At the same time, an important share of rural households, not allow for straightforward construction of an alternative measure across all across GDP levels, participate in non-farm (non-agricultural wage countries. We thus use the government definition of what constitutes rurality. labor and self-employment, Fig. 2). Globally, shares vary widely, Further, rurality is identified via household domicile, not the location of the job – a ranging from 24 percent (Ethiopia and Nigeria 2004) to over 90 number of labor activities identified as rural may actually be located in nearby urban areas. percent (Bolivia 2005). The simple mean non-farm participation 4 Details of the construction of the income aggregates can be found in Carletto et al. share for African countries is 44 percent, which is 10 percentage (2007). points lower than for non-African countries. Among African coun- 5 Agricultural income values all production, both consumed on farm and marketed; tries, the highest share is observed in Niger, at 65 percent. A similar transfers from both public and private sources (such as remittances) are included; share of households obtains income from public or private transfer other income covers a variety of non-labor sources of income, such as rental income or interest from savings. income, although it spans an even wider range, from 3 percent of 6 The two measures have different meanings. The mean of shares more accurately households in Nigeria in 2010 to almost 90 percent in Malawi in reflects a household-level income generating strategy, regardless of the magnitude of 2004. When including non-farm, transfers and other sources of income. The share of means reflects the importance of a given income source in the income, the vast majority of rural households across GDP levels aggregate income of rural households in general or for any given group of households. The two measures will give similar results if the distribution of the shares of a given 7 source of income is constant over the income distribution, which is clearly not always CGIAR (2014). 8 the case. If, for example, those households with the highest share of crop income are A household is considered to participate in an activity if it derives income out of also the households with the highest quantity of crop income, then the share of that activity. 9 agricultural income in total income (over a given group of households) using the We considered performing the analysis via a multivariate regression framework, share of means will be greater than the share using the mean of shares. but the sample size is too small. 156 B. Davis et al. / Food Policy 67 (2017) 153–174 Participation in on-farm activities 100 (3) + (4) + (5) + (6) + (7) 91% 77% 98% 74% 31% 97% Off-farm 80 total 60 Transfers (6) + (7) (%) & other 58% 29% 91% 47% 90% 8% 40 20 Non-farm (4) + (5) 54% 29% 92% 44% 24% 65% total 0 6 7 8 9 Group III GDP (log) On-farm (1) + (2) 85% 54% 99% 92% 86% 98% total Participation Africa Participation Non-Africa Overall Trend Agricultural Fig. 1. Percentage of rural households participating in on farm activities, by per + (6) + (7) (4) + (5) 83% 67% 96% capita GDP in 2005 PPP dollars. 70% 30% 93% Non- Total Participation in non-farm activities (1) + (2) + (3) 100 Agricultural 89% 64% 99% 92% 84% 99% Group II total 80 60 Other 24% 16% 59% 9% 1% 0% (%) (7) 40 Transfers 42% 89% 51% 26% 89% 3% 20 (6) 0 employment 6 7 8 9 Non-farm 34% 16% 29% 83% 60% GDP (log) 2% self- Participation Africa Participation Non-Africa (5) Overall Trend employment Source: Authors’ calculations based on the RIGA database. See Table A2 for full results by country. Fig. 2. Percentage of rural households participating in non-farm activities, by per Non-farm capita GDP in 2005 PPP dollars. 15% 25% 35% 18% 51% 6% wage (4) Participation in non-agricultural activities employment 100 Agricultural 18% 55% 27% 49% 1% 5% wage 80 (3) 60 Agriculture - Income-generating activity (%) Livestock Participation in income-generating activities, rural households. 61% 38% 67% 98% 80% 10% 40 (2) 20 Agriculture- 0 89% 81% 97% 79% 98% 40% Group I crops 6 7 8 9 (1) GDP (log) Participation Africa Participation Non-Africa Simple mean Simple mean Overall Trend Maximum Maximum Minimum Minimum Fig. 3. Percentage of rural households participating in non-agricultural activities, by per capita GDP in 2005 PPP dollars. 13 countries) have some form of off-farm income (see last column in Table 1), 9 countries) (14 surveys, (27 surveys, Non-Africa with rates higher in other regions (91 percent on average) than in Africa (74 percent). Participation in non-agricultural wage labor, Africa Table 1 on the other hand, shows a clear increase by levels of GDP (Fig. 4), with the African countries in our sample (shown in blue or darker B. Davis et al. / Food Policy 67 (2017) 153–174 157 Participation in non-agricultural wage labor 100 (3) + (4) + (5) + (6) + (7) 37% 15% 52% 67% 39% 88% Off-farm 80 total 60 Transfers (%) (6) + (7) & other 21% 18% 62% 8% 2% 3% 40 20 (4) + (5) 23% 40% 35% 13% 62% 6% Non- farm total 0 6 7 8 9 Group III (1) + (2) GDP (log) 63% 48% 85% 33% 12% 61% farm total On- Participation Africa Participation Non-Africa Overall Trend agricultural + (6) + (7) Fig. 4. Percentage of rural households participating in non-agricultural wage labor, (4) + (5) 31% 12% 45% 54% 28% 80% by per capita GDP in 2005 PPP dollars. Non- total hue) reporting relatively lower participation rates (from 2 percent Agricultural in Ethiopia to 25 percent in Kenya and Uganda 2009/10) than other Group II (1) + (2) 69% 55% 88% 46% 20% 72% countries at the same level of GDP. + (3) total Turning to income shares (Table 2, Figs. 5–10), the countries in our African sample show a tendency towards on-farm sources of income (i.e. agricultural income minus agricultural wages): they Other 13% 1% 3% 3% 0% 0% have higher shares of on-farm income (63 percent) and lower (7) shares of non-farm wage income (8 percent), compared with coun- tries of other regions (33 and 21 percent respectively), including Transfers 19% 15% 60% 7% 3% 0% those at similar levels of GDP. All the countries from Sub-Saharan (6) Africa in this sample earn at least 55 percent of their income from agricultural sources, reaching approximately 80 percent in a num- employment ber of countries (Ethiopia, Madagascar, Malawi, and Nigeria in Non-farm 2004). Similarly, on-farm income accounts for more than 50 per- 15% 29% 14% 36% 4% 1% cent in all but one country (Kenya, at 48 percent). Combined with self- (5) the observation above on the virtually universal level of participa- tion in agricultural activities in the Sub-Saharan Africa subsample, employment Source: Authors’ calculations based on the RIGA database. See Table A3 for full results by country. this reinforces the message of agriculture still dominating the rural Non-farm 15% 21% 39% economy on the continent. Despite the fact that non-agricultural 8% 2% 9% wage activities are ubiquitous (70 percent participation), they still (4) account on average for only about one third of total earnings. African countries, particularly those in West Africa, generally employment Agricultural have less income from agricultural wage labor (Fig. 9). For Sub- 15% 13% 25% 5% 1% 2% Saharan Africa overall, the maximum share is 15 percent in Malawi; wage in West Africa, it is a mere 3 percent in Niger. This is an important (3) insight, as some of the expected beneficial effects of high food Share of income-generating activities in total rural household income. prices for the poor have been hypothesized to materialize via higher Agriculture - Income-generating activity agricultural wages (Ivanic and Martin, 2008). In Africa this is less Livestock 16% 23% -1% 9% 3% 9% likely to be the case, compared to countries in Asia and Latin America where agricultural wage income shares in the order of (2) 15–25 percent are far more common. The features of agricultural wage employment are often linked to the peculiarities of the Agriculture- institutions of rural communities (e.g. ganyu labor in Malawi), 55% 32% 76% 25% 53% Group I 4% and possibly with the prevalence of plantations and cash crops. crops (1) Overall, the share of non-agricultural income among rural households increases with increasing levels of GDP per capita Simple mean Simple mean (Fig. 5). The importance of on-farm (crop and livestock) sources Maximum Maximum Minimum Minimum of income gradually decreases (Fig. 6) as they are replaced by non-agricultural wage income (Fig. 7) and public and private trans- fers (Fig. 8). In our sample of African countries, the largest share of income from non-farm sources is recorded in Nigeria (40 percent) 13 countries) and the lowest in Ethiopia (6 percent). Transfer income shares are 9 countries) (14 surveys, (27 surveys, Non-Africa highest in Kenya (19 percent) and lowest in Nigeria (1 percent), Africa and within this range several countries record substantial shares Table 2 of 9–10 percent, which is compatible with the documented impor- tance of migrant remittances from urban areas as well as from 158 B. Davis et al. / Food Policy 67 (2017) 153–174 Share of non-agricultural income Share of transfer income 100 100 80 80 60 60 (%) (%) 40 40 20 20 0 0 6 7 8 9 6 7 8 9 GDP (log) GDP (log) Africa Non-Africa Africa Non-Africa Overall Trend Overall Trend Fig. 5. Share of rural households’ non-agricultural income, by per capita GDP in Fig. 8. Share of rural households’ transfer income, by per capita GDP in 2005 PPP 2005 PPP dollars. dollars. Share of on-farm income Share of agricultural wage income 100 100 80 80 60 60 (%) (%) 40 40 20 20 0 0 6 7 8 9 6 7 8 9 GDP (log) GDP (log) Africa Non-Africa Africa Non-Africa Overall Trend Overall Trend Fig. 6. Share of rural households’ on farm income, by per capita GDP in 2005 PPP Fig. 9. Share of rural households’ agricultural wage income, by per capita GDP in dollars. 2005 PPP dollars. Share of non-agricultural wage income Participation in non-agricultural self-employment activites 100 100 80 80 60 60 (%) (%) 40 40 20 20 0 0 6 7 8 9 6 7 8 9 GDP (log) GDP (log) Africa Non-Africa Participation Africa Participation Non-Africa Overall Trend Overall Trend Fig. 7. Share of rural households’ non-agricultural wage income, by per capita GDP Fig. 10. Percentage of rural households participating in no-agricultural self- in 2005 PPP dollars. employment activities, by per capita GDP in 2005 PPP dollars. abroad. Broadly speaking, these values are comparable to the non-agricultural self-employment (Figs. 10 and 11), where there ranges observed in non-African countries. does not appear to be any clear association with GDP levels. Lastly, African and non-African countries do not appear to be One important difference between the African and non-African dissimilar in terms of participation in or shares of income from countries in this sample is in the composition of non-agricultural B. Davis et al. / Food Policy 67 (2017) 153–174 159 Share of non-agricultural self-employment income Among rural households in the countries of our African sample, 100 specialization in on-farm activities continues to be the norm (prac- ticed by 52 percent of households on average), ranging from one- third of households in Kenya to 83 percent in Ethiopia (Table 3). 80 Among all countries, with the exception of Niger, a majority of households specialize in on-farm activities. This result is quite dif- 60 ferent from the non-African households in our sample of countries, (%) where only 21 percent of households on average specialize in 40 farming. Within this group, in only two countries do the majority of households specialize in on farm activities. Diversification is 20 the norm; 45 percent of households fall into the diversified cate- gories, on average. The relative differences between the African 0 and non-African countries with increasing levels of per capita 6 7 8 9 GDP (log) GDP can be seen in Figs. 12 and 13. Rural households in the African Africa Non-Africa country are clustered above the trend line in the former graph, and Overall Trend below the trend line in the latter. When rural households in non-African countries do specialize, Fig. 11. Share of rural households’ non-agricultural self-employment income, by they mostly specialize in on-farm activities, although the percent- per capita GDP in 2005 PPP dollars. ages become lower as the per capita GDP increases. At higher GDP levels, specialization in non-agricultural wage labor becomes more income. While the shares of non-farm self-employment income are important for both African and non-African countries (Fig. 14). No comparable across countries in the two groups (14–15 percent), distinct association between GDP levels and specialization in agri- the average share of non-farm wage employment is generally cultural wage or self-employment is apparent for non-African coun- much smaller in SSA, with a maximum level of 15 percent in Kenya tries, while for African countries the share appears to increase in 2005, compared to an average of 21 percent (and peaks of nearly (Fig. 15). Taken together, these observations suggest a gradual tran- 40 percent) in the non-African component of the sample. This is in sition from heavy reliance on farming to a greater reliance on non- line with recent studies of the structural transformation of African farm wage employment, with non-farm self-employment the activ- economies that have used similar microdata and have found that ity of choice for a more or less constant share of households as rural employment in the industry and service sectors is largely in development occurs. This essentially confirms the trends observed own-account rather than wage occupations, and in services more based on the crude income shares data (Figs. 5–11 above). than in industrial sectors (McCullough, 2015). Interestingly, only one of the African countries in our sample has more than 5 percent of households specializing in transfer income 3.2. Diversification and specialization (Kenya, with 9 percent). Meanwhile, in non-African countries, it is not at all uncommon for more than 5 percent of households to The results presented thus far suggest that rural households receive more than three quarters of their earnings from transfers. employ a wide range of income-generating activities, although It is hard to generate robust conclusions from these observations, rural households in African countries are more dependent on agri- as transfer income is a mixed bag of several sources (e.g. social pro- culture then rural households in other countries. The question tection programs, pensions, migrant remittances, and more) with remains, however, whether households specialize in activities very different institutional and socio-economic determinants. How- (with diversity in activities across households in the rural space) ever, it is worth noting that very few African households are relying or, whether households themselves diversify income-generating mostly on these sources of income for their livelihoods. Despite activities. If we observe a decline in the share of agricultural widespread migration (De Brauw et al., 2014; Ratha et al., 2011) income, that could be the result of a few households moving out and the expansion of social programs (Garcia and Moore, 2012), pro- of agriculture entirely, or of many households marginally reducing ductive occupations are what keep most households afloat. their share of income from agriculture. To explore this question and understand the extent to which 3.3. Income sources, returns to different activities, and welfare levels households in Africa specialize in agricultural or other sectors rel- ative to households in other regions, we examine the degree of spe- The previous sections illustrated the diversified nature of the cialization and diversification by defining a household as rural economies in all the countries of our sample, including those specialized if it receives more than 75 percent of its income from of Sub-Saharan Africa. Exploring the composition of income at the a single source and diversified if no single source is greater than household level is essential to understanding the strategies and that amount.10,11 assets that households rely on in order to lift themselves out of poverty. The available literature shows that within both agricul- 10 Other definitions of diversification and specialization are possible. Davis et al. tural and non-agricultural income-generating activities, there is (2010) used 100% and 50% of income from a single source as alternative thresholds in often a dualism between high and low return sub-sectors (Nagler order to examine robustness. They find that the extent of diversification is affected by and Naudé, 2014). High-return activities often have significant bar- the choice of the threshold, which drops to around 10% or less in all cases when using riers to entry or require accumulation in terms of land, human cap- the 50% definition of specialization, climbing to around 90% with the 100% definition. The broad patterns by country and by level of welfare, however, did not change with ital, and other productive assets (Haggblade et al., 2007; Davis choice of the threshold. Alternative groupings of income categories are also possible, et al., 2010). In contrast, a low productivity segment usually serves such as joining together agricultural and non-agricultural wage labor, or non- as a source of residual income or subsistence food production and agricultural wage labor and non-agricultural self-employment, which would increase as a refuge for the rural poor.12 Entry barriers may end up confining the share of household specializing in these new categories. 11 Note that we are constrained from delving into the details of diversification due 12 to the way that household survey data are often collected. The apparent diversifi- See Lanjouw and Lanjouw (2001) and Lanjouw and Feder (2001) for a general cation may derive from aggregation across seasons (with seasonal specialization by discussion relevant to non-farm activities and Fafchamps and Shilpi (2003) for Nepal households) or across individuals (with specialization by individual household and Azzarri et al. (2006) for Malawi, for example, regarding the role of agricultural members). wage labor. 160 B. Davis et al. / Food Policy 67 (2017) 153–174 Table 3 Percent of rural household with diversified and specialized income-generating activities. Per Capita GDP, Principal Household Income Source (>=75% of Total Income) Diverse Income Country and year PPP Constant Non Ag Portfolio Ag Wage Self Emp Transfers Other Farm 2005, USD Wage Ethiopia 2012 453 10% 1% 1% 2% 1% 1% 83% Ghana 1992 949 22% 1% 4% 10% 3% 0% 60% Ghana 1998 1,051 24% 1% 6% 15% 3% 0% 50% Ghana 2005 1,222 23% 2% 6% 20% 5% 0% 44% Kenya 2005 1,340 35% 4% 10% 6% 9% 1% 36% African Countries Madagascar 1993 895 31% 1% 3% 4% 1% 0% 59% Malawi 2004 640 37% 3% 4% 3% 1% 0% 52% Malawi 2011 785 29% 7% 5% 3% 1% 0% 54% Niger 2011 535 46% 0% 2% 10% 3% 0% 38% Nigeria 2004 1,707 14% 0% 6% 7% 1% 0% 72% Nigeria 2010 2,120 20% 0% 8% 22% 0% 1% 49% Tanzania 2009 1,240 35% 1% 3% 5% 4% 0% 53% Uganda 2005/06 966 35% 6% 7% 8% 3% 0% 41% Uganda 2009/10 1,130 39% 3% 5% 8% 2% 0% 43% Simple mean 29% 2% 5% 9% 3% 0% 52% Albania 2002 4,710 51% 1% 7% 3% 11% 0% 27% Albania 2005 5,463 55% 1% 9% 5% 10% 0% 19% Bangladesh 2000 901 52% 11% 12% 10% 5% 2% 6% Bangladesh 2005 1,068 53% 9% 15% 8% 4% 2% 10% Bolivia 2005 3,758 51% 4% 11% 22% 5% 1% 7% Bulgaria 1995 6,930 50% 7% 15% 2% 21% 0% 5% Bulgaria 2001 7,348 41% 2% 9% 1% 43% 0% 3% Ecuador 1995 5,658 46% 13% 12% 9% 2% 1% 17% Ecuador 1998 5,862 30% 12% 11% 12% 6% 4% 24% Guatemala 2000 3,966 55% 9% 13% 6% 5% 0% 13% Non African Countries Guatemala 2006 4,178 52% 9% 17% 5% 7% 0% 9% Indonesia 1993 2,487 24% 5% 8% 15% 11% 1% 35% Indonesia 2000 2,724 42% 6% 14% 10% 11% 1% 16% Nepal 1996 829 52% 7% 6% 4% 3% 0% 27% Nepal 2003 926 53% 4% 12% 5% 7% 0% 19% Nicaragua 1998 1,961 35% 16% 15% 6% 3% 0% 25% Nicaragua 2001 2,145 44% 13% 14% 6% 1% 0% 22% Nicaragua 2005 2,311 42% 13% 10% 5% 4% 0% 25% Pakistan 1991 1,719 24% 3% 20% 14% 1% 0% 37% Pakistan 2001 1,923 36% 5% 19% 7% 9% 2% 22% Panama 1997 7,554 48% 8% 23% 6% 6% 1% 8% Panama 2003 8,267 49% 10% 20% 10% 7% 0% 5% Tajikistan 2003 1,283 54% 5% 4% 1% 5% 0% 32% Tajikistan 2007 1,656 50% 1% 5% 0% 1% 0% 43% Vietnam 1992 997 35% 3% 2% 15% 1% 0% 44% Vietnam 1998 1,448 44% 2% 2% 13% 1% 0% 38% Vietnam 2002 1,780 48% 2% 12% 10% 2% 0% 25% Simple mean 45% 7% 12% 8% 7% 1% 21% Note: Bordered cells indicate the category with the highest percentage in each country. Shaded cells indicate the specialization category (i.e. excluding diversified) with the highest percentage. more marginalized households in low-return sub-sectors, preventing Chawanote and Barrett (2013) find the existence of an ‘‘occupational them from taking advantage of the opportunities offered by the ladder” in rural Thailand, in which transitions into the rural non- more dynamic segments of the rural economy (Reardon et al., farm economy lead to increased income, and transitions into farm- 2000). In what follows, our focus will remain at the level of the more ing lead to reduced income. Using data similar to those in our African aggregated income-generating categories we described earlier, as subset, Nagler and Naudé (2014) find that the productivity of rural examining specific industries and occupations is intractable in a household enterprises suffers from the costs associated with large cross-country study such as this. distances, rural isolation, and low population density, and that The literature suggests that households participating in higher- household enterprises that emerge out of necessity rather than return rural non-farm activities are richer and have more upward opportunity are systematically less productive. income mobility (Barrett et al., 2001; Bezu et al., 2012; Bezu and To explore the relationship across countries between rural Barrett, 2012, among others), a relationship that holds up in cross income-generating activities and welfare, we start by examining country studies and across increasing levels of development (Davis activities by expenditure quintiles for each country. Fig. 16a charts et al., 2010; Winters et al., 2010). Recent studies focus on the dynam- income shares by expenditure quintile for all countries in the ics of household participation in rural non-farm activities. Bezu and African sample. Focusing on on-farm activities, the darkest color, Barrett (2012) find that households able to accumulate capital, or we see a sharp decrease in the share of on-farm income with that have more adult labor or better access to credit and savings, increasing levels of welfare, dropping from around 50 percent of are more able to access high-return rural non-farm activities. income in the poorest quintile in most countries, to less than 20 B. Davis et al. / Food Policy 67 (2017) 153–174 161 Share of specializing on-farm Share specializing non-agricultural self-employment 100 100 80 80 60 60 (%) (%) 40 40 20 20 0 0 6 7 8 9 6 7 8 9 GDP (log) GDP (log) Africa Non-Africa Africa Non-Africa Overall Trend Overall Trend Fig. 12. Share of rural households specializing on farm, by per capita GDP in 2005 Fig. 15. Share of rural households specializing in non-agricultural self-employ- PPP dollars. ment, by per capita GDP in 2005 PPP dollars. Share with diversified income portfolio countries in the African sample is not as clear in the non-African countries in Fig. 16b. Here Bangladesh, Bulgaria, Nepal, Pakistan 80 and Tajikistan show the opposite trend: the share of on-farm activ- ities increases with welfare. 60 On the other hand, participation in, and shares of income from, agricultural wage labor show for the most part a negative correla- tion with the level of expenditure, for both African and non-African (%) 40 countries. With the exception of those countries that have negligi- ble agricultural labor wage markets, poorer rural households tend 20 to have a higher rate of participation in agricultural wage employ- ment. Similarly, the share of income from agricultural wage labor is more important for poorer households in these countries, and 0 6 7 8 9 the relationship holds regardless of the level of development. GDP (log) Participation in rural non-farm activities can reflect engage- Africa Non-Africa ment in either high or low-return sub-sectors. Rural non-farm Overall Trend activities may or may not be countercyclical with agriculture, both within and between years, and particularly if not highly correlated Fig. 13. Share of rural households with diversified income portfolio, by per capita with agriculture, they can serve as a consumption smoothing or GDP in 2005 PPP dollars. risk insurance mechanism. Thus, the results raise the question of whether diversification is a strategy for households to manage risk and overcome market failures, or whether it represents specializa- Share specializing non-agricultural wage tion within the household, in which some members participate in 100 certain activities because they have a comparative advantage in those activities. If the latter is the case and it tends to be the young 80 who are involved in off-farm activities, diversification may simply reflect a transition period as the household shifts away from on- 60 farm activities. McCaig and Pavcnik (2014) investigate such an (%) hypothesis for Vietnam and find that less than 20 percent of the 40 shift of labor out of agriculture can be attributed to changing demographics (what they call a between-cohort as opposed to a 20 within-cohort effect). The empirical relationship between income-generating strategies, diversification and welfare is thus not straightfor- 0 6 7 8 9 ward. Lower diversification at higher levels of welfare could GDP (log) be a sign that those at lower income levels are using diversifi- Africa Non-Africa cation to overcome market imperfections (e.g. cash constraints Overall Trend to finance agriculture, or multiple activities to spread risk). Alternatively, a reduction in diversification at lower income Fig. 14. Share of rural households specializing in non-agricultural wage, by per capita GDP in 2005 PPP dollars. levels could be a sign of an inability to overcome barriers to entry in a second activity, thus indicating that poorer house- holds are limited from further diversification. Higher diversifica- percent in the richest quintile. The drop in on-farm sources of tion among richer households could be a sign of using income is made up by the increasing importance of off-farm (i.e. profitability in one activity to overcome threshold barriers to non-agricultural wage and self-employment) sources of income entry in another activity, or complementary use of assets for better-off rural households. The clear trend evident from the between activities. 162 B. Davis et al. / Food Policy 67 (2017) 153–174 Share of total income from main income generating activities (Africa) by expenditure quintiles 100 80 Shares of Income (%) 40 60 20 0 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 04 11 09 05 12 11 92 06 98 10 11 05 04 10 93 N A A A A A A A A A L L H ER D A A TZ G G G G G H H H ET A KE M M N G G G M U U U N N On-farm Acti vi ties Agric ultural W ages Transfers and Other Non-Labour Sou rces Non-farm Activ ities Note: Expenditure quintiles move from poorer to richer, countries are sorted by increasing GDP Fig. 16a. Share of total income from main income generating activities (Africa) by expenditure quintiles. Share of total income from main income generating activities (Non-Africa) by expenditure quintiles 100 80 Shares of Income (%) 40 60 20 0 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 BN 96 VI 3 97 03 00 TA 5 VI 1 01 5 00 AL 6 EC 5 BU 8 BU 5 PA 1 BN 2 03 TA 8 07 PA 2 N 8 01 IN 5 IN 3 BO 0 02 EC 5 0 9 9 0 9 0 0 9 9 0 9 0 0 0 9 9 0 K K A A L L U U N N G EP EP E G E E L IC IC IC D D J J B B PA U U VI PA AL N N N N G G On-farm Ac ti vities Agricultural W ages Transfers and Other Non-Labour Sou rces Non-farm Activiti es Note: Expenditure quintiles move from poorer to richer, countries are sorted by increasing GDP Fig. 16b. Share of total income from main income generating activities (non-Africa) by expenditure quintiles. The inability to conceptually sign a priori the correlation diversified portfolio of income-generating strategies shows few between diversification and household welfare status emerges consistent patterns by quintile of per capita consumption expendi- from the data. Fig. 17 explores the relationship between diversifi- ture in our sample countries, in both our African and non-African cation, specialization and household expenditure for the countries countries (Figs. 17a and 17b). A clear pattern emerges, however, in our African sample. The share of rural households with a among the African countries, in terms of the share of households B. Davis et al. / Food Policy 67 (2017) 153–174 163 specializing in on-farm activities. Here, the share of households in these activities are prominent, such as Nigeria, Ghana, Malawi most countries decreases with increasing consumption expendi- and Uganda. ture levels. Conversely, the share of households specializing in Measures of stochastic dominance can complement this analy- self-employment activities and non-agricultural wage labor sis by offering a more systematic approach at characterizing the increases with expenditures, at least for those countries where association between household income specialization strategies Share of households with diversified or specialized (Non-Africa) income portfolios, by expenditure quintiles 100 80 Share (%) 40 60 20 0 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 12345 BN 96 VI 3 BU 5 PA 1 PA 7 03 00 BN 92 IN 5 EC 5 TA 5 TA 8 VI 1 PA 2 N 1 98 N 1 93 G 05 BO 0 G 00 AL 6 AL 2 EC 5 98 VI 3 07 0 9 9 0 0 9 9 0 9 0 0 0 0 0 0 0 0 L L N N EP EP E E K E K A A G G IC IC IC D D L U U J B B J BU U U PA IN N N N Div ers ified Specialized in On- farm Activities Specializ ed in non-agr ic wage Specialized in agric wage Specialized in self-employment Specialized in non-labor income Specialized in transfers Notes: Surveys sorted by increasing per capita GDP. Fig. 17a. Share of households with diversified or specialized income portfolios, by expenditure quintiles (Non-Africa). Share of households with diversified or specialized (Africa) income portfolios, by expenditure quintiles 100 80 Share (%) 40 60 20 0 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 12 11 04 11 09 05 93 92 06 11 05 04 10 98 10 L AL A A A A A A A A A H ER D N A TZ H H H G G G G G ET A KE M M N M G G G U U U N N Diversified Specialized in On-farm Activities Specialized in non-agri c wage Specialized in agric wage Specialized in self-employment Specialized in non-labor income Specialized in transfers Notes: Surveys sorted by increasing per capita GDP. Fig. 17b. Share of households with diversified or specialized income portfolios, by expenditure quintiles (Africa). 164 B. Davis et al. / Food Policy 67 (2017) 153–174 and the level of household welfare. Stochastic dominance allows can either be diversified, or fall within one of six specialization cat- for comparing income from different sources and establishing egories.15 In the multinomial logit, k À 1 models are estimated for whether one source of income is associated with higher levels of any outcome consisting of k unordered categories. Parameter esti- welfare than others. For each of four of the African countries, cov- mates are then interpreted with reference to the excluded base cat- ering six data sets—Malawi (2011), Niger (2011), Tanzania (2009 egory (farm specialization in our case). For a unit change in the and 2010) and Uganda (2010 and 2011)—we plot cumulative den- regressor, the logit of the model outcome relative to the reference sity functions (cdf) of consumption expenditures for households in group is expected to change by its parameter estimate, holding other different specialization categories (excluding transfer and other variables constant (UCLA, 2014). income for clarity of presentation). If cdf lines do not intersect, Transforming a continuous variable (income, or income shares then we can say that one strategy stochastically dominates another which we could have used as the dependent variable) into a cate- in terms of per capita expenditure (Fig. 18).13 gorical one (specialization categories, which is what we use) leads Across all countries, specialization in off-farm activities (that is, to a loss of information, which should never be taken lightly. In this non-agricultural wage income and self-employment) stochasti- case, that loss of information is more than compensated for by the cally dominates other household income-generating strategies, in fact that using mutually exclusive categories allows us to interpret terms of per capita expenditure (the same analysis, not reported, the data not only in terms of greater or lower involvement in agri- performed over total household income returns the same order- culture, but also in terms of the sector towards which households ing). These are followed by on-farm specialization and diversified lean as they move away from on-farm specialization. The basic strategies, and then finally agricultural wage labor which is clearly question we aim to address is whether recent growth in rural associated with the lowest levels of welfare.14 Overall, these obser- Africa has been accompanied by less structural transformation of vations confirm the common finding in the literature that increased the rural economy than one would expect, given the secular trends reliance on non-farm income, particularly in wage employment, is observed elsewhere. One advantage of the multinomial logistic strongly associated with higher levels of overall household welfare, regression is that it allows for the use of farm specializers as the and lower likelihood of being in poverty. reference category. As we use on-farm specialization as the base category, the coefficients on the main variables of interest can be interpreted16 in terms of association with higher or lower likelihood 4. Modeling location and strategic income choices in LSMS-ISA that a household specializes in non-farm self-employment, non-farm countries wage, or agricultural wages relative to specializing in farming. Given the associations noted above between income strategies and welfare, 4.1. Estimation approach it clearly matters what households do if they do not specialize in farming.17 The other advantage is that since we are working with As we have noted earlier, much of the literature on rural non- six countries, employing categories that use the same cut-off points farm income in developing countries has sought to explain how increases the comparability of the results. asset endowments and barriers to entry tend to push or pull differ- Previous studies have discussed the role of other key household ent households and individuals into different activities. The signif- characteristics, namely different forms of capital (human, natural, icance for welfare and poverty analysis and policy has been physical, social), and these findings are relatively consistent and established in the previous section. Location is an important factor robust across studies. One concern with that evidence, however, in determining households’ income strategy decisions, but the lit- is the extent to which different levels and composition of assets erature is relatively silent on this point, primarily due to the lack of may in fact be endogenous to decisions regarding the income gen- data that would allow for spatially explicit analysis. The geo- eration strategy. In this paper, the primary interest is to gauge the referenced household data that we use makes it possible to begin extent to which truly exogenous factors like climate and distance filling this gap. Since we focus on the rural portion of the sample, from urban centers affect household specialization and diversifica- we do not discuss issues related to exits from agriculture through tion decisions. Admittedly, distance may itself be endogenous, as household migration to urban areas. existing employment opportunities clearly play a role in a house- In what follows, our approach is similar to a meta-regression hold’s decision on where to live, but we will for convenience put analysis in that: (i) common metrics are used for each of the coun- that consideration aside for this discussion. To gauge the effects tries analyzed, (ii) explanatory variables for each country have of distance, market access and agglomeration, we employ the vari- been created in a uniform manner, and (iii) a standard regression ables described in Section 2 that measure Euclidean distance in model is employed in each case. This approach minimizes the pos- kilometers to cities of 20, 100, and 500 thousands inhabitants. sibility that differences in results are driven by differences in the For each country regression, we therefore estimate four variants: variables used or in the empirical approach, and facilitates our one per each of the distance variables employed. The reason for dif- comparisons of results across countries. ferentiating the analysis of distance by city size is linked to the Our modeling approach is to employ a multinomial logit model consideration that secondary urban centers offer jobs that demand (separately for each country) to assess the association of location a different set of skills compared to jobs in large cities, with impli- with the likelihood that a household diversifies or specializes out cations for poverty reduction. Poor rural households with limited of farming, controlling for other household characteristics. The human capital may be better able to capture the opportunities choice of the multinomial logit is motivated by the fact that we offered by secondary towns than those linked to the metropoles have several unordered but mutually exclusive categories that or megacities, and the features of the structural transformation we use to characterize household income strategies: a household of the economy accompanying urbanization may differ depending 13 15 We performed pairwise tests of stochastic dominance and they confirm the For the econometric estimation we reduce the specialization categories to five, as overall message from Fig. 18 that non-agricultural wage and self-employment we collapse ‘transfers’ and ‘other income’ into one category. 16 specialization tend to stochastically dominate the other income generating strategies. With the necessary transformations needed to obtain relative risk ratios. 17 The tests are available from the authors. For interpretation of color in Fig. 18, the We also experimented with running a similar analysis using a standard OLS reader is referred to the web version of this article. regression with the share of income from agriculture as the dependent variable and 14 The one exception is specialization in agricultural wage labor in Niger, which the results (not reported but available on request) were compatible with those we includes less than one percent of households, but with relatively high incomes. present below but less informative. B. Davis et al. / Food Policy 67 (2017) 153–174 165 M a la wi 2 0 1 1 N i g er 2 0 1 1 1 1 .8 .8 .6 .6 .4 .4 .2 .2 0 0 0 50000 100000 150000 0 200000 400000 600000 800000 1000000 exp_pc exp_pc diverse farm diverse farm agr wge non agr wge agr wge non agr wge sel emp sel emp T an za n ia 2 0 0 9 T a n z a n ia 2 0 1 0 1 1 .8 .8 .6 .6 .4 .4 .2 .2 0 0 0 20000 40000 60000 80000 100000 0 200000 400000 600000 800000 1000000 exp_pc exp_pc diverse farm diverse farm agr wge non agr wge agr wge non agr wge sel emp sel emp Uganda 2010 Uganda 2011 1 1 .8 .8 .6 .6 .4 .4 .2 .2 0 0 0 20000 40000 60000 80000 100000 0 50000 100000 150000 exp_pc exp_pc diverse farm diverse farm agr wge non agr wge agr wge non agr wge sel emp sel emp Fig. 18. Cumulative per capita expenditure distributions, by income-generating strategy. on whether urbanization is dominated by the expansion of metro- Agricultural potential is proxied by an aridity index, also poles or accompanied by growth in secondary urban centers described in Section 2 above. To capture the non-linearities in (Christiaensen et al., 2013; Hamer and Linn, 1987). Using a cross the relationship between specialization/diversification and dis- section of 51 developing country data, Christiaensen et al. (2013, tance, we introduce both a quadratic term for distance, and inter- p. 444) find that ‘‘only rural diversification and migration to sec- action terms between distance and aridity. This analysis enables ondary towns is statistically contributing to poverty reduction, measuring the extent of impact of location effects (i.e. agricultural while migration to the metropoles is not.” potential, distance, and their interaction) on the choice of income- 166 B. Davis et al. / Food Policy 67 (2017) 153–174 generating strategies. In specifying our model using distance to urban centers of different sizes, we are also interested in gauging Integration how these relationships may vary when one considers distance Low High to small towns, as compared to distance to mid-size and large cities. Agricultural Potential The vector of regressors includes a range of additional house- Low hold characteristics that are known to impact decisions about occupational choice and income-generating strategies: separate (?) + agricultural and non-agricultural wealth indexes, and an index of access to basic infrastructure (all calculated using principal compo- nent analysis); household demographic and composition charac- teristics (household size, age and gender of the head, number of High working age members, share of female working age adults); and - + (?) variables to measure key households assets (education of the head, land owned).18 Based on the theoretical and empirical literature reviewed ear- lier in this paper, we have some clear expectations as per the sign Fig. 19. Matrix of expected relationship between specialization in non-agricultural of the correlation between household endowments and sectors of activities, agricultural potential, and integration into urban areas. specialization, with land strongly associated with agricultural activities, education strongly associated with non-farm (particu- larly) wage activities, and low levels of assets across the board the main results emerging from the analysis, we use graphs to being associated with agricultural wage employment. demonstrate the broad directions and non-linearities in the main To weigh the a priori expectations regarding the association variables of interest (Long and Freese, 2014). between the key location variables (distance and aridity) and Fig. 20 reports how the predicted probabilities of being in the diversification or specialization outside of agriculture, we provide diversified and in the main non-farm specialization categories a 2 Â 2 matrix organized around high/low integration and agricul- change with distance. To convey the effect of distance separately tural potential (Fig. 19). for high and low potential areas, we graph predicted probability In high potential, high integration19 areas, one expects both farm estimated at the 10th (solid line, low potential) and 90th (dashed and non-farm activities to thrive, with non-farm shares dominating line, high potential) percentile of the normalized aridity index. as integration levels increase. In low potential, high integration The same graphs are reported by distance to cities of different size areas, the expectation is for non-farm activities to dominate as peo- (20 thousand plus, 100 thousand plus or 500 thousand plus inhab- ple reap off-farm opportunities, as farming does not hold much pro- itants). Since one objective of the study is to characterize how (and mise given the unfavorable conditions. Meanwhile, in low which) households transition from agriculture to other sectors, we integration, high potential areas, the expectation is for farming to focus on the sectors that identify more engagement in activities be relatively more important. Deichmann et al. (2008) find that in outside of agriculture (non-agricultural wage specializers and Bangladesh, returns to self and wage employment outside of agricul- non-agricultural self-employment specializers), as well as on ture tend to decline with distance to the main urban centers, and to diversified households, as these constitute a significant share of decline faster as the agricultural potential increases. the total (Table 3). It should be noted that since the sum of the The low-potential low-integration areas are more difficult to probabilities of households falling into any of the six diversifica sign a priori, as on the one hand households will have to rely to a tion/specialization categories is equal to one, one should interpret large extent on subsistence farming for their own survival, while the trends in the three reported categories as the mirror image of on the other hand they will also try to complement the expected the probability of being in one of the other categories, with farming meager returns from farming with (possibly equally meager) attracting the lion’s share of specializing households (again, refer returns from non-farm activities, including migration. The distinc- to Table 3 for the distribution of household into these categories). tion between diversification from necessity as opposed to from The graphs convey the combined effect of the quadratic and choice proposed by Ellis (2000) is useful in characterizing the situ- interaction terms that are otherwise difficult to interpret from a ation in these areas. standard table of coefficients. The first result that emerges is that Our use of a quadratic distance term and of interactions non-linearities are clearly present in most of the estimated rela- between distance and aridity reflect these expected non- tionships. For most countries and sectors of specialization, the role linearities. For the reasons detailed above, the magnitude and signs of distance changes markedly with potential and with city size, but of these relationships may vary with the size of the urban centers it is difficult to gauge far-reaching regularities. There does not one considers when measuring urban integration. seem to be any universal law governing how the probability of households moving into the non-farm sector varies with distance from urban centers and with agricultural potential. Even within 4.2. Results: The impact of distance from urban centers and the same country, how the likelihood of households selecting into agricultural potential on household income generation strategies different categories changes with distance is hardly ever constant across city size or across level of agricultural potential. As summarized in the above discussion, we effectively estimate To facilitate the interpretation of these graphs, we turn to the 5 logit models using 4 different city size categories. We focus the relationship between income strategies and distance from cities. discussion on the extent to which we found presence of non- As expected, most lines are downward sloping, indicating that linearities, their extent and direction, and on the regularities and the probability of household diversifying or specializing in key differences we find across countries, between the role of urban non-farm activities declines as the distance from cities increases. centers of different sizes, and by agricultural potential. To convey There are, however, several exceptions. In Malawi’s low potential areas for instance, the probability of a household being in the 18 Summary statistics for the variables are included in the Appendix Table A4. diversified category declines from around 50 percent to below 40 19 In what follows, we loosely use the term integration as the inverse of distance. percent as distances from towns of 20 thousand plus inhabitants Diversification Diversification Diversification Distance from the nearest city (20K) Distance from the nearest city (100K) Distance from the nearest city (500K) Ethiopia 2012 Malawi 2011 Niger 2011 Ethiopia 2012 Malawi 2011 Niger 2011 Ethiopia 2012 Malawi 2011 Niger 2011 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 Distance Distance Distance -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 Distance Distance Distance Distance Distance Distance low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot Nigeria 2010 Tanzania 2009 Uganda 2011 Nigeria 2010 Tanzania 2009 Uganda 2011 Nigeria 2010 Tanzania 2009 Uganda 2011 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 0 .1 .2 .3 .4 .5 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 Distance Distance Distance Distance Distance Distance Distance Distance Distance low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot Specialization in Non-Agricultural Wage Specialization in Non-Agricultural Wage Specialization in Non-Agricultural Wage Distance from the nearest city (20K) Distance from the nearest city (100K) Distance from the nearest city (500K) Ethiopia 2012 Malawi 2011 Niger 2011 Ethiopia 2012 Malawi 2011 Niger 2011 B. Davis et al. / Food Policy 67 (2017) 153–174 Ethiopia 2012 Malawi 2011 Niger 2011 0 .02 .04 .06 .08 .1 0 .02 .04 .06 .08 .1 0 .02 .04 .06 .08 .1 0 .02 .04 .06 .08 .1 0 .02 .04 .06 .08 .1 0 .02 .04 .06 .08 .1 0 .02 .04 .06 .08 .1 0 .02 .04 .06 .08 .1 0 .02.04.06.08 .1 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 Distance Distance Distance -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 Distance Distance Distance Distance Distance Distance low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot Nigeria 2010 Tanzania 2009 Uganda 2011 Nigeria 2010 Tanzania 2009 Uganda 2011 Nigeria 2010 Tanzania 2009 Uganda 2011 0 .02 .04 .06 .08 .1 0 .02 .04 .06 .08 .1 0 .02 .04 .06 .08 .1 0 .02 .04 .06 .08 .1 .08.1 .06 .1 .1 .08 .04 .08 .1 .08 .1 .08 .06 .06 .06 .06 0.02 .04 .04 .04 .04 .02 .02 .02 .02 0 0 0 0 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 Distance Distance Distance Distance Distance Distance Distance Distance Distance low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot Specialization in Self-Employment Specialization in Self-Employment Distance from the nearest city (20K) Specialization in Self-Employment Distance from the nearest city (100K) Distance from the nearest city (500K) Ethiopia 2012 Malawi 2011 Niger 2011 Ethiopia 2012 Malawi 2011 Niger 2011 0 .05 .1 .15 .2 .25 .3 0 .05 .1 .15 .2 .25 .3 0 .05 .1 .15 .2 .25 .3 Ethiopia 2012 Malawi 2011 Niger 2011 0 .05 .1 .15 .2 .25 .3 0 .05 .1 .15 .2 .25 .3 0 .05 .1 .15 .2 .25 .3 0 .05 .1 .15 .2 .25 .3 0 .05 .1 .15 .2 .25 .3 0 .05 .1 .15 .2 .25 .3 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 Distance Distance Distance Distance Distance Distance -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 Distance Distance Distance low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot Nigeria 2010 Tanzania 2009 Uganda 2011 Nigeria 2010 Tanzania 2009 Uganda 2011 Nigeria 2010 Tanzania 2009 Uganda 2011 0 .05 .1 .15 .2 .25 .3 0 .05 .1 .15 .2 .25 .3 0 .05 .1 .15 .2 .25 .3 0 .05 .1 .15 .2 .25 .3 0 .05 .1.15 .2.25.3 0 .05.1.15.2.25.3 0.05.1.15.2.25.3 .25.3 .25.3 .15.2 .15.2 0.05.1 0.05.1 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 -6 -4 -2 0 2 4 Distance Distance Distance Distance Distance Distance Distance Distance Distance low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot low pot high pot Fig. 20. Multinomial logit results: The effect of distance on income strategies, by agricultural potential (aridity) – Base category: Farm specialization. 167 168 B. Davis et al. / Food Policy 67 (2017) 153–174 increases. In Niger, a broadly similar trend is observed. Ethiopia is true in Niger and no difference is observed in Ethiopia and Tan- and Nigeria also have downward sloping curves, but here the lines zania (where the levels of self-employment specialization are for high and low potential areas are virtually overlapping. In Tan- smallest). Where non-agricultural wage and self-employment spe- zania and Uganda, on the other hand the curves are of an cialization probabilities increase with distance, this is usually inverted-U shape: they overlap in Uganda, while in Tanzania the ‘compensating’ for a decline in diversification from relatively high probability of being diversified is higher for households in high levels (Malawi, Niger). potential areas at any given distance. We have noted above how the differences by potential (the gap In several cases, the slope of the curves also increases when dis- between the two lines in each graph) is sometimes very sizeable, tance to larger cities is considered, but again, the trend is by no sometimes non-existent. High/low potential areas are associated means universal. Specialization in non-agricultural wage in Malawi with different probabilities depending on country, city size and for instance is rather flat as distance to small towns increases, but category, with few regularities to speak of. The only country where clearly downward sloping for cities of half a million people or more. we never observe a substantial difference between the two ‘strata’ This is consistent with the expectation that larger centers play more is Ethiopia (note that this is also the country where specialization of a stimulus factor for non-agricultural occupations, but at the in farming is dominant in the data) whereas in all other countries same time we observe cases where the slope is not much affected, the difference matters in at least some of the category/city size or is affected in an opposite direction to what was expected, when combinations. Also, there is a substantial amount of switching of the size of the cities being considered increases (e.g. Ethiopia, and the dominant ‘stratum’ across specialization/diversification cate- Tanzania for self-employment and diversification). gories, less so across city sizes. One aspect to note is that the difference in predicted probabil- These findings speak to different dynamics when the role of ities, whether across high and low potential areas, or over the dis- small towns is considered and when large cities come into play. tance continuum, is often of sizeable magnitude, meaning that For small towns, we find support to the hypothesis that high- understanding these relationships does matter for understanding potential, low-integration areas see less specialization in off-farm how these factors play out and interact in shaping household activities, the reverse being true for high-integration low- strategies. In Niger, Nigeria and Uganda for instance, the probabil- potential areas. These were the two cells in Fig. 19 for which we ities of specializing in self-employment activities decline by 20–30 had clear a priori expectations, but we also found that the role of points as distance from cities of half a million people or more potential is not particularly strong, at least when the off-farm spe- increases. In Niger, the probability of households diversifying is cialization categories are considered. The two cells where we had about twice as large in low potential areas as compared to high unclear expectations were the high potential-high integration, potential areas, and differences of similar scale can be observed and low-potential low-integration areas. For the former, we find for non-agricultural wage specialization in Nigeria. that at least in Tanzania and Uganda the combination of favorable A few considerations can be made when looking at the income conditions for agriculture and lower distance from urban centers generation strategies individually. Diversification, as defined tends to create the conditions for more households to specialize above, is generally more likely close to urban centers, with Tanza- in off-farm activities. When integration is lower and agricultural nia and to some extent Ethiopia being the exceptions. In Tanzania conditions more difficult, the picture is mixed, with households note however the corresponding steep decrease in non-agricultural more likely to engage more fully in non-farm activities in Niger, wage specialization as distance from cities increase, as the two but less likely to do so in Uganda and Tanzania. trends are probably two sides of the same story (agricultural wage When distance to large cities is considered, the impact of dis- specialization being replaced by more diversification, a mixed bag tance is generally more marked, as signaled by the relatively stee- of income sources, as distance from cities increases). Where differ- per negative slope for both self-employment and non-agricultural ences in probabilities across high and low potential areas are size- wage work. In low-potential, low-integration areas, the sign was able (i.e. where the two lines in each graph lie apart), uncertain a priori and we find that the impact of distance prevails. diversification is usually higher in low potential areas (Tanzania In high potential areas, we still find the effect of distance generally being the exception). In Malawi and Niger, the difference in prob- more than offsetting the effect of potential, which results in abilities between low and high potential areas decreases with dis- decreased odds of being specialized off-farm relative to agriculture tance, but does not disappear completely. In Tanzania, where as distance from major cities increases. In both cases, Tanzania and households in high potential areas are more likely to be diversified, Ethiopia counter the trends in at least some of the income the gap with high potential areas increases with distance. categories. For non-agricultural wage specialization, the probabilities tend All in all, these results point to evidence that appears to be to decline with distance in cities of half a million plus, the excep- broadly consistent with the predictions of the theory. There is no tion being Ethiopia. For smaller cities the story is mixed, with sign of African households adopting income generation strategies mostly flat curves when distance from the smaller towns (20 thou- that differ from those observed elsewhere in terms of their rela- sand) is considered. There is also a mix of country situations with tionship to basic exogenous determinants such as agricultural probabilities of specializing in non-agricultural wage higher in potential and distance from urban centers. There is however evi- high potential areas in Malawi, Niger, and Tanzania, but lower in dence that theory alone cannot be relied upon to predict the net Nigeria and Uganda. effects of these forces, and that careful, location-specific and spa- For self-employment, the relationship with distance is still pre- tially explicit diagnostic work is needed to inform policies to facil- sent but less generalized. It is consistently downward sloping only itate the transformation of rural livelihoods. in Niger and Uganda, where the levels of specialization in self- employment activities are relatively high near all urban centers. In the other countries it is either flat (Malawi), moderately upward 5. Conclusion sloping (Ethiopia), or changes from upward to downward sloping as city size increases (Niger). Specialization in self-employment Is Africa’s rural economy transforming as its economies grow? also tends to be more likely in high potential areas in three of Is it trapped in a growth pattern based on natural resources that the six countries (Malawi, Nigeria, Uganda), whereas the opposite may prove unsustainable in the long run? Is there evidence B. Davis et al. / Food Policy 67 (2017) 153–174 169 of the share of agriculture in the economy decreasing, following have a larger share of specialization into non-agricultural wage the familiar secular pattern followed by the vast majority of the employment. countries now enjoying middle and high-income status? The Conversely, agricultural sources of income are generally most analysis in this paper has explored the latest microdata evidence important for the poorest households. Income from crop and to respond to some of these questions from the perspective of livestock activities, as well as from agricultural wage labor, rep- the rural economy. resents a higher share of total income for poorer households in The analysis of the income-generating activities of rural house- almost all countries. Furthermore, a higher share of households holds based on a large cross-country dataset paints a clear picture specializing in on-farm activities, and particularly agricultural of multiple activities across rural space and diversification across wage employment, is found at the low end of the welfare rural households. This diversification is true across countries at distribution. all levels of development and in all four continents, although less For both African and non-African countries, diversification may so in the African countries included in the sample. Bearing in mind function as a household strategy to manage risk and overcome the caveat that our sample is not representative of the whole of market failures, or represent specialization within the household Sub-Saharan Africa, the evidence seems to point towards African deriving from individual attributes and comparative advantage. patterns of household level income diversification as having the Therefore, diversification can be into either high or low-return sec- potential to converge towards patterns similar to those observed tors, reflect push or pull forces, and represent a pathway out of in other developing regions. While African households are still gen- poverty or a survival strategy. erally more likely to specialize in farming compared to households The results offered here suggest the need to carefully consider in other regions, after controlling for the level of GDP, the shares of how to promote rural development, particularly in Sub-Saharan income and participation in non-agricultural activities are not dis- Africa. Even if development, in the long run, does entail exit from similar from those found elsewhere. agriculture, the age-old (Johnston and Mellor, 1961) conclusion For most countries outside Africa (generally with higher levels that this transition needs to happen through investment in the sec- of GDP), the largest share of income stems from off-farm activities, tor, and not its neglect, is still valid today. It is unlikely that inclu- and the largest share of households have diversified sources of sive growth and poverty reduction can happen in rural Africa, income. However, for the African countries in the sample, most where half the households specialize in agriculture, without pro- income still derives from on-farm sources. In terms of participation ductivity growth in the sector. rates, a striking 92 percent of rural households are involved in The spatial analysis of the factors that drive specialization away farming to some extent. Even more remarkably, agricultural from on-farm activities demonstrates that the constraints to off- income represents 69 percent of total income for the average rural farm specialization are likely to differ between high- and low- household in Africa, meaning it is by far the most important source potential and high- and low-integration areas. Additionally, small of household income. As a result, the median African rural house- and large urban centers are likely to exert different influences on hold earns three fourths of its income from agriculture. the transformation of the rural economy. While this adds complex- Specialization in on-farm income-generating strategies is thus ity to the formulation of policies to promote rural non-farm the norm among the African countries in the sample. growth, it also testifies to a series of trends that are not uncommon Agricultural-based sources of income remain critically important in other countries, and suggests that after all the African specificity for rural livelihoods in all countries, in terms of both the overall in terms of higher incidence of farming activities may be due more share of agriculture in rural incomes and the large share of house- to a GDP-level effect than to a different response by households to holds that still specialize in agricultural and on-farm sources of the incentives and opportunities coming from agricultural and income. non-agricultural growth opportunities. While the outcome of a given income-generation strategy will vary by a given household, overall greater reliance on non-farm Appendix sources of income is associated with households being richer, in all countries. In almost all cases, better-off households in rural (See Tables A1–A4) areas have a higher level of participation in (and greater share of income from) non-farm activities. Similarly, richer households 170 B. Davis et al. / Food Policy 67 (2017) 153–174 Table A1 Countries included in the analysis. Country Name of survey Year collected Number of observation Per capita GDP, PPP constant 2005, USD Total Rural Urban African countries Ethiopia Rural Socioeconomic Survey 2011/12 3,969 3,969 N/A 454 Ghana Living Standard Survey 1992 4,552 2,913 1,639 949 Ghana Living Standard Survey 1998 5,998 3,799 2,199 1,051 Ghana Living Standard Survey 2005 8,687 5,069 3,618 1,222 Kenya Integrated Household Budget Survey 2005 13,212 8,487 4,725 1,340 Madagascar Enquete Permanente Aupres des Menages 1993/94 4,504 2,652 1,852 895 Malawi Integrated Household Survey 2004/05 11,280 9,840 1,440 640 Malawi Integrated Household Survey 2010/11 12,271 10,038 2,233 785 Nigeria Living Standard Survey 2004 17,425 13,634 3,791 1,707 Nigeria Living Standard Survey 2010 4,682 3,182 1,500 2,120 Niger Enquête Nationale sur les Conditions 2011 3,968 2,430 1,538 535 de Vie des Ménages et l’Agriculture Uganda National Household Survey 2005/06 7,424 5,714 1,710 966 Uganda National Household Survey 2009/06 2,975 2,206 769 1,130 Tanzania National Panel Survey 2009 3,265 2,063 1,202 1,240 Non African countries Albania Living Standards Measurement Study 2002 3,599 1,640 1,959 4,710 Albania Living Standards Measurement Study 2005 3,640 1,640 2,000 5,463 Bangladesh Household Income-Expenditure Survey 2000 7,440 5,040 2,400 901 Bangladesh Household Income-Expenditure Survey 2005 10,080 6,400 3,680 1,068 Bolivia Encuesta de Hogares 2005 4,086 1,751 2,335 3,758 Bulgaria Integrated Household Survey 1995 2,468 824 1,664 6,930 Bulgaria Integrated Household Survey 2001 2,633 877 1,756 7,348 Ecuador Estudio de Condiciones de Vida 1995 5,810 2,532 3,278 5,658 Ecuador Estudio de Condiciones de Vida 1998 5,801 2,535 3,266 5,862 Guatemala Encuesta de Condiciones de Vida 2000 7,276 3,852 3,424 3,966 Guatemala Encuesta de Condiciones de Vida 2006 13,693 7,878 5,808 4,178 Indonesia Family Life Survey-Wave 1 1993 7,216 3,786 3,430 2,487 Indonesia Family Life Survey-Wave 3 2000 10,435 5,410 5,025 2,724 Nepal Living Standards Survey I 1996 3,370 2,655 715 829 Nepal Living Standards Survey III 2003 5,071 3,655 1,416 926 Nicaragua Encuesta de Medición de Niveles de Vida 1998 4,236 1,963 2,273 1,961 Nicaragua Encuesta de Medición de Niveles de Vida 2001 4,191 1,839 2,352 2,145 Nicaragua Encuesta de Medición de Niveles de Vida 2005 6,864 3,400 3,464 2,311 Pakistan Integrated Household Survey 1991 4,792 2,396 2,396 1,719 Pakistan Integrated Household Survey 2001 15,927 9,978 5,949 1,923 Panama Encuesta de Niveles de Vida 1997 4,945 2,496 2,449 7,554 Panama Encuesta de Niveles de Vida 2003 6,363 2,945 3,418 8,267 Tajikistan Living Standards Survey 2003 4,156 2,640 1,520 1,283 Tajikistan Living Standards Survey 2007 4,860 3,150 1,710 1,656 Vietnam Living Standards Survey 1992 4,800 3,840 960 997 Vietnam Living Standards Survey 1997/98 6,002 4,272 1,730 1,448 Vietnam Living Standards Survey 2002 29,380 22,621 6,909 1,780 Table A2 Participation in income-generating activities by country, rural households. Country Per capita GDP, Income-generating activity and year PPP constant Group I Group II Group III 2005, USD (1) (2) (3) (4) (5) (6) (7) (1) + (2) (4) + (5) (1) (4) (6) + (7) (3) + (4) + (3) + (6) + (7) + (2) + (5) + (5) + (6) + (7) Agriculture- Agriculture - Agricultural Non-farm Non-farm Transfers Other Agricultural Non- On- Non- Transfers Off-farm crops Livestock wage wage self- total agricultural farm farm & other total employment employment employment total total total African countries Ethiopia 2012 454 87% 80% 24% 6% 19% 22% 19% 89% 47% 92% 24% 38% 60% Ghana 1992 949 87% 54% 4% 14% 45% 37% 6% 88% 73% 88% 54% 40% 75% Ghana 1998 1,051 88% 51% 4% 18% 40% 41% 13% 89% 75% 89% 49% 49% 76% Ghana 2005 1,222 85% 43% 4% 13% 41% 36% 4% 88% 69% 87% 49% 38% 70% Kenya 2005 1,340 89% 79% 13% 25% 21% 53% 13% 94% 74% 92% 41% 57% 79% Madagascar 1993 895 93% 78% 26% 18% 21% 43% 11% 96% 67% 95% 36% 50% 75% Malawi 2004 640 96% 65% 55% 16% 30% 89% 7% 98% 93% 97% 42% 90% 97% Malawi 2011 785 93% 48% 49% 13% 16% 66% 11% 97% 79% 93% 28% 71% 91% Niger 2011 535 96% 77% 11% 8% 60% 58% 0% 98% 84% 98% 65% 58% 86% Nigeria 2004 1,707 85% 38% 1% 9% 16% 6% 4% 86% 30% 86% 24% 9% 31% B. Davis et al. / Food Policy 67 (2017) 153–174 Nigeria 2010 2,120 81% 53% 3% 14% 47% 3% 5% 84% 57% 86% 53% 8% 56% Tanzania 2009 1,240 97% 61% 22% 15% 34% 57% 1% 99% 77% 98% 43% 58% 82% Uganda 2005 966 88% 65% 20% 16% 38% 43% 2% 92% 72% 90% 49% 44% 79% Uganda 2009 1,130 89% 67% 23% 25% 43% 32% 24% 92% 77% 91% 56% 49% 83% Simple mean 89% 61% 18% 15% 34% 42% 9% 92% 70% 92% 44% 47% 74% Minimum 81% 38% 1% 6% 16% 3% 0% 84% 30% 86% 24% 8% 31% Maximum 97% 80% 55% 25% 60% 89% 24% 99% 93% 98% 65% 90% 97% Non African countries Albania 2002 4,710 92% 86% 5% 28% 10% 68% 4% 93% 85% 93% 35% 69% 87% Albania 2005 5,463 95% 85% 5% 30% 11% 74% 19% 95% 90% 95% 39% 76% 92% Bangladesh 2000 901 82% 39% 35% 32% 26% 49% 55% 87% 91% 79% 53% 75% 97% Bangladesh 2005 1,068 85% 73% 29% 35% 22% 42% 59% 93% 90% 82% 53% 76% 96% Bolivia 2005 3,758 79% 48% 7% 18% 83% 27% 4% 84% 96% 81% 92% 29% 98% Bulgaria 1995 6,930 65% 41% 22% 37% 4% 66% 14% 73% 86% 66% 39% 69% 92% Bulgaria 2001 7,348 68% 64% 8% 26% 2% 89% 13% 78% 95% 76% 29% 91% 97% Ecuador 1995 5,658 74% 76% 39% 34% 39% 27% 48% 93% 85% 88% 57% 62% 94% Ecuador 1998 5,862 68% 78% 35% 34% 38% 28% 15% 89% 71% 85% 56% 38% 86% Guatemala 2000 3,966 88% 66% 43% 35% 31% 65% 4% 93% 84% 91% 53% 67% 95% Guatemala 2006 4,178 81% 46% 31% 51% 33% 71% 3% 85% 90% 81% 67% 72% 97% Indonesia 1993 2,487 57% 29% 20% 26% 30% 71% 11% 72% 85% 61% 50% 74% 89% Indonesia 2000 2,724 54% 10% 19% 32% 33% 85% 14% 64% 93% 54% 55% 87% 94% Nepal 1996 829 93% 82% 42% 35% 20% 26% 8% 98% 69% 95% 50% 32% 85% Nepal 2003 926 93% 86% 38% 36% 21% 38% 27% 98% 82% 96% 52% 53% 91% Nicaragua 1998 1,961 71% 68% 42% 38% 22% 33% 4% 90% 67% 83% 50% 36% 85% Nicaragua 2001 2,145 85% 72% 39% 35% 26% 39% 19% 95% 73% 92% 52% 43% 87% Nicaragua 2005 2,311 82% 67% 43% 30% 38% 33% 6% 94% 70% 90% 56% 36% 84% Pakistan 1991 1,719 60% 76% 25% 47% 32% 31% 3% 84% 80% 80% 68% 33% 86% Pakistan 2001 1,923 40% 65% 20% 48% 18% 31% 16% 75% 78% 70% 58% 41% 85% Panama 1997 7,554 87% 98% 27% 44% 53% 69% 8% 99% 94% 99% 79% 71% 98% Panama 2003 8,267 78% 65% 30% 42% 56% 64% 12% 87% 87% 82% 58% 67% 94% Tajikistan 2003 1,283 89% 69% 49% 29% 3% 58% 1% 95% 73% 93% 32% 58% 91% Tajikistan 2007 1,656 98% 78% 28% 45% 17% 48% 3% 99% 78% 99% 56% 49% 88% Vietnam 1992 997 95% 88% 15% 22% 41% 35% 5% 97% 72% 94% 54% 38% 77% Vietnam 1998 1,448 98% 91% 20% 32% 38% 36% 19% 99% 80% 98% 59% 48% 86% Vietnam 2002 1,780 79% 68% 11% 39% 40% 83% 25% 85% 96% 83% 64% 87% 96% Simple mean 79% 67% 27% 35% 29% 51% 16% 89% 83% 85% 54% 58% 91% 171 Minimum 40% 10% 5% 18% 2% 26% 1% 64% 67% 54% 29% 29% 77% Maximum 98% 98% 49% 51% 83% 89% 59% 99% 96% 99% 92% 91% 98% Table A3 172 Share of income-generating activities in total rural household income, by country. Country and year Per capita Income-generating activity GDP, PPP Group I Group II Group III constant 2005, USD (1) (2) (3) (4) (5) (6) (7) (1) + (2) + (3) (4) + (5) (1) + (2) (4) + (5) (6) + (7) (3) + (4) + (5) + (6) + (7) + (6) + (7) Agriculture- Agriculture - Agricultural Non-farm Non-farm Transfers Other Agricultural Non- On-farm Non-farm Transfers Off-farm Crops Livestock wage wage self- total agricultural total total & other total employment employment employment total African countries Ethiopia 2012 454 73% 11% 4% 2% 4% 3% 3% 88% 12% 85% 6% 6% 15% Ghana 1992 949 66% 3% 2% 8% 16% 6% 0% 71% 29% 69% 23% 6% 31% Ghana 1998 1,051 55% 4% 1% 10% 21% 9% 1% 61% 39% 59% 30% 9% 41% Ghana 2005 1,222 49% 3% 3% 9% 26% 10% 0% 55% 45% 52% 35% 10% 48% Kenya 2005 1,340 32% 16% 7% 15% 9% 19% 2% 55% 45% 48% 24% 21% 52% Madagascar 1993 895 57% 13% 6% 6% 8% 6% 2% 77% 23% 71% 15% 8% 29% Malawi 2004 640 56% 9% 11% 7% 9% 6% 0% 77% 23% 66% 16% 7% 34% Malawi 2011 785 59% 6% 15% 8% 6% 6% 0% 80% 20% 65% 13% 6% 35% Niger 2011 535 48% 9% 3% 4% 26% 10% 0% 60% 40% 57% 30% 10% 43% Nigeria 2004 1,707 76% 5% 1% 7% 10% 1% 1% 81% 19% 81% 17% 2% 19% Nigeria 2010 2,120 48% 9% 1% 11% 29% 0% 1% 58% 42% 57% 40% 2% 43% B. Davis et al. / Food Policy 67 (2017) 153–174 Tanzania 2009 1,240 53% 13% 4% 7% 13% 10% 0% 70% 30% 66% 19% 11% 34% Uganda 2005 966 47% 7% 11% 10% 16% 9% 0% 65% 35% 54% 26% 9% 46% Uganda 2009 1,130 48% 11% 8% 12% 16% 6% 0% 66% 34% 58% 28% 6% 42% Simple mean 55% 9% 5% 8% 15% 7% 1% 69% 31% 63% 23% 8% 37% Minimum 32% 3% 1% 2% 4% 0% 0% 55% 12% 48% 6% 2% 15% Maximum 76% 16% 15% 15% 29% 19% 3% 88% 45% 85% 40% 21% 52% Non African countries Albania 2002 4,710 15% 34% 2% 15% 5% 28% 0% 51% 49% 49% 21% 28% 51% Albania 2005 5,463 17% 23% 3% 18% 7% 28% 3% 43% 57% 41% 26% 31% 59% Bangladesh 2000 901 15% 1% 20% 20% 16% 13% 13% 37% 63% 17% 36% 27% 83% Bangladesh 2005 1,068 18% 9% 16% 22% 13% 9% 12% 43% 57% 27% 36% 21% 73% Bolivia 2005 3,758 29% 7% 5% 13% 36% 9% 1% 41% 59% 36% 49% 10% 64% Bulgaria 1995 6,930 13% 8% 13% 24% 2% 37% 2% 35% 65% 21% 27% 39% 79% Bulgaria 2001 7,348 4% 12% 5% 17% 1% 60% 1% 20% 80% 16% 18% 62% 84% Ecuador 1995 5,658 9% 3% 10% 39% 23% 9% 6% 23% 77% 12% 62% 15% 88% Ecuador 1998 5,862 22% 11% 20% 18% 18% 5% 5% 54% 46% 33% 37% 10% 67% Guatemala 2000 3,966 28% 3% 20% 20% 12% 17% 0% 50% 50% 30% 33% 17% 70% Guatemala 2006 4,178 21% 3% 17% 27% 13% 18% 0% 41% 59% 24% 40% 19% 76% Indonesia 1993 2,487 25% 5% 11% 16% 15% 26% 2% 41% 59% 30% 31% 28% 70% Indonesia 2000 2,724 23% 2% 10% 20% 18% 23% 4% 35% 65% 26% 38% 27% 74% Nepal 1996 829 32% 14% 18% 17% 9% 10% 1% 64% 36% 46% 26% 11% 54% Nepal 2003 926 20% 18% 13% 21% 9% 17% 2% 51% 49% 38% 30% 19% 62% Nicaragua 1998 1,961 23% 11% 25% 23% 9% 8% 1% 59% 41% 34% 32% 9% 66% Nicaragua 2001 2,145 21% 14% 21% 21% 11% 6% 5% 57% 43% 35% 32% 11% 65% Nicaragua 2005 2,311 30% 0% 24% 17% 15% 12% 1% 55% 45% 30% 32% 13% 70% Pakistan 1991 1,719 31% 14% 6% 27% 19% 3% 1% 51% 49% 46% 45% 3% 54% Pakistan 2001 1,923 21% 11% 9% 29% 11% 15% 5% 41% 59% 33% 39% 19% 67% Panama 1997 7,554 15% 7% 14% 29% 17% 16% 1% 37% 63% 23% 46% 18% 77% Panama 2003 8,267 16% 2% 17% 27% 23% 15% 1% 35% 65% 18% 50% 16% 82% Tajikistan 2003 1,283 37% 17% 17% 12% 1% 15% 0% 72% 28% 55% 13% 16% 45% Tajikistan 2007 1,656 52% 9% 7% 19% 6% 7% 0% 68% 32% 61% 26% 7% 39% Vietnam 1992 997 53% -1% 5% 11% 23% 8% 0% 57% 43% 52% 34% 8% 48% Vietnam 1998 1,448 41% 15% 6% 9% 21% 7% 0% 62% 38% 56% 30% 7% 44% Vietnam 2002 1,780 26% 3% 10% 38% 15% 7% 2% 39% 61% 29% 52% 9% 71% Simple mean 25% 9% 13% 21% 14% 15% 3% 46% 54% 33% 35% 18% 67% Minimum 4% -1% 2% 9% 1% 3% 0% 20% 28% 12% 13% 3% 39% Maximum 53% 23% 25% 39% 36% 60% 13% 72% 80% 61% 62% 62% 88% B. Davis et al. / Food Policy 67 (2017) 153–174 173 Table A4 Table A4 (continued) Summary statistics of the variables used in the multinomial logit model. Variable Obs Mean Std. Dev. Variable Obs Mean Std. Dev. Nigeria Ethiopia Log of Euclidean Distance 20k 3,177 3.06 1.06 Log of euclidean distance 20k 3,969 3.34 0.80 Log of Euclidean Distance 100k 3,177 3.58 0.90 Log of euclidean distance 100k 3,969 4.45 0.81 Log of Euclidean Distance 500k 3,177 4.35 0.79 Log of euclidean distance 500k 3,969 5.43 0.56 Log of Euclidean Distance 1000k 3,177 4.88 0.71 Log of euclidean distance 1000k 3,969 5.51 0.59 Aridity 3,177 0.83 0.46 Aridity 3,969 0.71 0.23 Ag wealth index 3,177 0.00 1.02 Ag wealth index 3,969 0.00 1.00 Wealth index (non Ag) 3,157 0.00 0.99 Wealth index (non Ag) 3,969 0.00 1.00 Infrastructure Index 2,691 0.00 0.96 Infrastructure index 3,969 0.00 1.00 Labor force in the HH 3,177 3.20 2.10 Labor force in the HH 3,969 2.95 1.45 Female share of labor force 3,177 0.50 0.27 Female share of labor force 3,969 0.52 0.31 Female household head 3,177 0.14 0.35 Female household head 3,969 0.22 0.41 Household head age 3,169 50.41 15.77 Household head age 3,969 44.30 15.84 Household head years of education 3,102 4.86 5.18 Household head years of education 3,969 1.79 3.06 Household size 3,177 5.84 3.13 Household size 3,969 5.11 2.26 Land owned, Ha 3,177 0.49 2.05 Land owned, Ha 3,969 4.09 6.19 Specialization/Diversification Specialization/Diversification Diversified 3,112 20.33% 0.40 Diversified 3,812 13.72% 0.30 Specialized in farm activities 3,112 48.80% 0.50 Specialized in Farm activities 3,969 68.48% 0.43 Specialized in Ag wage 3112 0.35% 0.06 Specialized in Ag wage 3,812 2.40% 0.10 Specialized in NON Ag wage 3,112 7.88% 0.27 Specialized in Non Ag wage 3,812 3.80% 0.11 Specialized in self-employment 3,112 21.85% 0.41 Specialized in Self-employment 3,969 7.36% 0.15 Specialized in other income activities 3,112 0.79% 0.08 Specialized in other income activities 3,969 4.24% 0.12 Tanzania Malawi Log of Euclidean Distance 20k 2,008 3.42 0.86 Log of Euclidean Distance 20k 9,816 3.42 0.64 Log of Euclidean Distance 100k 2,008 4.26 0.84 Log of Euclidean Distance 100k 9,816 4.08 0.64 Log of Euclidean Distance 500k 2,008 5.51 0.74 Log of Euclidean Distance 500k 9,816 4.30 0.83 Log of Euclidean Distance 1000k 2,008 5.90 0.59 Log of Euclidean Distance 1000k 9,816 6.34 0.14 Aridity 2,008 0.67 0.23 Aridity 9,800 0.68 0.17 Ag wealth index 2,006 0.00 1.01 Ag wealth index 9,816 0.00 0.70 Wealth index (non Ag) 2,006 0.00 1.00 Wealth index (non Ag) 9,816 0.00 0.99 Infrastructure index 2,008 -0.13 0.86 Infrastructure index 9,816 0.00 1.00 Labor force in the HH 2,008 2.52 1.56 Labor force in the HH 9,816 2.11 1.18 Female share of labor force 2,008 0.51 0.27 Female share of labor force 9,816 0.47 0.31 Female household head 2,008 0.24 0.43 Female household head 9,816 0.25 0.43 Household head age 2,008 47.22 15.92 Household head age 9,816 42.89 16.78 Household head years of education 1,982 4.48 3.40 Household head years of education 9,816 5.35 3.51 Household size 2,008 5.44 2.92 Household size 9,816 4.60 2.18 Land owned, Ha 2,008 1.60 1.79 Land owned, Ha 9,816 0.60 0.50 Specialization/Diversification Specialization/Diversification 9,816 Diversified 2,008 34.85% 0.48 Diversified 9,816 35.92% 0.48 Specialized in farm activities 2,008 52.99% 0.50 Specialized in farm activities 9,816 44.04% 0.50 Specialized in Ag wage 2,008 0.67% 0.08 Specialized in Ag wage 9,816 7.78% 0.28 Specialized in non Ag wage 2,008 2.95% 0.17 Specialized in non Ag wage 9,816 6.69% 0.25 Specialized in self-employment 2,008 4.53% 0.21 Specialized in self-employment 9,816 3.50% 0.18 Specialized in other income activities 2,008 4.02% 0.20 Specialized in other income activities 9,816 2.07% 0.14 Uganda Niger Log of Euclidean Distance 20k 2,629 2.73 1.14 Log of Euclidean Distance 20k 2,409 3.79 0.64 Log of Euclidean Distance 100k 2,629 4.15 1.18 Log of Euclidean Distance 100k 2,409 4.42 0.59 Log of Euclidean Distance 500k 2,629 4.72 1.22 Log of Euclidean Distance 500k 2,409 5.13 0.43 Log of Euclidean Distance 1000k 2,629 4.83 1.26 Log of Euclidean Distance 1000k 2,409 5.35 0.53 Aridity 2,605 0.73 0.11 Aridity 2,409 0.19 0.05 Ag wealth index 2,672 0.00 1.66 Ag wealth index 2,409 0.00 0.98 Wealth index (non Ag) 2,688 -0.15 1.10 Wealth index (non Ag) 2,409 0.00 0.94 Infrastructure index 2,687 0.00 0.90 Infrastructure index 2,409 -0.01 0.98 Labor force in the HH 2,688 2.38 1.49 Labor force in the HH 2,409 0.96 0.18 Female share of labor force 2,688 0.50 0.30 Female share of labor force 2,409 0.94 0.23 Female household head 2,688 0.30 0.46 Female household head 2,409 0.10 0.30 Household head age 2,687 43.64 15.90 Household head age 2,409 44.61 15.14 Household head years of education 2,686 4.47 4.04 Household head years of education 2,409 1.37 3.55 Household size 2,688 5.13 2.88 Household size 2,409 6.42 3.46 Land owned, Ha 2688 1.65 11.11 Land owned, Ha 1,978 4.17 3.88 Specialization/Diversification Specialization/Diversification Diversified 2,607 35.39% 0.48 Diversified 2,409 46.27% 0.50 Specialized in farm activities 2,607 33.65% 0.47 Specialized in farm activities 2,409 37.90% 0.49 Specialized in Ag wage 2,607 2.81% 0.17 Specialized in Ag wage 2,409 0.46% 0.07 Specialized in non Ag wage 2,607 11.61% 0.32 Specialized in non Ag wage 2,409 2.23% 0.15 Specialized in self-employment 2,607 12.44% 0.33 Specialized in self-employment 2,409 10.47% 0.31 Specialized in other income activities 2,591 4.10% 0.19 Specialized in other income activities 2,409 2.66% 0.16 174 B. Davis et al. / Food Policy 67 (2017) 153–174 References FAO, 1998. The State of Food and Agriculture: Rural Non-Farm Income in Developing Countries. Food and Agriculture Organization, Rome. Garcia, M., Moore, C., 2012. The Cash Dividend: The Rise of Cash Transfer Programs Azzarri, C., Carletto, C., Davis, B., Fatchi, T., Vigneri, M., 2006. Food and nutrition in Sub-Saharan Africa. The World Bank, Washington, DC. security in Malawi. In: Background Paper Prepared for the World Bank Malawi Haggblade, S., Hazell, P., Reardon, T. (Eds.), 2007. Transforming the Rural Nonfarm Poverty and Vulnerability Assessment. FAO, Rome. Economy. Johns Hopkins University Press. Barrett, C., Reardon, T., Webb, P., 2001. Nonfarm income diversification and Hamer, A.M., Linn, J.F., 1987. Urbanization in the developing world: patterns, issues, household livelihood strategies in Rural Africa: concepts, dynamics and policy and policies. In: Mills, E.S. (Ed.), Handbook of Regional and Urban Economics, implications. Food Policy 26 (4). vol. II. Elsevier. Bezu, S., Barrett, C., 2012. Employment dynamics in the rural nonfarm sector in Ivanic, M., Martin, W., 2008. Implications of higher global food prices for poverty in Ethiopia: do the poor have time on their side? J. Develop. Stud. 48 (9), 1223– low-income countries. Agricult. Econ. 39 (s1). November. 1240. Johnston, B., Mellor, J., 1961. The role of agriculture in economic development. Am. Bezu, S., Barrett, C., Holden, S., 2012. Does nonfarm economy offer pathways for Econ. Rev. 51 (4), 566–593. upward mobility? World Develop. 40 (8), 1634–1646. Lanjouw, P., Feder, G., 2001. Rural Nonfarm Activities and Rural Development: From Binswanger-Mkhize, H.P., McCalla, A.F., Patel, P., 2010. Structural transformation Experience Towards Strategy. Rural Development Strategy Background Paper and African agriculture. Global J. Emerg. Market Econ. 2, 113–152. #4. The World Bank, Washington, DC. Carletto, G., Covarrubias, K., Davis, B., Krausova, M., Winters, P., 2007. Rural Income Lanjouw, J., Lanjouw, P., 2001. The rural nonfarm sector: issues and evidence from Generating Activities Study: Methodological Note on the Construction of developing countries. Agricult. Econ. 26 (1), 1–23. Income Aggregates. Agricultural Sector in Economic Development Service, Long, J.S., Freese, J., 2014. Regression Models for Categorical Dependent Variables Food and Agriculture Organization . McCaig, B., Pavcnik, N., 2014. Moving Out of Agriculture: Structural Change in CGIAR, 2014. Global Geospatial Potential EvapoTranspiration & Aridity Index. Vietnam, NBER Working Paper Series. Working Paper 19616. National Bureau of Methodology and Dataset Description (accessed McCullough, E.B., 2015. Labor Productivity and Employment Gaps in Sub-Saharan 4 April 2014). Africa, Policy Research Working Paper 7234. The World Bank, Washington, DC. Chawanote, C., Barrett, C., 2013. Farm and Non-Farm Occupational and Earnings Nagler, P., Naudé, W., 2014. Non-Farm Enterprises in Rural Africa, New Empirical Dynamics in Rural Thailand. Cornell University, Mimeo. Evidence, Policy Research Working Paper 7066. The World Bank, Washington, Chenery, H., Syrquin, M., 1975. Patterns of Development. Oxford University Press, DC. London. Ratha, D., Mohapatra, S., Ozden, C., Plaza, S., Shaw, W., Shimeles, A., 2011. Christiaensen, L., De Weerdt, J., Todo, Y., 2013. Urbanization and poverty reduction: Leveraging Migration for Africa: Remittances, Skills, and Investments. The the role of rural diversification and secondary towns. Agricult. Econ. 44, 435– World Bank, Washington, DC. 447. Reardon, T., Taylor, J.E., Stamoulis, K., Lanjouw, A., Balisacan, P., 2000. Effects of non- Davis, B., Winters, P., Carletto, G., Covarrubias, K., Quinones, E., Zezza, A., Stamoulis, farm employment on rural income inequality in developing countries: an K., Bonomi, G., Di Giuseppe, S., 2010. Assets, activities and rural poverty investment perspective. J. Agric. Econ. 51 (2), 266–288. alleviation: evidence from a multicountry analysis. World Develop. 38 (1), 48– UCLA, 2014. UCLA: Statistical Consulting Group, Stata Annotated Output. 63. Multinomial Logistic Regression . Accessed 2 April 2014. structural transformation of Sub-Saharan Africa. World Develop. 63, 33–42. Winters, P., Davis, B., Carletto, G., Covarrubias, K., Quinones, E., Zezza, A., Stamoulis, de Ferranti, D., Perry, G.E., Foster, W., Lederman, D., Valdés, A., 2005. Beyond the K., Bonomi, G., Di Giuseppe, S., 2009. A cross country comparison of rural City. The Rural Contribution to Development. The World Bank, Washington, DC. income generating activities. World Develop. 37 (9), 1435–1452. Deichmann, U., Shilpi, F., Vakis, R., 2008. Spatial Specialization and Farm-Nonfarm Winters, P., Essam, T., Davis, B., Zezza, A., Carletto, G., Stamoulis, K., 2010. Patterns of Linkages. World Bank Policy Research Working Paper No. 4611. rural development: a cross-country comparison using microeconomic data. J. Ellis, F., 2000. The determinants of rural livelihood diversification in developing Agric. Econ. 61 (3), 628–651. countries. J. Agric. Econ. 51 (2). May. Yamano, T., Kijima, Y., 2010. The associations of soil fertility and market access with Fafchamps, M., Shilpi, F., 2003. The spatial division of labour in Nepal. J. Develop. household income: evidence from rural Uganda. Food Policy 35, 51–59. Stud. 39 (6), 23–66. Fafchamps, M., Shilpi, F., 2005. Cities and specialisation: evidence from South Asia. Econ. J. 115 (503), 477–504.