Uganda: A Reform Agenda for More and Better Jobs through Agriculture
                                       Christopher Delgado
                                        October 16, 2018


Summary of Recommendations

  •   Strengthen public institutions and policies for agricultural productivity.
  •   Increase both the efficiency and amount of public spending for agricultural research and
      extension, especially for smallholder farmers.
  •   Coordinate and strengthen the work of line ministries, extension agencies, and agricultural
      research institutes.
  •   Reform the regulatory environment to ensure consistent quality of inputs and outputs without
      undue compliance burdens.
  •   Promote commercialization of agriculture and trade through improving institutions and
      infrastructure.
  •   Improve access to agricultural finance along entire agrobusiness value chains.
  •   Assess policies related to transport and other transfer costs for agricultural commodities
      entering regional trade.
  •   Analyze and carefully plan investments in infrastructure—roads, electricity, communications,
      water—which are especially critical to how agribusinesses will locate.
  •   Invest in information systems and databases.
  •   Improve rural ICT.
  •   Continue to foster land tenure security through better land demarcation and delivery of
      adequate documentation to land owners.
  •   Pursue initiatives to promote agro-processing and encourage agro industry vertical
      coordination.
  •   Support development of rural cooperatives.
  •   Increase resilience in agro ecosystems and of rural livelihoods.
  •   Increase collaboration between the many resiliency-building actors.
                                                      2


Introduction: Why Agricultural Growth is Critical for Economic Transformation with More
and Better Jobs in Uganda

Economic transformation in low-income countries (LICs) starts in agriculture, where most workers earn
their living.1 Growth in agricultural labor productivity releases labor for off-farm work, typically increasing
overall labor productivity. A vibrant agriculture sector stimulates downstream jobs in production of farm
inputs (plant propagation, seeds, fertilizers, tools and equipment), creates waged jobs on commercial
farms, and generates downstream off-farm jobs in transportation, storage, and pack houses. In time,
growing commercial surpluses of farm products lead to development of downstream off-farm jobs in agro-
processing plants, including for export. In addition, income flowing to remote areas from agricultural sales
outside the local region increases demand for local services and non-tradable goods, such as locally
processed and perishable prepared foods. This effect can be substantial in remote areas where labor is
underemployed, whether due to seasonality or due to high costs of transferring local products out.2

The Ugandan economy is very dependent on agriculture. The agricultural sector accounts for 70 percent
of employment—overwhelmingly on small farms and often at low productivity—occupies half of all land
area, and provides half of all exports and one-quarter of Uganda’s GDP. Agriculture is considered a leading
sector for future economic growth and economic inclusion in Uganda’s National Development Plan.3
Agricultural income growth in Uganda has fueled consumption, stimulated demand for off-farm services,
and has led poverty reduction and inclusive growth compared to other sectors by a wide margin.4 This is
consistent with studies of a number of other agriculturally-dependent LICs, where agriculture has shown
to be two to three times more effective at reducing poverty than growth originating in industry or
services.5 The potential for agricultural growth to reduce poverty compared to other sectors is greatest in
the poorest countries, where agricultural productivity tends to be much lower than in industry and
services, and labor resources are underemployed for much of the year.6 It Is not surprising therefore that
development of higher value agricultural value chains through private sector development continues to
be a fundamental tenet of Ugandan national strategy, and World Bank advice, for both economic growth
and creation of more and better jobs.7

Drivers of agricultural growth and poverty reduction for more than a decade in Uganda, largely
unrelated to specific Ugandan agricultural policy, are unlikely to repeat. Examples of such events include
the relative regional food price spike after 2008; cropland expansion into dwindling forests and wetlands;
a peace dividend after the Lord’s Resistance Army was pushed back from the north; emergence of newly
independent trading partner, South Sudan, eager to import Ugandan processed foods; the opening up of
the East African Community (EAC) to free regional agricultural trade; and a change to local procurement
and processing of food aid in Uganda by the World Food Program (WFP) for its operations in South Sudan.

Prospects for being able to continue to promote overall growth and rapid job creation through
agricultural growth in Uganda are uncertain. Average Total Factor Productivity (TFP)8 growth in Ugandan
agriculture has been negative for the last two decades.9 This suggests that on balance the country is now
getting less for equal or greater effort. National agricultural output has grown at only two percent per
year over the last five years, compared to agricultural output growth of three to five percent in other EAC
                                                    3


members, while the Ugandan population has grown 3.3 percent per annum over the same period.
Furthermore, food insecurity, poverty, and nutritional quality remain major challenges in rural Uganda,
and national food imports have increased in the last decade.

Uganda has a number of concerning issues related to economic growth and agricultural employment.
These concerns include continued high rural population growth and youth unemployment rates, despite
growing outmigration to towns, and the potential for rapid urban population growth of 5.8 percent per
annum in 2017 (highest in EAC) to exceed absorption capacity.10 This has increased rural population
density on arable land and continued to degrade land and water resources in the absence of adequate
on-farm investments in soil and water management. Exposure to climate-related hazards is especially high
in Uganda, and the danger to rural resilience is exacerbated by low adaption capacity in rural communities.
This stems from having few assets, few alternatives due to low education levels, and little access to
financial or other safety nets.

It is questionable whether the Uganda agricultural sector can continue to drive poverty reduction and
economic growth and provide enough more and better jobs.11 Accommodating projected Ugandan youth
who will be completing school from 2020 to 2030 will require well over 3 percent per annual growth in
jobs, and progress will require that a substantial share of these be higher productivity non-farm jobs in
rural areas.12 Uganda has relatively few alternatives to agriculture to create a large number of jobs and
widespread growth in domestic consumer income capable of stimulating demand for local services,
products, and foreign exchange.

We can draw on evidence and lessons from what has worked in other countries to support Ugandan
agriculture sector growth and jobs. Fundamental policy choices are at hand and the stakes are high. If
the measures set out in this paper are taken, Uganda could achieve agricultural GDP growth at the high
end of the range achieved by other EAC members in recent years of about per year, compared to the
present two percent. Total factor productivity in Ugandan agriculture has been declining more than one
percent annually, but could increase to the more than two percent per year experienced by neighboring
Rwanda. Labor productivity growth in agriculture in Uganda could rise from an annual average of about
0.2 percent per year over the last decade to perhaps even more than the 3.5 percent Tanzania has
experience given the 6 percent annual growth achieved by Kenya and Rwanda.

Progress in rebooting agricultural productivity growth and continued agricultural trade growth is tied
to growth in formal jobs. About 125,000 Ugandans are employed in formal sector jobs in food
manufacturing, and in 131,000 formal jobs in food and beverage services, as measured in the most recent
Census of Business Enterprises in 2010. Extrapolating to 2018), about 362,000 Ugandans have formal jobs
in these two areas, and about 560,000 will be employed in formal sector jobs extrapolating estimates to
2028. Formal sector manufacturing jobs in agricultural, food, and beverage processing in 2010 accounted
for two-thirds of all manufacturing employment in Uganda that year. Since food manufacturing has been
growing faster than overall manufacturing, this share is likely to increase. Agro-processing and food
services could exceed the estimated 560,000 jobs by 2028 if processed agricultural exports grow faster
than the more than 10 percent annual experienced from 2006 to 2016. In addition, growth in persons
                                                    4


employed in food services can exceed the conservative 2028 estimate if Ugandan urbanization rates rise
closer to rates seen in other EAC countries.

The agro, food processing, and food services together can generate about new 200,000 formal jobs in
Uganda over the next decade, which represent better, more regularly paid work than in smallholder
farming. If overall economic growth and agricultural trade expand more quickly than in the last decade
under a new agricultural productivity and marketing thrust, 200,000 represents a low estimate. If,
however, agriculture is treated like a low-growth holding sector for the majority of the population,
deserving of equity transfers but without major growth prospects, this promise is unlikely to be attained,
with serious consequences for the overall economy.


Labor Productivity, Employment, and Structural Change for Agriculture in Uganda


        Insights from trends across all sectors and select countries

An in-depth review (Merotto, Weber, and Aterido, 2018) examined structural changes in employment—
sustained increases in labor productivity from work transitions—during 16,000 economic growth
episodes in IDA-eligible countries The study synthesized patterns from separate individual “Jobs
Diagnostics� for each country that take into account countries’ structural characteristics in terms of
demography, growth patterns, and economic transformation. Household and firm-level data are then
analyzed to understand how workers transition into higher labor productivity jobs (value added per
worker), and how firms hire as they grow. Considered together, they yield insights on employment growth
patterns in different kinds of countries by level of formality, gender, age group and sub-sector. The data
generated also permit analysis of determinants of labor transitions to higher productivity jobs.

The Jobs Diagnostic work above yields a wide variety of labor market insights for growth, but four are
of especial relevance for Ugandan agriculture. First, and foremost, structural change in poor countries
involving labor transfers to higher-productivity jobs (typically outside agricultural production) tends to
happen fastest, and economic growth tends to be higher, when agricultural productivity is rising. Second,
while aggregate productivity rises as low-productivity agricultural labor moves to higher-productivity non-
farm—typically urban—jobs, average labor productivity in the receiving sectors tends to fall. Third, most
creation of new “better� (higher value-added per worker) jobs in the non-farm production sectors is by
new, smaller firms. Larger established agri-business firms maintain their position during growth spurts,
but tend to not create new non-farm jobs. Fourth, formal non-farm jobs tend to be associated with very
high capital intensity relative to the resource endowments of the poor countries concerned. The
implication of these four findings is that a jobs-oriented growth strategy for poor countries requires: (i)
boosting agricultural productivity, (ii) improving the skill base of rural-to-urban migrants to capture the
benefit of the demographic transition, (iii) promoting more labor-intensive technologies, and (iv)
spreading capital more evenly across jobs in both agriculture and non-agriculture.
                                                    5


Although similar geographically and demographically to other Sub-Saharan African countries, three key
relevant trends in Uganda stand out compared to most of its neighbors. First, real GDP growth per capita
in Uganda has been quite volatile, generally high from the 1999 to 2011 average of over 6.6 percent per
year, but with a pronounced slowdown to an average 1.1 percent per year from 2013 to 2017. This is low
compared to the 3.3 percent in Tanzania, 3.2 percent in Kenya, and the 2.8 percent in Rwanda.13 Uganda
also stands out in terms of the youthfulness of its workforce. Births per mother declined from about 7 per
mother in the mid-1990s to 5.7 in 2015, a decline much slower than in neighboring countries. Labor force
growth from 2013 to 2017 was 3.8 percent per year, compared to 3.3 percent in Tanzania, 3.2 percent in
Kenya, and 2.8 percent in Rwanda. This suggests that Uganda needs to be even more concerned than its
neighbors about creating large numbers of jobs for young people for some time to come, even though its
economy has slowed.14

The Ugandan pattern of industrial (and especially manufacturing) stagnation and growth of the service
sector is widespread in Africa. In a few cases, the services in question are “tradables�, in the sense of
tourism and ICT exports (or import substitutes) that bring in (save) foreign exchange. In most cases,
however, services are urban-based, most often informal, and “non-tradable�.15 Moreover, they have been
stimulated by public spending on urban wages and consumption financed by foreign aid and trade taxes.16

A recent study analyzed economic and overall labor market trends in Uganda from 1992 to 2016 and
their relationship to GDP and demographic trends.17 Labor force participation rose impressively during
the decade of rapid real GDP growth after 1999, rising from 74 percent in 1999 to 89 percent in 2012, with
particular gains for women and youth in urban areas. However, it fell with the economic slowdown back
to 79 percent in 2016. Labor force participation for youth fell from 80 percent in 2012 to 60 percent in
2016, a result that should elicit deep concern. The share of employment directly in agriculture fell from
77 percent in 1999, to 72 percent in 2013, and again to 64 percent in 2016.18 The eight percent fall in the
share of agriculture between the 2013 and 2016, derived from the Uganda National Household Surveys
(UNHS), is matched by an eight percent increase in the share of services in the Ugandan economy.19
Despite these sectoral changes, the share of rural employment (ages 15 to 64) was still 75 percent in 2016
(according to the UNHS), down only two percent from 2012 but down 10 percent from 2005.20

Thus, Ugandan employment remains mostly rural and informal, but trends differ from EAC neighbors in
only a few respects. Overall, the share of rural employment and the share of employment in agricultural
production in Uganda are about 10 percent higher compared to other EAC members, differing widely only
with the low share of employment in agriculture in Kenya, estimated at 38 percent in 2017.21 Further,
Uganda does not stand out in East Africa in terms of its high share of informality in work (all sectors).
Wage employment was 24 percent of total employment in 2016, and the share of wage employees with
formal contracts was only 32 percent for all workers aged 25 to 64 (12 percent for youth).22 The share of
all employment that is waged in other EAC members in 2017 ranged from 14 percent in Tanzania to 38
percent in Kenya.23
                                                    6


        Trends in value-added, employment, and labor productivity in agriculture

Similar to neighboring countries, while the number of people employed in agriculture in Uganda has
increased, labor productivity (per worker, per year) in agriculture is much lower than the rest of the
economy.24 Using household survey data per worker, per annum (not per hour worked), labor productivity
per agricultural worker, including smallholders, was estimated to be at 13 percent of that for workers in
other sectors in 2013. However, labor productivity per person, per hour worked was found to not be much
different, with 1,850 hours/year spent in non-agriculture (seven hours/day) vs. 700 hours/year (2.7
hours/day) in agriculture.25 The apparent disparity in labor productivity results mainly from the
seasonality of agriculture. Rural households often have insufficient opportunities to access wage
employment in non-agricultural businesses to smoothen labor calendars. In contrast, urban households
have more access to countercyclical work, resulting in smoother labor calendars and lower poverty rates.26

Uganda exhibits two significant differences compared to neighbors over the last two decades that
suggest a growing problem with agricultural labor productivity. The share of agriculture in GDP in
Uganda is falling, while the share of agriculture in total employment continues to be high. The evolution
from 1999 to 2017 of sectoral shares in GDP in Uganda and three comparable EAC members are shown in
Table 1.27 The absolute decline in share for agriculture in Uganda over the entire period was roughly 6
percent, compared to a decrease of 2 percent in Tanzania, and an absolute increase in share of 3 percent
in Kenya. The annual compound rate of decline of the sectoral share of agriculture in GDP from 2009
onwards (-1.1% per annum) is also shown in Table 1, along with positive growth rates for the same period
in Kenya, Rwanda, and Tanzania.

Surprisingly, Uganda’s employment share in agriculture in 2015-2017 was virtually the same as in 1999-
2001, despite its 6 percent absolute loss of share in value-added. By contrast, the absolute share of
agriculture in total employment fell 8 percent in Kenya, 10 percent in Tanzania, and nearly 22 percent in
Rwanda over the same period. The implied loss of labor productivity in agriculture for Uganda relative to
neighbors, and the failure to move low-productivity farm workers into higher productivity non-farm jobs,
illustrates a slower rate of structural transformation than for neighbors likely to become more competitive
in agricultural production relative to Uganda over time.

Labor productivity growth in Ugandan agriculture appears to have slowed from around 2010 onwards,
unlike in neighboring countries. A “macro� approach can be used to roughly estimate trends in sectoral
labor productivities over 1999-2017, using the same national accounts data and International Labor
Organization (ILO) imputed data on employment (see Tables 1 and 2). This is useful for seeing macro
trends, but is not a substitute for more accurate results culled from household-level employment data.
Table 3’s columns A and B repeat the compound annual growth rates estimated respectively in Tables I
and 2 for sectoral growth in value-added shares and employment shares 1999-2017. Adding the
corresponding cells together in column C yields rates of change in sectoral shares of labor productivity
growth.28 Adding in the compound annual growth rates of overall value-added per person employed from
ILO reported in the World Development Indicators (WDI) (shown in Table 3) yields a rough estimate of
compound annual labor productivity growth rates for sectors and time periods specified, in column D. For
                                                      7


Uganda, estimated labor productivity growth in agriculture was decent at 2.1 percent annually from 1999-
2011 (bearing in mind that this only covered persons that ILO counted as “employed�, which does not
appear to include most of self-employed smallholder farmers). Nonetheless, growth in labor productivity
by the same measure was almost nil for agriculture in Uganda from 2009 to 2017, unlike 1999 to 2011. By
contrast, labor productivity growth in agriculture was over 6 percent in Kenya, almost 6 percent in
Rwanda, and 3.5 percent in Tanzania annually from 2009 to 2017.

Labor productivity in Ugandan firms in agricultural supply chains differs based on sub-sector and size of
firm. From survey data in 2001 and 2010,29 Table 4 shows results for four ISIC revision 4 2-digit sub-sectors
in agricultural value chains from production through food and beverage retail. It details changes in
employment, labor cost per worker, and total real value-added per worker in 2010. Similar results for the
financial sector (not restricted to agriculture or rural areas but excluding pensions and insurance) are
shown for comparison because they represent a highly urban and capital-intensive sector with different
drivers of value added per worker compared to agricultural supply chains30.

Labor productivity growth (change in real value added per worker) in formal agricultural (primary)
production has been strong on average in medium-sized firms. Labor productivity for these firms grew
more than five-fold between 2001 and 2010, but declined for small firms (less than 20 workers) by about
one-quarter over the same decade (see Table 4). Conversely, labor productivity for medium-sized firms in
food manufacturing grew about 66 percent over the decade, and labor productivity in small firms grew
284 percent, but declined by 21 percent for large firms (greater than 99 workers, of which there are
relatively few). In food and beverage retail, medium-sized firms exhibited relatively higher labor
productivity growth. Finally, for comparison, financial services firms of all sizes together experienced
negative labor productivity growth (mostly urban) over the same period. Even so, average labor
productivity in financial services in 2010 was on average almost seven times higher when compared to
formal sector agricultural production, and 1.6 times higher compared to food manufacturing. In both
cases, these labor productivity ratios were about half what they were in 2001, showing that average labor
productivity in formal sector agriculture and urban non-agriculture converged significantly between 2001
and 2010. However, they are still far apart in the case of primary agricultural production vis-à-vis other
activities.



Agricultural growth trends and total factor productivity in agriculture since 1961

Aggregate agricultural production growth has been low and erratic in Uganda since 1961.
Understanding why requires decomposing agricultural growth into its various components to isolate
growth due to a combination of increased efficiency and technological change.31 Figure 1 decomposes
agricultural growth into components due to area expansion (in orange); increased use of inputs other than
land, including labor (in grey); and total factor productivity growth (TFP). TFP is calculated as the residual
obtained after netting out output from all other measurable sources of growth, such as agricultural land
expansion, increased numbers of workers living on farms, and increased use of purchased inputs like
                                                      8


fertilizer.32 Growth in the TFP residual reflects a combination of increased technical efficiency, allocative
efficiency, and (or) technological progress. Improved technical efficiency derives from redeploying existing
inputs, land, and labor regardless of prices in a way that leads to net physical gains using existing
technologies33; gains can be thought of as catching-up to good practices. Greater allocative efficiency
arises from taking into account resource allocation, the costs of using different inputs and factors in
addition to technical issues, to maximize profitability; gains here can be thought of as the “art of business�,
since private sector gains stem from using inputs and choosing outputs more profitably. Finally,
technological change embodies scientific and technical innovation to get more from less.

Between 1991 and 2000, overall agricultural growth averaged 2.5 percent annually, lower than average
annual population growth above 3 percent in the same period. From 2001 to 2014, agricultural growth
was negative, averaging -0.2 percent year, with per capita growth decreasing an average of -.3 percent
per annum. Although crop estimates for 2005-2014 show that cereal yields improved steadily from 1.6 to
2.0 metric tons per hectare (mt/ha), root crop yields halved during this period. Cash crops also performed
poorly. For example, tobacco yields declined on average by 3.2 percent annually and coffee yields
stagnated.34 Within Sub-Saharan Africa, Uganda has one of the lowest adoption levels of improved seeds,
inputs, or mechanized traction.35 In 2014, only 16 percent of farmers used purchased inputs of fertilizer
or pesticide.36 The use of inorganic fertilizers remains concentrated on larger and more commercially-
oriented farms in the central region, on which cash crops such as tea, coffee, and increasingly sugarcane
or oil palms are grown. Fertilizer-use has been increasing slowly in Uganda in recent years, but the average
application level was still below 2 kg/ha in 2014, well below the already low average for Sub-Saharan
Africa of 16 kg/ha.37

Total factor productivity (TFP) growth has been largely absent from Ugandan agriculture for the last
three decades. As illustrated in Figure 1, TFP weakness has been worsening, especially since 2010.
Although not clear what underlies Uganda’s massive losses in TFP since 2000, it seems that these losses -
particularly large outside the cereals sector - stem from growing pest and disease incidence and are likely
related to policy distortions that harm allocative efficiency. They also likely reflect insufficient public
priority on innovation in Ugandan agriculture.


Trends in Agricultural Commercialization, Agro-processing, and Agro-industrial
Development in Uganda

Uganda has created jobs and higher labor incomes from agri-business, agro-processing, and food
exports. Between 2001 and 2010, the number of businesses in Uganda expanded almost threefold to
458,106, with 78,753 businesses specifically identified as being in agriculture, forestry, fishing, food
processing, and restaurants (UBOS 2003, 2011). This count does not include any of the 87,236 businesses
identified in the 2010 census as having more than one employee and being primarily engaged in trade,
although perhaps half of the latter could be counted as downstream or upstream from agriculture based
on the importance of agricultural and fish trade. The 2010 census also shows that 268,317 Ugandans were
formally employed through businesses in agricultural production (excluding smallholder farm families
                                                      9


working on their own farm), food and beverage processing, and food and beverage services retail (see
table 4). This represents a 14.6 percent annual growth rate in employment in these sectors from 2001 to
2010, and more than a quarter of all formal business employment in 2010.38

Across the food and beverage value chains, large food manufacturing has created the most jobs in
Uganda since 2001. As shown in Table 4, employment growth from food manufacturing alone grew nearly
400 percent from 2001 to 2010, at an annual compounded rate of 16.5 percent. The Uganda Bureau of
statistics (UBOS) databases underlying Table 4 (UBOS 2003, 2011) show that 47 percent of employment
in food manufacturing in 2001 was in firms with at least 100 employees. In 2010, the corresponding share
was 76 percent, illustrating that large firms accounted for most of the growth in employment, even though
employment in all size firms grew. Furthermore, Table 5 shows that in 2010, food manufacturing
accounted for half of all manufacturing jobs in Uganda, and more broadly 2-digit sub-sectors downstream
from agricultural production accounted for two-thirds of all manufacturing jobs.

Around 30 to 40 larger companies import and wholesale agricultural inputs, and hundreds of small
traders re-sell to individual farmers (1,992 agro-input dealers according to a 2008 census). The Uganda
National Agro-lnput Dealers Association (UNADA) has 1,300 members, including 48 larger scale ones. Seed
traders are another important stakeholder, organized in the Uganda Seed Trade Association (USTA) with
18 ordinary members. As discussed below, ensuring the quality of traded seeds, fertilizers, and other
inputs is a key bottleneck for increasing sector performance. However, growth of this sector is held back
by institutional uncertainties in the public sector, most particularly by free distribution to selected farmers
of agricultural inputs by State actors, and widespread fake ingredients contamination.39

Substantial jobs growth has stemmed in recent years from establishment of large-scale commercial
farms, especially in northern Uganda, devoted to grain and pulse production for sale to the World Food
Programme (WFP) and exports to South Sudan and Democratic Republic of the Congo (DRC). These have
typically been developed by foreign investors, while farms of 10-200 ha are being developed by domestic
investors. These domestic commercial farms mostly came into being in the past decade. To some extent,
uncertainties over property rights have constrained this mode of job creation. Land conflicts are more
prevalent in districts with a substantial number of refugees, or where there is high population growth and
ethnically diverse communities. One study found that yields were 22 percent lower on land parcels
associated with disputes compared to parcels without.40 Land disputes not only affect productivity, but
also overstretch legal institutions, as it takes on average about 32 months to settle disputes (Justice Law
and Order Sector 2016). Disputes also negatively affect the viability of commercial farming investments,
particularly in Northern Uganda where 93 percent of land is under customary tenure and where refugee
influx is most pronounced.41

The value-added standouts in the food and drink sector from 2002 to 2011 have been beer, soft drinks
and bottled water, and edible oils. Since 2011, while beer has fallen into negative growth, sugar
processing has come into its own with about 15 percent growth annually, soft drinks and bottled water
have continued grower steadily at 9 percent growth annually, as has edible oils processing at 6.1 percent
annually. Coffee processing has also made a growing contribution to manufacturing value-added in the
later in the period at 7.5 percent per year, compared to negative growth in the earlier period. Sugar, soft
                                                     10


drinks, coffee, and edible oils appear to be the most likely candidates to attract increased private
investment if supply continues to grow to meet apparent demand.

Food and beverage manufacturing increased in importance in Uganda, driven mainly by domestic and
regional demand. These sectors were responsible for 56.8 percent of all manufacturing value-added from
2011/12 to 2015/16 period, a growth in share from the 53 percent in 2010 (see Table 6). Less than 16
percent of total manufacturing value-added in this period came from traditional coffee and tea; if
including processing of the other traditional export crops (cotton ginning and textiles, not shown), this
would rise to about 19 percent. This illustrates that the contribution of agriculture to manufacturing value-
added is not driven by traditional export crops.

Rapid growth of food and beverages is beginning to provide a bridge between low-wage agricultural
sector work and high-wage, low employment sectors such as banks. At the high end of the dualistic labor
economy, banking is a good example of labor costs in a primarily urban and primarily capital-intensive
industry. It represents one pole of the African dual economy with smallholder agriculture at the other
pole. Average labor costs per employee for large employers in financial services in 2001 and 2010 were
ten times more than for large employers in food manufacturing. Yet the latter were roughly double the
average labor costs in formal sector primary agricultural production or the food and beverage service
industries. The emergence of food manufacturing and food service jobs over the past two decades
provides a middle ground for wage employment in Uganda.


Trends in Agricultural Trade in Uganda

Agriculture as a sector is highly sensitive to the overall economic and trade climate. Overall economic
growth, both domestically and among trading partners, determines demand for agricultural output,
influences fiscal space and exchange rates, and affects the opportunity cost of labor and capital used in
agriculture. In Uganda, as elsewhere, the share of agriculture in the economy is shrinking as the economy
develops, as is to be expected. Despite remaining by far the largest source of livelihoods at roughly 70
percent of employment in 2016, the share of agriculture in GDP in Uganda declined from 55 to about 23
percent between 1990 and 2016. Meanwhile, the share of services jumped from 31 to 47 percent, and
the share of industry rose from 14 percent to 31 percent.42 Core inflation since 2013 has ranged between
five to six percent per annum. Food prices, a critical part of core inflation, decreased significantly in 2016
to 3.1 percent for the year compared to 6.7 percent in 2015.43 Nonetheless since 2006 food prices have
outstripped the overall CPI.

Agricultural products, primary and processed, accounted for 54 percent of total exports over the last
decade, and for 49 percent in 2016 (Table 7). Trade decreased globally in 2016 to a low point. While a
gradual decreasing over the long term as a share of total exports, agricultural products have nonetheless
displayed solid growth in nominal value, and unlike share of GDP, the role of agriculture in exports remains
high. As of 2016, total agricultural exports were more than fourfold their early 1990s level in nominal
terms, and more than threefold their early 2000s level. They also represent about 20 percent of Uganda’s
                                                    11


total foreign exchange earnings. Note that the figures represent only recorded (formal) exports, and UBOS
estimates that informal, unrecorded, exports amount to about 15 percent of all exports, but no
disaggregated data is available.44 Unrecorded exports are primarily regional.

Uganda remains a traditional exporter to world coffee markets, and coffee remains the country’s main
export, along with tea, tobacco, and cotton. Aggregate exports of these four crops tripled in nominal
value and increasing 2.4 times in real US dollar terms between the early 2000s and the early 2010s (Table
7). Coffee exports almost doubled, while the other three traditional exports rose between five and 10
times. Compared to the same baseline, traditional exports in 2016 more than doubled, even with coffee
exports decreasing 11 percent compared to the previous five-year average. After an export boom in the
first decade for the other three traditional commodities, with a peak of a 476 percent increase for tobacco,
their export growth continued at a more moderate pace in the recent decade, ranging from 98 percent
increase for tobacco to 126 percent for tea.

The country has also become a major supplier of non-traditional agricultural products. Including fish and
fish products, which have grown nine-fold in nominal terms during the past two decades, non-traditional
agricultural products have become the largest non-traditional export category (Table 7). Traditional
agricultural exports were larger in value terms than non-traditional ones through the 2000s, although
steadily diminishing. After 2010, non-traditional agricultural exports began to dominate, a trend likely to
grow. Significant concern arose about decline in export volumes in the late 2000s, a consequence of
declining catches, falling stocks, and overfishing45 especially in Lake Victoria. Ugandan fish exports and
fish products leveled off in the early and mid-2010s, consistent however with observed stable overall
catch from the country’s open fresh water bodies during the last five years. The largest markets for
Uganda’s fish and fish products exports are Hong Kong, OECD countries, Gulf countries, Israel, and the
U.S.., and increasingly include neighboring countries such as Rwanda, Kenya, and the DRC.

There are other rapidly expanding, non-traditional exports. These include sugar and sugar confectionary,
cocoa beans, vegetable oils and sesame seeds, cereals (maize, sorghum and rice and their flours), hides
and skins, beans and other legumes, flowers, and vegetables (Table 7). In some cases, the cumulative
growth of these exports is in the thousands of percent, increasing from a very small level to significant
levels in recent years. An example is sugar and sugar confectionary, now representing the second largest
non-traditional agricultural export at around US$100 million. Over the same baseline, maize exports
tripled by the early 2010s, and increased four-fold by 2016.

Growth in most non-traditional exports is driven by increasing demand in neighboring countries. This is
the case except for cocoa beans, hides and skins, and flowers, mostly shipped to OECD countries and China
along with traditional exports. The largest markets for Uganda’s cereal exports in 2016, for example, are
South Sudan ($70 million), Kenya ($38 million), Rwanda ($18 million) and the DRC ($12 million). Other
smaller markets are Burundi, Tanzania, and Sudan (ranging from $3 million to $1.5million) (ITC 2018a).

In recent years, the most significant cause of distress to Uganda’s regional export performance. By 2013,
South Sudan had become the largest destination for Ugandan exports. However, this market has become
                                                     12


extremely unpredictable due to serious unrest, and the resulting intermittent blocking of trade routes. In
addition, severe drought in Uganda in 2016 decreased production of agricultural commodities for export,
particularly maize and beans.46

Uganda is also a large importer of processed foods, and a growing one for fresh food, including cereals.
This trend is widespread in Eastern and Southern Africa.47 Processed foods accounted for 9.3 percent of
all Ugandan recorded imports from 2012 to 2016 period; fresh unprocessed) food accounted for 3.4
percent.48 While imports of processed food declined by 5 percent annually from 2012 to 2016, fresh food
imports increased at an annual rate of 21 percent. In particular, wheat imports, mostly from Russia,
peaked at US$165 million in 2014, ten times as much as recorded in 2012. Rice imports also peaked in
2014 at US$72 million and remained high in 2016 at US$47 million. Rising food import dependency
reached almost 13 percent by 2016,49 which increases Uganda’s vulnerability to global price fluctuations,
and its ability to generate foreign exchange.


Domestic, Regional, and Global Demand Shifts Will Create Further Opportunities for Job Creation in
Uganda

Demand-side opportunities for agriculture and food in Uganda and its neighbors are the strongest they
have ever been. This demand is both domestic and regional. Domestically, it is certainly driven by the high
population and urbanization rates. Urban income growth is also rapidly expanding the middle class, similar
to most countries in the region (Tschirley et al. 2015b). A detailed, comprehensive analysis of the 2013
household panels for rural and urban areas in Uganda in 2012/13 shows that household consumption is
very responsive to income growth (ie income elastic, ref Boysen 2016). As expected, income response for
food consumption expenditures are higher for the poorest quintile of households than for the richest, and
on average are higher in rural areas than urban ones. However, it is striking how high (>1 or elastic) mean
urban consumption response with respect to income is in both urban and rural areas for meat, fish, milk,
and fruits. Widespread demand for these items will likely continue to grow more quickly than income in
both rural and urban areas.

Increasing regional demand for food, and dietary shifts into higher-value and more processed foods,
offer massive opportunities for Ugandan farmers and value chains. Unlike domestic demand, which will
always be constrained by the relatively small size of domestic markets, regional and global demand is huge
and growing. Africa’s demand for food is projected to more than double by 2050, driven by population
growth, rising incomes, rapid urbanization, and more open intra-regional trade policies. The value of the
African food market is predicted to rise to US$ 1 trillion by 2030 from US$ 300 billion currently, with rapid
growth of both the urban and rural middle class.50 Diets increasingly move away from cereal and tuber
staples towards greater consumption of animal protein, fruits, and vegetables a(Ibid.).

Uganda has subscribed to a growing number of regional treatises and commitments concerning
agriculture. Regional treatises and commitments include those linked to the EAC and to Common Market
for Eastern and Southern Africa (COMESA), the East African Common Market Protocol (CMP) 2010, and
                                                    13


the East African Community’s Agriculture and Rural Development Strategy 2005-2030, and the COMESA
Seed Harmonization Implementation Plan (COMSHIP) validated by COMESA Member States in 2014.
While these regional commitments reflect agriculture having become a priority among many African
countries, implementation of regional commitments tends to lag among most members.

Trade patterns show the current expansion of regional agricultural trade. Examination of agricultural
incentives in the policy section below shows that Uganda has potential to expand regional agricultural
trade. Uganda benefits from both a very favorable resource endowment and a terrific location to meet
rapidly expanding demand from inland neighbors like South Sudan and the DRC. Uganda is especially well
placed for meeting rapidly rising regional demand for processed foods, maize, dairy, fish, and animal
products.

Uganda’s agricultural export performance is likely still well below potential. According to the ITC Trade
Performance Index’s global performance rankings, Uganda is ranked 44th in fresh food and 68th in
processed food. The comparable rankings for Kenya are 37 and 94, respectively; for Rwanda, 94 and 119;
and for Tanzania 40 and 99. Thus, compared to its neighbors, Uganda’s performance is average to better
than average. However, significant export potential for most traditional and non-traditional agricultural
products is left unrealized. To traditional destinations such as the OECD, ITC primarily sees substantial
underutilized potential for unroasted coffee, and to a small extent for cocoa beans. To both Sub-Saharan
Africa and other Non-OECD destinations, ITC calculates underuse of export potential in the high 40 to 80
percent range depending on the commodity.51

Domestic and regional trends in overall economic growth in the last year have been positive after a
difficult 2016 for Uganda. Growth in Sub-Saharan Africa as a whole is estimated to have rebounded to 2.4
percent in 2017 after slowing sharply to 1.3 percent in 2016 as commodity prices recovered, global
financing conditions remained favorable, and slowing inflation lifted household demand.52 This trend has
been especially pronounced in Uganda, where real GDP growth was estimated at 2.5 percent in CY 2016,
twice as much in CY 2017, and is likely to be well above 5 percent annually for the next two calendar
years.53 In contrast, neighboring Kenya, Ethiopia, Rwanda, and Tanzania are projected to have real GDP
growth rates ranging from 5 to 8 percent annually in 2017, well above the figures for Sub-Saharan Africa
as a whole. These rates are likely at least twice as high growth expected in the industrialized Europe and
North America countries representing traditional agricultural export markets (Ibid.). Furthermore, income
elasticities of demand for higher-priced, more processed foods and beverages are much larger in East
Africa than in the OECD, as the dietary transition is only just beginning to get underway in Africa. This
illustrates the stakes Ugandan agriculture has in contesting regional and other emerging country markets
for agricultural products that increase in demand with income.

Regional trade, particularly for agricultural commodities, has potential to stimulate growth and
improve living conditions for a high proportion of Ugandans employed in the agricultural sector.54 At
the same time, Uganda who must buy their food, mainly residents in urban areas, may find it more difficult
to buy higher value food items that are increasingly exported. The balance of overall costs and benefits is
expected to be positive for Uganda. However, there will most likely be adjustment costs for some, at least
                                                     14


initially. Over time, the added national income going to rural producers and urban processors and shippers
will generate growth that adds broadly to income opportunities. This means that it is vital to engage in
dialogue with all relevant actors in Uganda, for which the recently revived public-private dialogue
mechanism offers an appropriate framework.

Constraints and Prospects for Further Growth of Agro-industrial Development and Trade
in Uganda

Agribusiness development can increase farm productivity, but also create better job opportunities
along supply chains for the predominantly young African population. Dietary change is driving structural
changes in labor demand, a critical issue in African policy debates given that over 700 million youth are
predicted to enter the labor market over the next three decades.55 About 60 percent of the agricultural
labor force in Africa is already between 15 and 35 years of age, and the share of this age group is growing.56
An analysis of six Sub-Saharan African countries showed that transforming their food systems could add
more jobs than the rest of the economy between 2010 and 2025.57 Already in 2010, the number of jobs
in agribusiness amounted to 10 percent of the number of jobs in agriculture in Eastern and Southern
Africa.58 Labor productivity in agribusiness was up to seven times higher than in agriculture, depending on
the type of activity.

About 362,000 Ugandans were employed in formal jobs in the food manufacturing and food and
beverage services sectors in 2018. We can extrapolate that the 125,000 formal sector jobs in food
manufacturing and the 131,000 formal jobs in food and beverage services, as measured in 2010,59 now
represents 362,000 jobs. Although 2010 is the last comprehensive census of businesses, this simple
extrapolation is derived from adding two separate calculations using available data. First, sample surveys
provide annual data from 2011 to 2016 on the numbers of formal sector workers in food manufacturing
(UBOS 2017, and preceding issues). These suggest that the compound annual growth rate of food
manufacturing jobs in Uganda from 2010 to 2016 was 5.1 percent annually. This can be reliably
extrapolated two years to get a present estimate for food manufacturing jobs in 2018 of 186,000.
Similarly, a UBOS survey of employed urban labor force in 2015 allows an estimate of 3.7 percent annual
compound growth in that variable. Applying this to demand for formal food services (irrespective of any
income effects) yields a conservative estimate of growth in demand for food services of 3.7 percent
annually. This yields the estimate for 2018.

By 2028, we estimate that about 560,000 Ugandans will be employed in formal jobs in food
manufacturing and food and beverage services. Applying the same annual percentage to 2028 is less
accurate, suggesting the wisdom of using a conservative approach and growth rates. At 306,000,
employment in food manufacturing will exceed employment of 253,000 in food services, consistent with
the rise of a service-based urban economy. Formal sector manufacturing jobs in agro, food, and beverage
processing in 2010 accounted for two-thirds of all manufacturing employment in Uganda that year. Since
food manufacturing has grown faster between 2010 and 2016 at 5.1 percent compared to overall
manufacturing at 4.4 percent, this share is likely to continue to increase. More than the estimated 560,000
jobs in agro-processing and food services in 2028 can be created if processed agricultural exports grow
                                                     15


even faster than the more than 10 percent annual growth in real value added from agricultural exports
from 2006 to 2016 (Table 7). In addition, growth in persons employed in food services in 2010 would likely
be greater than the conservative estimate incorporated in the total for 2028 if Ugandan urbanization rates
rises closer to rates seen in other EAC countries.


Assessment of Challenges for Developing Agricultural Value Chains60

Numerous barriers to agricultural commercialization continue to exist for smallholder farmers in
Uganda. A study assessing factors driving commercialization in Uganda found that a main determinant
was access to physical, human, and financial capital.61 Larger farm holdings were more eager to
commercialize since they could realize economies of scale by adopting modern technologies. Farmers with
access to assets and connectivity to markets actively engaged in markets. The geographic dispersion of
smallholder farmers and poor infrastructure quality, in contrast, present barriers to commercialization.
The lack of rural roads and poor road maintenance hamper smallholders’ access to input and output
markets, drive up transaction costs, and lead many to pursue more subsistence-oriented practices.62 High
transportation costs are moreover a significant barrier to trade, and lead to lower farm gate prices or
higher market prices since traders exploit their market power over farmers.63

Access to finance is critical for smallholders to invest in better farming equipment and practices, but
also to improve their livelihoods more generally. Commercial banks provided 95 percent of all
agricultural finance in Uganda in 2013, but generally not to smallholders64, as in Uganda smallholders
rarely hold collateralizable property rights to land. While the financial inclusion of Ugandans in general
has improved markedly, agriculture-related finance made up only 8.4 percent of commercial bank lending
in 2013.65 Since the 2008 food crisis, formal financing for marketing has tended to grow more quickly than
for production or processing; however, production loans more than doubled in 2013 and processing loans
also doubled in 2014. As of 2014, formal loans for production, processing, and marketing each accounted
for about one-third of agricultural finance.66 This is unusual in East Africa, where the production share
(but not amount) of formal sector agricultural finance has tended to decrease over time due to the rise of
agribusiness post-harvest value chain lending67. The difference in Uganda may be explained by the surge
of new commercial production of maize in the north after the peace agreement.

Private sector investment in Ugandan agriculture—whether by smallholders on their own plots or by
large-scale investors—has been modest compared to Kenya, Tanzania, or Ethiopia. For smallholders,
this appears to be linked to difficulties in securing finance and uncertainties about whether they can retain
their land. Less than 20 percent of land can be said to have secure tenure of the type required as collateral
for formal loans. The guaranteed market for quality maize provided by long-term WFP procurement in
Uganda, coupled with sudden availability of contiguous larger plots of land from peace in the north, has
enabled significant private sector supply response to the Government’s liberalization of maize markets,
but insecurity of usufruct rights has limited this expansion as well.
                                                     16


Uganda has fragmented set of agricultural institutions with overlapping and unclear responsibilities,
leading to inefficient public spending. Public spending on agriculture and the institutional base for public
engagement in stimulating agricultural development in Uganda have gone through wide swings in
approaches over the past 20 years. This resulting inefficiencies in public investment exacerbate the deficit
in private investment in agriculture. Together the effects are manifested in poor sector productivity
growth, which ultimately must be at the heart of a pro-jobs growth strategy.

Agricultural extension services, which account for up to half of all public agricultural expenditure, are
ineffective. Extension services, once a world-renowned model of innovation in Uganda, are today
emblematic of the failure of public support to foster agricultural productivity growth. Services include
subsidized inputs distributed by the military for redistributive purposes rather than productive ones.
There is widespread recognition among agricultural stakeholders at all levels for the need to use scarce
funds to provide non-capturable public goods, such as improved research, extension, and institutional
development to facilitate widespread private investment (including smallholders) in the sector.

The combination of Uganda’s natural resources and agricultural skills place it in a strong position to
stimulate overall economic growth. Ugandans can sell into a new regional agricultural commodity boom
and growing demand for products such as confectionary, fish, maize, animal feed, vegetable oil, and dairy,
which have grown rapidly in Uganda’s exports since 2000 (Table 7). Likewise, Ugandan export growth has
been strong globally for high-value traditional coffee and tea exports, and for non-traditional ones such
as fish, cocoa, spices, and flowers. This demand is powered by high population growth, urbanization, and
income growth per capita throughout the region. Agricultural value chains everywhere—including in
Uganda—are rapidly becoming higher value, more processed, longer, wider, and more anonymous. They
are also becoming more demanding in terms of quality and reliability of shipments. Private firms in
Uganda and elsewhere in the region struggle to secure raw material supplies. Farmers, on the other hand,
have increasing difficulty achieving market recognition for their products sold in spot markets.

Larger sized food processing establishments in Uganda locate where infrastructure is accessible. The
spatial patterns of food processing establishments by size in Uganda indicate the strong correlation of size
of firm with infrastructure availability and other supply side factors, but demand is also clearly important.
A forthcoming mapping study shows that roughly half of the largest food processors (greater than 99
workers), and many of the smaller ones, are located in and around Kampala, the economic capital.68
Kampala, a major growing market, is also endowed with energy supplies and roads, and fosters economies
of agglomeration. National distribution of smaller firms is highly correlated spatially with the density of
smallholder farming, another supply-side and agglomeration driver. Not surprisingly, fish processing firms
tend to congregate at Lake Victoria.

Firm location data illustrates the importance of demand-side drivers in understanding the location of
agricultural processing firms. The large majority of firms outside the major Kampala domestic market are
located on the few major national all-weather roads. Yet it is striking that about one-third of the largest
food processing firms, and many of the smaller ones, are located close to export markets and national
borders, even relatively remote ones. There is a heavy concentration in Eastern province, near the Kenyan
                                                     17


border, another concentration in the Western province (Bundibugyo and Kabarole) near the DRC border,
another cluster in Mbarara on roads leading to and passing through Rwanda to the DRC, and, perhaps
most surprising, is the remote cluster in the northwest (Yumbe and Arua) close to the border with South
Sudan. The locations on the borders clearly are driven by demand form the other side of the border.

The primary constraint East African food processors face is reliable supply of raw material.69 Capitalizing
on growing urban and regional demand for processed and higher value foods requires vertical
coordination of smallholders into higher value supply chains. Under such arrangements, smallholders will
have incentive to produce more and better, and have access to the tools to do so. This would support
better post-harvest handling, particularly storage, at smallholder and aggregation/warehouses to reduce
post-harvest losses and improve prices received by smallholders. More skills and capital transfers, and
better branding can benefit smallholder suppliers.

Improving the implementation of Uganda’s agricultural regulatory environment will be key. This
includes a wide variety of agricultural issues such as import licenses, grading, and phyto-sanitary, sanitary,
and food safety regulation. The quality of agricultural inputs available in smaller quantities to
smallholders is a particular problem, including, but not limited to, those distributed by government.
Doubts about the quality of seed and fertilizers sold or distributed represent a serious barrier to producer
investment. They also harm productivity and explain why Uganda has one of the lowest inorganic fertilizer
use rates in the world. The weak regulatory system is directly related to the poor state of agriculture-
related public institutions. A pro-farmer, pro-business, and pro-jobs objective requires helping the “Made
in Uganda� label to strengthen sales of agricultural inputs and outputs.

ICT development in rural areas is critical to overcoming the remoteness of the multitude of diverse and
widespread small farms. These issues create high transaction costs for firms attempting to work with
smallholders, and vice versa. Rapidly expanding high-value agricultural supply chains require good
connectivity and reliable, accessible databases between suppliers and integrators for passing market and
technical information in both directions in near real time. Relatedly, it requires building trust, currently
lacking, among different actors along the supply chain. Vertical coordination works best over time for all
parties when market incentives are transparent, competitive, and both sides have recourse in the case of
non-performance.

Finally, a pro-jobs growth strategy in a natural resource-based sector such agriculture will increasingly
need to deal with climate change risks. Average temperatures in Uganda have increased by 1.3°C since
1960, and could rise by up to 2.5° by 2050. Seasonal rainfall has become more variable and less
predictable, with consequences for fluctuating yields. Crop and animal pests and diseases linked to climate
change are increasing problematic. While most countries in the same tropical latitudes as Uganda face
similar dangers from climate change, Uganda is one of the least adapted agro-economic system, and thus
one of the most vulnerable. This impacts the welfare of smallholders directly, and increasingly discourages
potential investors in more formal, long-term agricultural enterprises.
                                                   18


Recommendations70

Strengthen public institutions and policies for agricultural productivity . Uganda needs a rapid
turnaround in agricultural productivity growth, where it lags other countries in the region despite having
better agricultural resources. This is a necessary precondition for sustainable agricultural growth and,
arguably, for any truly pro-jobs growth strategy in the country.

Increase both the efficiency and amount of public spending for agricultural research and extension,
especially for smallholder farmers. Present extensive public procurement funds should be repurposed
for agricultural inputs for redistribution, on a free or subsidized basis, by agents other than qualified
extension personnel. The current extension system should move away from free distribution of sub-
standard inputs and should be rebuilt to increase its efficiency and effectiveness. The Government may
consider achieving equity support goals through means other than inputs distribution, for example using
social protection linked to climate-smart soil and water management practices. Extension should be
supported by adequate staffing, data collection systems and capacities, and could be improved by non-
governmental approaches, investments in radio programs, and farmer field schools.

Coordinate and strengthen the work of line ministries, extension agencies, and agricultural research
institutes. The Ministry of Agriculture (MAAIF) needs to coordinate public activities in the sector with
many public agencies and private firms. MAAIF needs to monitor public spending and funding, including
donor funding. It will be critical for MAAIF and its agencies to regain capability to design and carry out
agricultural policy and economics analysis, and monitoring and evaluation, especially under a shifting
macroeconomic environment. It will be vital to improve the institutional and human capacity of line
ministries to carry out activities such as agricultural public expenditure reviews, policy analysis
monitoring, and document experiences in this area. They need to be able to generate and use policy
analyses, monitor programs, and carry out credible technical evaluations. Without this capacity they will
always be at a disadvantage in dealing with domestic and external funding sources, and will find it
difficult to lead effectively. This has implications for staffing and institutional roles, and development
partners should assist with start-up costs.

Reform the regulatory environment to ensure consistent quality of inputs and outputs without undue
compliance burdens. To improve input quality, regulatory burden should be shifted from controlling
registration (such as licensing of traders), to controlling actual operations through random sampling, in
addition to regular controls of seed companies. Reforms should help support successful quality-
certification initiatives such as AgVerify (a pilot program to allow checking the authenticity of seed
package labels on the internet) and assess the potential for expanding its procedures from seeds to
fertilizer. Licensing procedures and import processing for improved inputs and new seed varieties should
be also be enhanced to reduce delays and to foster inputs for agribusiness development.

Promote commercialization of agriculture and trade through improving institutions and infrastructure.
To date, private sector successes are modest in agriculture in Uganda, except in processing driven by
urban and regional demand. An improved public framework for research, extension, regulation of input
quality, rule of law, infrastructure, and access to regional markets is needed. Under these circumstances,
the private sector is central for taking agriculture forward in a market-led economy, whether through
                                                     19


supplying, producing where plantation or other large-scale agriculture makes sense, developing markets,
or processing for export.

Improve access to agricultural finance along entire agrobusiness value chains. This includes better land
titling and enforcement, warehouse receipt programs, banking reforms, and the rule of law in commercial
matters. Mobile money transfers, value-chain financing, digitizing land titles, and warehouse receipt
systems (WRS) are promising approaches to de-risk the sector, to overcome the lack of collateralizable
land titles for loans, and to increase farmers’ access to finance. The over 500 Savings and Credit
Cooperatives (SACCOs) currently registered in Uganda could be better supported through inclusion in legal
banking frameworks, governance, and supervision mechanisms.

Assess policies related to transport and other transfer costs for agricultural commodities entering
regional trade. Costs for transporters, such as high fuel taxes and expensive equipment, should be
reduced and associated fiscal drag dealt with through measures that do not discourage exports . High-
level leadership is needed to build mutually beneficial trade relationships with neighbors that can endure
weather shocks, local shortages, and electoral cycles, and to contain border closures and other
administrative hinderances to trade, including arbitrary ones imposed at short notice by local officials.

Analyze and carefully plan investments in infrastructure—roads, electricity, communications, water—
which are especially critical to how agribusinesses will locate.71 Considerations related to raw material
source locations, perishability, intended markets, potential economies of agglomeration, economies of
scope and scale are all important factors. While specific recommendations go beyond the scope of this
note, we strongly recommend that Government urgently focus on how to promote secondary and tertiary
towns as growth poles for agriculture.

Invest in information systems and databases. Agribusinesses and government need better market
information, including agricultural transport flows and costs for producers by radio and cell to
demonstrate least-cost pathways and target areas needing attention. Public databases on key household-
level data and agricultural variables, such as market prices, are critical for boosting national
competitiveness, food security, and for decision makers to steer pragmatic policies based on evidence.
Consideration at a high level is needed to identify what data pertaining to food and agriculture are most
relevant, and how they can be collected and analyzed cost-effectively in near real time. The MAAIF should
provide leadership for data collection (for instance, about inputs, yields, post-harvest losses, soil quality)
management and use by a range of stakeholders.

Improve rural ICT. Larger-scale agricultural operations, and both vertical and horizontal agribusiness
coordination, require significant improvements in rural ICT. Identification and implementation of a set of
investments and policies specific to rural ICT for agriculture is critical. Modern infrastructure is needed to
support emerging technologies and services.72 Suggested measures include optimizing connectivity of
rural users by, for instance, using international undersea fiber optic cables arriving in the interior of East
Africa; encouraging private sector participation in ICT infrastructure development; and hastening
enforcement and awareness of ICT related (property) laws.

Continue to foster land tenure security through better land demarcation and delivery of adequate
documentation to land owners. Multiple initiatives ongoing for this purpose should be supported and
                                                     20


upscaled. Examples include the Systematic Land Adjudication and Certification (SLAAC) program, or the
GIZ-partnership with the Ministry of Lands, Housing and Urban Development (MLHUD). One way to
support faster resolution of land disputes would be to support relevant courts. Greater use could be made
of GPS-data, and of technologies such as drones to reduce the time and cost of data collection about field
boundaries. New technologies are already transforming the digitization and securitization of land records
in neighboring countries.

Pursue specific initiatives to promote agro-processing and encourage agro industry vertical
coordination. Agribusiness will benefit from all the above initiatives, but agro-processing typically requires
aggregation of products from many small farms. Initiatives to foster successful vertical coordination of
production into whole value chains capable of meeting increasingly strict quality and safety requirements
are needed. We suggest a three-pronged approach for the public sector to foster private vertical
integration. First, create a knowledge platform for recording and diffusing good practices in agricultural
vertical coordination arrangements. Second, clarify the legal status of vertical coordination agreements
and obligations, such as under contract farming. Third, identify and empower a public authority with tools
to oversee a level playing field in vertical coordination between farmer groups and aggregators, and across
firms, and assist in contract enforcement. Establishment and enforcement of grading and quality
standards for inputs and outputs will also facilitate vertical coordination by making it easier for
aggregating firms to offer incentives to suppliers to provide quality raw materials. These approaches can
also help attract multi-national agricultural investors willing and able to foster inclusive growth.

Support development of rural cooperatives. Policies limiting development of farmer cooperatives as
economic actors should be reviewed and revised. Policies have been in place to limit development of
rural cooperatives to avoid economic groups straying into politics, but the economic role of cooperatives
needs to be supported. Independent rural coops are important to strengthen the bargaining power of
smallholders under vertical coordination, and they also simplify aggregator interface with suppliers. They
also spread the costs and simplify the governance of ownership and maintenance of technical equipment
too expensive for individual smallholders, such as electronic soil scanners.

Increase resilience in agro ecosystems and of rural livelihoods. It is vital that Uganda take steps
immediately to increase agro-ecological systems resilience to climate change to protect rural livelihoods
and firms. The ongoing efforts to increase irrigation from its current very low level are critical given the
breadth of risks and the urgency low resiliency demands, but a much broader multi-stakeholder
approach is needed for these efforts to succeed High-level national leadership is needed to build
widespread consensus on the dangers and to develop a practical and coordinated action plan to
increase resilience through technology and better land and water management.

Increase collaboration between the many resiliency-building actors. The activities of climate and
agriculture-related ministries, research institutes, civil society and farmers’ groups, development
partners, and private sector actors need to be better coordinated. For instance, establishment of a multi-
stakeholder platform is needed. This can help identify and prioritize the most cost-efficient Climate-Smart
Agricultural (CSA) practices for a given agro-ecological zone and local context. This includes
implementation of the New National Irrigation Plan in consultation with stakeholders, building the
capacity of user groups, and adapting planning based on economic feedback from users and upstream
                                                 21


and downstream value-chain participants. All stakeholders need adequate and timely access to credible
meteorological, price, and disease-related information and early warning information. To support
vulnerable communities to develop local emergency response mechanisms aligned with national-level
institutions, it will be critical to coordinate a simple communication campaign across government
authorities, private sector entities such as telecommunication companies, academia and civil society.
                                                    22


References

Adjognon, S., Liverpool-Tasie, S. and T. Reardon 2017. Agricultural input credit in Sub-Saharan Africa:
   Telling myth from facts, Food Policy (67) February: 93–105.
African Growth Initiative at Brookings 2018. Foresight Africa: Top priorities for the continent in 2018.
Washington. D.C.: Brookings Institution.

Alliance for a Green Revolution in Africa (AGRA) 2017. Africa Agriculture Status Report: The Business of
    Smallholder Agriculture in Sub-Saharan Africa (Issue 5), Nairobi, Kenya: Alliance for a Green
    Revolution in Africa.
Bank of Uganda 2015. Agricultural Finance Year Book Bank of Uganda: Innovations and Research in
  Agricultural Finance, Ministry of Agriculture, Animal Industry and Fisheries (MAAIF) and Economic
  Policy Research Center, Entebbe: Uganda.
Christiaensen, L. and J. Kaminski 2015. Structural change, economic growth and poverty reduction –
  Micro-evidence from Uganda. Working Paper Series No. 229. African Development Bank Group,
  December. Available at: https://ideas.repec.org/p/adb/adbwps/2322.html
Christiaensen, L. and W. Martin 2018a. “Agriculture, structural transformation and poverty reduction:
  Eight new insights�. World Development (109) 413-416.
Christiaensen, L. and W. Martin 2018b. “Five new insights on how agriculture can help reduce poverty�.
  Washington, D.C., International Food Policy Research Institute Blog: Research Post, July 26. Available
  at: http://www.ifpri.org/blog/five-new-insights-how-agriculture-can-help-reduce-poverty
Delgado, C., J. Hopkins, V. Kelly et al. 1998. Agricultural Growth Linkages in Sub-Saharan Africa.
  International Food Policy Research Institute, Research Report No. 107. Washington, D.C., IFPRI.
  Available at: http://ageconsearch.umn.edu/bitstream/37908/2/rr107.pdf
Diao, X., P. Hazell, and J. Thurlow 2010. “The Role of Agriculture in African Development�. World
   Development (38) 10 October 1375-1383. https://doi.org/10.1016/j.worlddev.2009.06.011
Fardoust, S., Y. Kim and C. Sepulveda, eds. 2011. Post-Crisis Growth and Development: A Development
Agenda for the G20. (Washington. D.C.: The World Bank).

Food and Agricultural Organization of the United Nations (FAO) 2017a. The food security and nutrition–
  conflict nexus: building resilience for food security, Nutrition and peace, Regional Overview of Food
  Security and Nutrition in Africa 2017, Accra: Ghana.

Food and Agricultural Organization of the united Nations (FAO) 2017b. Agricultural Policy and Market
  Distortions in Uganda: A Synthesis of the Evidence, Monitoring and Analyzing Food and Agricultural
  Policies (MAFAP) program of the FAO, unpublished manuscript.

Garlati, P.; Mejia-Mantilla, C.; Merotto, D.; and Weber, M. 2018. “Towards a Better Labor Market in
Uganda: An analysis of labor markets transitions between 2009 and 2016�, Draft manuscript, May 17.

Gollin, D.; D. Lagakos, and M. Waugh 2014. “The Agricultural Productivity Gap�, Quarterly Journal of
Economics (129) 2, May 2014, 939–993. https://doi.org/10.1093/qje/qjt056
                                                     23


International Trade Commission (ITC) 2018a. Trade Map, (with data through 2016, accessed July 2018)
   at:
www.trademap.org/countrymap/Country_SelProductCountry_TS.aspx?nvpm=1|800||||TOTAL|||2|1|
   1|2|2|1|2|1|1

International Trade Commission (ITC) 2018b. Uganda – General Trade Performance (accessed March
2018) at: http://www.intracen.org/country/Uganda/General-Trade-Performance

Ivanic, M. and W. Martin 2018. “Sectoral Productivity Growth and Poverty reduction: National and
Global Impacts�. World Development. (109) 429-439. http://dx.doi.rg./10.1016/j.worlddev.2017.07.004

Joughin, J. and L. Adupa 2017. Governance and Incentive Aspects of Implementing Agricultural Reforms
in Uganda: Consultant’s report. unpublished manuscript.

McCullough, E. 2017. Labor productivity and employment gaps in Sub-Saharan Africa, Food Policy 67:
  133–152.
Merotto, D.; Weber, M.; and Aterido, R. 2018. Jobs Diagnostics: Facts and findings. Washington, D.C.:
World Bank, June.

Mwijagye, Paul. (14 September 2009) Uganda: Fish Exports Drop by U.S.$20 Million. East African
  Business Week. http://allafrica.com/stories/200909160722.html
Mwesigye, F. and T. Matsumoto 2016. The Effect of Population Pressure and Internal Migration on Land
  Conflicts: Implications for Agricultural Productivity in Uganda. World Development 79 (March): 25-39.
  Available at: https://doi.org/10.1016/j.worlddev.2015.10.042
Meyer, R. 2015. Financing agriculture and rural areas in sub-Saharan Africa: Progress, challenges and
  the way forward, Working Paper, International Institute for Environment and Development: London.

Nivievskyi, O., von Cramon-Taubadel, S. and S. Zorya 2010. Stages of Agricultural Commercialization in
   Uganda: The Role of the Markets, Discussion Papers No. 51, Georg-August University of Goettingen:
   Germany.

Oryokot, J. 2017. Uganda - Agricultural Technology and Agribusiness Advisory Services: P109224 -
  Implementation Status Results Report: Sequence 13 (English). Washington, D.C.: World Bank Group.

Sheahan, M. and C. B. Barrett 2014. Understanding the agricultural input landscape in sub-Saharan
   Africa: Recent plot, household, and community-level evidence. World Bank Policy Research Working
   Paper 7014, August 2014.

Timmer, C. P. 1988. The agricultural transformation. Chapter 8, Handbook of Development Economics,
   Volume 1, 275-331. Available at: https://doi.org/10.1016/S1573-4471(88)01011-3

Townsend, R., Benfica, R. M., Prasann, A., Lee, M. 2017. Future of Food: Shaping the Food System to
  Deliver Jobs, World Bank: Washington, DC.
Tschirley, D. L. et al. 2015. Africa's unfolding diet transformation: implications for agrifood system
   employment. Journal of Agribusiness in Developing and Emerging Economies 5(2), 102–136.
                                                   24


Uganda Bureau of Statistics (UBOS) 2003, A Report on the Uganda Business Register, 2001/2002. Uganda
  Bureau of Statistics, Kampala, Uganda, January.

Uganda Bureau of Statistics (UBOS) 2011, Report on the Census of Business Establishments 2010/11.
  Uganda Bureau of Statistics, Kampala, Uganda, December.

Uganda Bureau of Statistics (UBOS) 2017, Statistical Abstract. Uganda Bureau of Statistics, Kampala,
  Uganda.

World Bank (2007), Agriculture for Development. World Development Report 2008. World Bank,
  Washington, D.C.

World Bank 2013. Bridges across borders: unleashing Uganda's regional trade potential (Vol. 2): Main
  report (English). Uganda Economic Update; no. 1. Washington DC: World Bank.
  http://documents.worldbank.org/curated/en/432271468310733487/Main-report

World Bank 2015. Republic of Uganda: Financial Sector Review. World Bank, Washington, D.C. Report
  No. ACS13673 June 14.

World Bank 2016. The Uganda Poverty Assessment Report 2016. Farms, cities, and good fortune:
  assessing poverty reduction in Uganda. Washington, D.C., World Bank, Report No. ACS18391, 2016.
  Available at: http://pubdocs.worldbank.org/en/381951474255092375/pdf/Uganda-Poverty-
  Assessment-Report-2016.pdf

World Bank (2017), Sebudde, Rachel K.; Goffe, Valeriya; Daka, Dorothy; Safavian, Mehnaz S.. 2017. Step
  by step: let’s solve the finance puzzle to accelerate growth and shared prosperity (English). Uganda
  Economic Update; No. 8. Washington, D.C. : World Bank Group.
  http://documents.worldbank.org/curated/en/662191486394023103/Step-by-step-let-s-solve-the-
  finance-puzzle-to-accelerate-growth-and-shared-prosperity

World Bank. 2018a. Closing the Potential-Performance Divide in Ugandan Agriculture. Washington, D.C.:
  World Bank. ©World Bank.

World Bank, 2018b. World Development Indicators. Various, accessed online.

World Bank 2018c. Global Economic Prospects 2018. Washington, D.C., World Bank. January. Available
  at: http://pubdocs.worldbank.org/en/575011512062621151/Global-Economic-Prospects-Jan-2018-
  Sub-Saharan-Africa-analysis.pdf

World Bank 2018d. Uganda Economic Update—11th Edition World Bank, Washington, D.C., May

World Bank forthcoming. “Uganda Jobs Diagnostic�. Draft report March 22, 2018.

Yeboah, F. and T. Jayne 2016. Africa’s evolving employment structure: Trends and drivers of change.
  Presentation at the Eye on Africa Seminar Series, October 20, 2016, African Studies Center: East
  Lansing, MI, Available at
  http://fsg.afre.msu.edu/gisaia/Yeboah_and_Jayne_Africa's_employment_structure_eye_on_Africa.p
  df
25
                                                  26


 Table 1: Sectoral Shares of GDP 1999-2017 Uganda and Selected Other EAC Countries

                                         (A) Annual% value-            (B) Compound Annual
                                          added (% of GDP)                  Growth Rates
                                        1999- 2009- 2015-              1999- 2009- 1999-
                                         2001    2011    2017           2011    2017     2017
   Uganda           Agriculture          30.0    25.8     24.2         -1.5% -1.1% -1.3%
                    Services             42.4    47.4     47.5          1.1%    0.0%     0.7%
                    Industry             20.2    19.5     20.2         -0.3%    0.6%     0.0%
                    Other                  7.4     7.3     8.0         -0.1%    1.7%     0.5%

   Kenya            Agriculture          28.4       24.8      31.3      -1.3%      3.9%       0.6%
                    Services             45.3       48.2      45.4       0.6%     -1.0%       0.0%
                    Industry             15.1       18.7      17.4       2.2%     -1.2%       0.9%
                    Other                11.2        8.3       5.9      -3.0%     -5.6%      -4.0%

   Rwanda           Agriculture          37.3       29.2      29.4     -2.4%       0.1%     -1.5%
                    Services             45.5       48.4      47.2      0.6%      -0.4%      0.2%
                    Industry             16.3       15.2      16.4     -0.7%       1.3%      0.0%
                    Other                 0.9        7.3       7.0     23.4%      -0.6%     13.8%

   Tanzania         Agriculture          31.4       29.8      29.5      -0.5%     -0.2%      -0.4%
                    Services             44.2       44.1      38.9       0.0%     -2.1%      -0.8%
                    Industry             18.1       20.6      25.2       1.3%      3.4%       2.1%
                    Other                 6.4        5.5       6.4      -1.5%      2.8%       0.1%


Source: Data are from World Bank, World Development Indicators, last updated June 28, 2018. The sectoral
definitions are as follows: "Agriculture" covers activities under ISIC rev. 4 codes 1-5, and includes cropping,
livestock, forestry and fishing; "Services" includes ISIC rev. 4 codes 50-99; "Industry" covers ISIC rev. 4 codes
10-45, including both construction and manufacturing. Together the three categories cover well over 90%
of economic activity in the countries and years in question (with the narrow exception of Kenya in 1999-
2001). “Other� is a residual calculated by the author that permits summing all value-added shares to 100%
and mainly covers mining and extractives and certain transportation services (ISIC codes 6-9 and 46-49).
The compound growth rates in Columns (B) are computed between mid-points of the three-year averages
shown.
                                                     27


                       Table 2: Changes in Employment Share by Sector in Uganda
                              and Selected other EAC Countries 1999-2017

                                                 (A)% of All          (B) Compound Annual
                                               Employment                    Growth
                                          1999- 2009- 2015-           1999- 2009- 1999-
                                          2001     2011 2017           2011    2017 2017
 Uganda              Agriculture            70.4    71.0     70.1      0.1% -0.2% 0.0%
                     Services               21.9    21.7     23.0     -0.1%    0.8% 0.3%
                     Industry                7.8     7.2      6.9     -0.7% -0.8% -0.8%

 Kenya               Agriculture            45.8     39.1     37.9    -1.5%     -0.4%   -1.2%
                     Services               40.6     45.4     47.7     1.1%      0.7%    1.0%
                     Industry               13.7     15.4     14.3     1.2%     -1.0%    0.3%

 Rwanda              Agriculture            88.6     79.6     66.9    -1.1%     -2.5%   -1.7%
                     Services                8.7     14.8     25.0     5.4%      7.8%    6.8%
                     Industry                2.7      5.6      8.1     7.6%      5.5%    7.2%

 Tanzania            Agriculture            77.2     72.1     67.2    -0.7%     -1.0%   -0.9%
                     Services               18.1     22.4     26.8     2.2%      2.6%    2.5%
                     Industry                4.8      5.4      6.0     1.3%      1.5%    1.5%
Source: World Bank, World Development Indicators (WDI), last updated June 28, 2018.
Columns (A), “% of All Employment� by sector are three-year averages computed from ILO modeled values in
ILOSTAT, reported in WDI.
Columns (B) are compound annual growth rates between the mid-year points of the averages in Columns (A).
                                                              28


     Table 3: A Macro Approach to Estimating Trends in Labor Productivity 1999-2017 by Sector in Uganda
                                      and Other Selected EAC Countries
                                                                             (C) Change in Sectoral       (D) Annual Sectoral
                           (A) Growth in % of          (B) Growth in
                                                                                 Share of Labor                 Labor
                              value-added            Employment Shares
                                                                              Productivity Growth        Productivity Growth
                         1999-    2009-    1999-    1999-   2009-    1999-   1999- 2009- 1999-          1999- 2009- 1999-
                          2011     2017     2017     2011    2017     2017    2011 2017 2017             2011    2017     2017
Uganda    Agriculture    -1.5%    -1.1%    -1.3%    0.1%    -0.2%     0.0%   -1.6%    -0.9%    -1.3%     2.1%     0.2%       1.5%
          Services        1.1%     0.0%     0.7%    -0.1%    0.8%     0.3%    1.2%    -0.8%    0.4%      4.9%     0.4%       3.2%
          Industry       -0.3%     0.6%     0.0%    -0.7%   -0.8%    -0.8%    0.4%     1.4%    0.8%      4.1%     2.5%       3.6%
          Overall growth in value-added per person employed
                          3.7%     1.1%     2.8%
Kenya     Agriculture    -1.3%     3.9%     0.6%    -1.5%   -0.4%    -1.2%    0.2%     4.4%    1.8%      2.1%     6.2%       3.7%
          Services        0.6%    -1.0%     0.0%    1.1%     0.7%     1.0%   -0.5%    -1.7%    -1.0%     1.4%     0.1%       1.0%
          Industry        2.2%    -1.2%     0.9%    1.2%    -1.0%     0.3%    1.0%    -0.2%    0.6%      2.9%     1.6%       2.6%
          Overall growth in value-added per person employed
                          1.9%     1.8%     2.0%
Rwanda    Agriculture    -2.4%     0.1%    -1.5%    -1.1%   -2.5%    -1.7%   -1.4%     2.6%    0.3%      4.0%     5.9%       5.1%
          Services        0.6%    -0.4%     0.2%    5.4%     7.8%     6.8%   -4.8%    -8.2%    -6.5%     0.6%    -4.9%       -1.7%
          Industry       -0.7%     1.3%     0.0%    7.6%     5.5%     7.2%   -8.3%    -4.2%    -7.1%    -2.9%    -1.0%       -2.3%
          Overall growth in value-added per person employed
                          5.4%     3.3%     4.8%
 Tan-
 zania    Agriculture    -0.5%    -0.2%    -0.4%    -0.7%   -1.0%    -0.9%    0.2%     0.8%    0.5%      3.5%     3.5%       3.7%
          Services        0.0%    -2.1%    -0.8%    2.2%     2.6%     2.5%   -2.2%    -4.7%    -3.3%     1.1%    -2.0%       0.0%
          Industry        1.3%     3.4%     2.1%    1.3%     1.5%     1.5%    0.0%     2.0%    0.6%      3.3%     4.6%       3.9%
          Overall growth in value-added per person employed
                           3.3%     2.7%      3.2%
     Data: All underlying data are from World Bank, World Development Indicators (WDI), last updated June 28, 2018.
     The figures in the table are computed values as detailed below.
     Methodology: All figures shown are compound annual growth rates between the mid-point years of annual three-
     year averages of the underlying variables: 1999-2001, 2009-2011, and 2015-2017.
     The sectoral definitions are as follows: "Agriculture" covers activities under ISIC rev. 4 codes 1-5, and includes
     cropping, livestock, forestry and fishing; "Services" includes ISIC rev. 4 codes 50-99; "Industry" covers ISIC rev. 4
     codes 10-45, including both construction and manufacturing. Together the three categories cover well over 90% of
     economic activity in the countries and years in question. Not covered are ISIC rev. 4 codes 6-9 and 46-49, mainly
     mining and extractive industries and certain transport services.
     Columns (A), “Growth in % of value-added� are computed directly from estimates of annual value-added shares by
     sector in WDI (see Table1).
     Columns (B), “Growth in Employment Shares� are computed from annual modeled ILO estimates of annual
     sectoral shares in total employment reported in WDI (see Table 2). In this case, the three main sectors sum to
     100%, so there is a minor discrepancy in definition with the sectoral definitions used for value-added.
     Columns (C), “Change in Sectoral Share of Labor Productivity Growth�, are computed as (A) + (B) for the
     corresponding cells, which is an identity if the minor discrepancy in definitions noted above is ignored, and bearing
     in mind that labor productivity growth is used synonymously with growth in value-added per employed worker,
     and that the sum of two growth rates is equivalent to the growth rate of the product.
                                                      29


Columns (D), “Annual Sector Labor Productivity Growth� are obtained by (C) + the corresponding annual
compound rate of growth in overall real value-added per person employed, using the same logic as in Columns (C).
The annual compound rate of growth in overall real value-added per person employed is computed for the time
periods of interest from ILO estimates of annual GDP per person employed per country in constant 2011 dollars,
reported in WDI, and are reported in the table for each country.
                                                        30



     Table 4: Labor Productivity Change 2001-2010 in Private Agriculture Related Firms in Uganda
                             (in constant 000’s 2010 UgSh, workers and%)

                   Crop &              Food               Beverage             Food &             Financial
                   Livestock           Manuf.             Manuf.               Bev. Svcs.         Services
                   Production          ISIC4=10           ISIC4=11             ISIC4=56           ISIC4=64
                   ISIC4=1                                                                        (for comp.)
 Employment
  in 2001          14,965           18,888            2,463                        42,562                4,363
 (workers)
 Value Added     LC/      VA/    LC/       VA/    LC/       VA/                LC/      VA/        LC/          VA/
                     1        2      1         2      1
 in 2001 (2010   wkr      wkr    wkr      wkr     wkr       wkr2               wkr1     wkr2       wkr1         wkr2
 UGS 000)
   All firms   1,971 4,584 2,981 16,624 15,307 84,044                          1,795    4,511 39,881        57,988
   Small firms 1,492 7,046 1,715 6,543 2,068               10,182              1,787    4,517 19,335        33,569
 (1-19 wkrs.)
   Med. Firms 1,047 3,282 3,415 23,165 3,015               29,799              1,943    4,520 49,429        65,836
 (20-99)
   Large Firms 2,174 4,305 4,013 24,124 17,770 95,910                          none     none      40,206    59,450
 (100+)
 Employment
                   36,980           93,988            5,755                       131,594             31,024
 in 2010
 Aggregate Change 2001 to 2010 (real terms)%
 Employment
 in 2010 All          147              398              134                           209                 611
 firms (% chg)
 Value Added Per Worker in 2010 (% real change from 2001)
   All firms     -27      106    -37        29    -89       -56                 -34         20     -60           -42
   Small firms
                 -52      -26     28       284    -28       105                 -35         15     -75           -71
 (1-19 wkrs.)
   Med. Firms
                 208      559     6         66    -42       -10                  6          127    -66           -74
 (20-99)
   Large Firms
                 -12      182    -59       -21     7        -53                none     none       -37           -1
 (100+)
Source: Data from Business Censuses carried out by the Uganda Bureau of Statistics (UBOS 2003, 2011).
Notes: All financial variable shown are in thousands of constant 2010 UgSh. In mid-year2010, US$1 exchanged for
UGS 2,230. Labor productivity in the table can be viewed from two perspectives:
(1) The average labor cost per person employed (LC/wkr) is the wage bill and associated costs divided by the
number of employees. This can also be thought of as the labor share of value-added on a per employee basis. (2)
Alternatively, total value-added per worker is real total value-added divided by the number of employees. It
includes remuneration to other factors of production, such as capital. It helps illustrate the relative capital (and
possibly skill) intensity of production. While it does not represent what each employee adds to production, it does
help illustrate what each employee is entrusted with on average. The financial services column is given for
comparison purposes; it excludes insurance and pensions.
                                                     31


                  Table 5: Manufacturing Employment in Uganda by 2-Digit Sub-Sector 2010

      ISIC4 2 digit Sub-Sector                                                    Employment        Share%
 10. Manufacture of food products                                                      93,988          50%
 11. Manufacture of beverages                                                           5,755           3%
 12. Manufacture of tobacco products                                                      321           0%
 13. Manufacture of textiles                                                            4,665           2%
 14. Manufacture of wearing apparel                                                    18,879          10%
 15. Manufacture of leather and related products                                        1,925           1%

     Sub-total downstream from agriculture                                             125,533        66%

 16. Manufacture of wood and of products of wood except furniture                           1,470      1%
 17. Manufacture of paper and paper products                                                  665      0%
 18. Printing and reproduction of recorded media                                            5,333      3%
 19. Manufacture of coke and refined petroleum products                                       635      0%
 20. Manufacture of chemicals and chemical products                                         4,616      2%
 21. Manufacture of basic pharmaceutical products                                           1,659      1%
 22. Manufacture of rubber and plastics products                                            1,704      1%
 23. Manufacture of other non-metallic mineral products                                     4,983      3%
 24. Manufacture of basic metals                                                            2,348      1%
 25. Manufacture of fabricated metal products, except machinery and
 equipment                                                                                 16,476      9%
 26. Manufacture of computer, electronic and optical products                                  56      0%
 27. Manufacture of electrical equipment                                                      778      0%
 28. Manufacture of machinery and equipment                                                 2,718      1%
 29. Manufacture of motor vehicles, trailers and semi-trailers                                 95      0%
 30. Manufacture of other transport equipment                                                 110      0%
 31. Manufacture of furniture                                                              18,990     10%

 Manufacturing employment (ISIC4 10-32)                                                188,169       100%
Source: UBOS (2011) and background data files from UBOS (2011).
                                                       32




         Table 6: Changes in Real Value-Added of Food and Beverage Manufacturing in Uganda
                                         2011/12 to 2015/16

                                                                                  Average
                                                                                                   Average
                                                                                  annual%
                                                                                                   annual%
                         Weight in total                                         compound
                                                 2011/12          2015/16                         compound
                       manufacturingvalue-                                         growth
       Item                                                                                      growth over
                             added                                               over 2002
                                               (2002 = 100)     (2002 = 100)                    2011/2012 to
                          over period                                                to
                                                                                                  2015/2016
                                                                                 2011/2012
                                                                                                    period
                                                                                   period
 All food and
 drink                        56.8%                 176              218            6.5%             5.5%
 processing1
                       Share of food and beverage sub-sectors in total manufacturing:
              Sugar
                              13.9%                 124              216            2.4%            14.9%
        processing
               Beer            9.9%                 282              216           12.2%            -6.5%
             Coffee
                               8.9%                  84              112            -1.9%            7.5%
        processing
   Soft drinks and
                               6.9%                 362              509           15.3%             9.0%
    bottled water
   Tea processing              6.8%                 122              131            2.2%             1.8%
   Edible oils and
                               4.2%                 275              349           11.9%             6.1%
                fats
              Total
   manufacturing              100%                  186              224            7.1%             4.8%
        all sectors
Source: Calculated from data in UBOS Statistical Abstract 2017
Note: (1) Weighted index that excludes tobacco. The sub-categories below this are the main sub-components in
terms of weight of this entry. Only sub-categories that had a weight of at least 4% of total manufacturing are
shown here.
                                                       33


     Table 7: Recorded Annual Agricultural and Food Exports in Nominal US$ Millions 1990 - 2016

                                1991-1995     1996-2000     2001-2005      2006-2010      2011-2015        2016
  Traditional Export
                                   225.2         331.4         191.3          398.3         569.3          507.2
  Crops
  Coffee                           209.2         285.7         118.3          284.6         415.4          371.7
  Tea                                9.2         28.3           34.2           54.7          77.3           71.5
  Tobacco                            6.7         17.4           38.7           59.0          76.6           64.1
  Cotton                             7.7         18.9           21.7           19.2          47.1           31.6
  Non-Traditional
                                    56.6         88.0          183.7          378.8         666.4          703.8
  Agric/Food Exports
  Fish and products                 12.8         35.0          100.1          128.0         128.7          121.5
  Sugar and
                                     0.1          5.8            2.3           38.0          84.9          100.3
  Confectionary
  Cocoa Beans                        0.5          1.3            5.5           22.3          50.8           75.0
  Maize                             16.6         10.4           16.2           26.7          52.1           70.3
  Animal/Veg Fats and
                                     0.1          2.3            5.5           46.2          98.6           62.1
  Oils
  Sorghum                            0.0          0.1            0.1           1.5           20.2           55.3
  Hides and Skins                    7.1          8.6           10.5           12.3          55.2           51.4
  Beans and other
                                     9.7          9.5            5.6           12.3          28.9           50.5
  Legumes
  Flowers                            1.1          6.0           21.5           24.3          25.8           24.6
  Other Non-Trad Agric
                                     8.5          8.9           16.4           67.2         121.2           93.0
  and Food Exports1
  Total Agricultural and
                                   281.7         419.4         375.0          777.2        1,235.7        1,211.0
  Food Exports2
  % Share of Total
                                    n.a.          n.a.          63%            54%           54%            49%
  Exports All Sectors
Source: COMTRADE database accessed via WITS and UBOS (2017).
Note: (1) In declining order of importance in 2016: rice, vegetables, sesame seeds, beer, mineral water, soybeans,
fruits, pepper, vanilla, groundnuts, live animals, bananas. Note that UBOS (2017) has the same commodity specific
totals, but also lists exports of “other commodities� of US$583.5 million in 2016, after accounting separately for
the items lumped together in this line item and for big-ticket non-agricultural items such as gold and petroleum.
None of this “other category� in UBOS (2017) is included here, although it seems likely that some of the items in
question are agricultural in origin.
(2) The cumulative rate of inflation of United States consumer prices in US$ was 76.2% from 1991 to 2016 (U.S.
Bureau of Labor Statistics Consumer Price Index). Therefore, the inflation-adjusted growth in US$ terms of
agricultural exports from 1991 to 2016 was +176% or a multiplicative factor of 2.76. This implies that in US$
inflation-adjusted terms, non-traditional agricultural and food exports grew by a factor of 7 from 1991 to 2016,
while total agricultural and food exports grew by a factor of 2.4 over the same period.
                                                        34


    Figure 1. Decomposition of sources of agricultural growth in Uganda 1961-2014 (% per annum)

           (Net agricultural output growth in% per annum at top of column in period indicated)
     %
      8
               5.5%
      6
                              -2.5%                                              1.0%          -1.1%
      4                                       2.7%
                                                              2.5%
      2

      0

      -2

      -4

      -6

      -8
              1961-70         1971-80         1981-90         1991-00           2001-10      2005-14

                                  TFP     Area expansion     Inputs intensification

Source: World Bank (2018a).
Notes: (1) The three sources of growth listed sum to output growth in the period in question. Output growth may
be different than column height because of negative contributions to growth in the period in question. Growth not
explained by area expansion or increased use per ha of inputs (including labor) is attributed to TFP. TFP is a
combination of increased allocative and technical efficiency, and technological change.
Source: Compiled from the USDA International Agricultural Productivity database, available at:
https://www.ers.usda.gov/data-products/international-agricultural-productivity/
                                                         35




1
  Both these findings were found to hold on a global basis by World Bank (2007), which had assembled a massive
amount of evidence to this effect.
2
  See Delgado et al. (1998), a partial equilibrium demonstration of agricultural growth multipliers at work in four
African countries in the 1990s based on household survey data. A far more comprehensive, elegant, up-to-date
economy-wide modeling investigation of the multiplicative impact on overall growth of growth of agricultural
tradables in selected LICs is found in Diao, X., P. Hazell, and J. Thurlow (2010).
3
  Unless otherwise referenced, the original sources of data and analysis for Ugandan agriculture in this policy note
are detailed in the new report, “Closing the Potential-Performance Divide in Ugandan Agriculture� (World Bank,
2018a). Note that the share of employment in agriculture fluctuated up and down between 77% in 1999 and 72%
in 2013 according to the Uganda National Household Surveys in various years but fell suddenly to 64% in the 2016
survey.
4
  World Bank 2015. Also see Christiaensen and Kaminski (2015).
5
  This is the conclusion of a series of peer-reviewed empirical studies recently published in a special issue of World
Development, see Christiaensen and Martin (2018a), also consistent with insights published in World Bank (2007).
6
  Ibid, also see Christiaensen and Martin 2018b for a summary of key points for a lay audience.
7
  Garlati et al. (2018) and Merotto et al. (2018)
8
  The difference between aggregate output growth and the growth of all inputs and factors of production that
produced it.
9
  World Bank (2018a)
10
   World Bank, World Development Indicators, accessed October 12, 2018, shows a growth rate for urban
population in Uganda of 5.8% per annum in 2017, compared to 5.7% in Burundi, 5.3% in Tanzania, 4.3% in Kenya,
and 2.8% in Rwanda. See: https://data.worldbank.org/indicator/SP.URB.GROW
11
   Following Merotto et al. (2018), “better jobs� is used synonymously with “higher labor productivity� jobs.
12
   Merotto et al. (2018)
13
   Arithmetic means of annual real growth rates per capita in domestic currency in the period concerned, from
data in World Bank (2018b).
14
   Garlati et al. (2018)
15
   African Growth Initiative at Brookings (2018); a similar view is developed in Fardoust et al. (2011).
16
   Ibid.
17
   World Bank forthcoming. Garlati et al. (2018) also uses the results of that work; labor transitions in Uganda were
analyzed from 2009 to 2016, using five waves of the Uganda National Panel Survey (UNPS) of households.
18
   These results from the Uganda National Household Surveys rounds in 1999, 2003, 2005, 2009, 2013 and 2016,
reported in Garlati et al. 2016. As suggested in note 1 above, the rapid transition out of agriculture since 2013
implied by the 2016 survey result is surprising, but conceivable. It does not match urban employment trends
however, and suggests rapid growth in informal service jobs in rural areas.
19
   Garlati et al. (2018)
20
   Ibid.
21
   World Bank, World Development Indicators.
22
   Garlati et al. (2018)
23
   World Bank, World Development Indicators. Data from the different countries may contain small differences in
what is counted in waged employment, but the main insights from the comparison is judged to be robust.
24
   This paragraph draws on “Closing the Potential-Performance Divide in Ugandan Agriculture� (World Bank,
2018a).
25
   McCullough (2017).
26
   Gollin et al. (2014) and McCullough (2017).
27
   Variable levels and compound growth rates in Tables 1, 2, and 3 are computed between the mid-points of 3 year
annual averages, to mitigate the arbitrary volatility of annual outcomes that affects rainfed agriculture and that
can greatly influence computed growth trends between any pair of years.
                                                         36



29
   Descriptive statistics cited by firm sub-sector, size group, and time period were computed by the World Bank
Jobs Group and supplied to the present author by Dino Merotto.
30
   Average labor cost per worker is a measure of average labor productivity that effectively nets out the share of
value-added due to other factors such as capital and land. Average total value-added per worker gives a sense of
average overall production per worker and is taken as average labor productivity for present purposes.
31
   This section is taken from World Bank (2018a), which includes more detail and a methodological appendix.
32
   This approach using national level data implicitly calculates a weighted average (by field area) of data for all
fields in the country, and assumes that the stock workers living on farms is a reasonable proxy for the flow of labor
to agriculture and that this relationship is constant over time. While the latter assumption in particular is
questionable, only panel data at the level of individual farms would allow escaping these limitations. The results
are sufficient for present uses.
33
   Conversely, it could also reflect losses from systematic decreases over time in the average quality of inputs, such
as land degradation or declining quality of fertilizer used, but this is not measured in the data.
34
   FAO (2017a)
35
   Sheahan, M. and C. B. Barrett (2014).
36
   Adjognon et al. (2017)
37
   Joughin, J. and L. Adupa 2017.
38
   Data are taken from ISIC rev 2 digit estimates for 2010 underlying more aggregated figures in UBOS (2013),
although there is some uncertainty in disaggregating “other manufacturing� and “trade� and thus these are not
included here. The Census of Business Enterprises in question reports only jobs in businesses enumerated in the
census.
39
   The issues here are complex and explored in detail in World Bank (2018a).
40
   Mwesigye and Matsumoto (2016).
41
   Joughin and Adupa (2017).
42
   Uganda Bureau of Statistics (2017).
43
   Ibid.
44
   Ibid.
45
   Mwijagye (2009).
46
   World Bank (2017).
47
   See Tschirley et al. (2015). To some extent, importing more cereals, especially feed grains, is an anticipated
counterpart to exporting more higher value foods such as vegetables and fruits more suited to Uganda’s
topography and climate. Importing more processed foods however may be a lost opportunity for development of
secondary cities, see World Bank (2018a).
48
   International Trade Commission (2018a).
49
   Ibid.
50
   Alliance for a Green Revolution in Africa (2017).
51
   International Trade Commission (2018b).
52
   World Bank (2018c).
53
   World Bank (2018d).
54
   See World Bank (2013).
55
   AGRA (2017).
56
   Yeboah and Jayne (2016).
57
   Townsend et al. (2017).
58
   Tschirley et al. (2015).
59
   As measured in the most recent Census of Business Enterprises in 2010.
60
   This section summarizes the main conclusions of a comprehensive study that itself is summarized in 92 pages in
World Bank (2018a).
61
   Nivievskyi et al. (2010)
62
   Oryokot (2017).
63
   FAO (2017b).
64
   World Bank (2015).
65
   Bank of Uganda (2015).
                                                         37



66
   Ibid.
67
   Meyer (2015)
68
   Preliminary results from forthcoming work at the World Bank by B. Blankespoor, D. Merotto, and T. Norman
(personal communication from D. Merotto) that uses the COBE database and the UBoS 2014 Population Census.
69
   This is a widespread result from firm surveys in East and Southern Africa, that is documented for Uganda in
World Bank (2018a).
70
   Like the previous section, this section summarizes the main recommendations of a comprehensive study that
itself is summarized in 92 pages in World Bank (2018a), to which the reader is referred for detail and evidence.

71
   This important topic and its spatial aspects goes beyond the scope of the present note, but is being investigated
in Uganda by World Bank teams and will be the subject of a future note.
72
   Interested readers ae referred to World Bank (2018a)