1
EMPLOYMENT INTENSITY OF NON-AGRICULTURAL GROWTH IN RWANDA Analyzing the links between Growth, Employment, and Productivity in Rwanda
Dickson Malunda (PhD)1 Abstract
Despite the intuitive notion that non-farm employment matters for poverty reduction, there is insufficient empirical research in this area in Rwanda. While there is a broad consensus that not all growth spells have the same impact on poverty, there have been relatively few attempts to systematically unpack the relationship between economic growth, employment and poverty reduction in Rwanda. The purpose of this study is to determine the employment intensity of economic growth in Rwanda in order advise government on which sectors have a higher potential to generate productive off-farm farm jobs that will drive a higher proportion of the population out of poverty. In this study, we use a three step framework to determine the main mechanisms through which economic growth translates into poverty reduction and job creation in the non-agricultural sector in Rwanda.
In this paper, we decompose changes in GDP and attribute to each component (employment, output per worker, capital and TFP) or to each sector a share of total observed growth. Our methodology uses Shapley decompositions, which is a simple additive method that links changes in a particular component to changes in total per capita GDP, by taking into account the relative size of the sector or component, as well as the magnitude of the change. The aim of this methodology is to understand how growth is linked to changes in employment, output per worker and population structure at the aggregate level and by sectors. In order to draw the profile of Rwanda's growth we disentangle the sources of output per worker growth: either Total Factor Productivity (TFP) growth, movements of employment from one sector to another, or changes in the capital-labor ratio. This framework will enable us to determine: first, the extent to which growth is associated with changes in non-farm employment (the quantity of jobs) or productivity (the quality of jobs) by sector of the economy.
1 Dickson Malunda is a Senior Research Fellow at the institute of Policy Analysis and Research (IPAR)- Rwanda Email: d.malunda@ipar-rwanda.org OR dicksonmalunda@yahoo.co.uk
2
Introduction
There has been a growing concern with “jobless growth” as a major obstacle for the poor to benefit from the positive growth performance experienced by many Low Developed Countries. (LDCs).Given that the poor derive most of their income from work (as employees, as the self-employed, or in subsistence activities), the impact of growth on poverty depends on the extent to which growth generates employment and good earning opportunities.
However, if employment growth is achieved at the expense of wage reductions, it may have little impact on poverty. Given that the poor in low income countries like Rwanda cannot afford to be employed, government policies should be more concerned with raising the income of the working poor, rather than, or in addition to, reducing the unemployment rate. In Rwanda underemployment is a bigger problem compared to unemployment defined as a person who worked at least an hour a day within the week prior to the survey, according to the ILO. This standard ILO definition of unemployment of an unemployed person being one who has not had atleast one hour of work in the last 7 days conceals a great deal of both hidden unemployment and under-employment. Cichello et al characterize Rwanda’s labor market as one of high labor force participation and low unemployment. Their recommendation of improving the quality rather than overall quantity of jobs further reinforces the problem of under-employment in Rwanda (World Bank, 2010).
Another issue for the policy discussion is whether poverty is more effectively reduced by a growth pattern that favors the sectors of the economy in which the poor are found (i.e., agriculture) in order to enhance employment opportunities or by a pattern that disproportionately advances the sectors in which the poor are not found, so that more of the poor can be drawn into the higher earning parts of the economy. If the poor face extensive barriers to gaining access to the higher earning sectors, it could be the case that measures to increase productivity of agriculture where most of the poor are employed, may take policy priority coupled with increasing the number of non-farm jobs in the non-agricultural sectors.
3 In addition, the question of how productivity improvements translate into higher earnings
for the poor is also at the forefront of the debates. The main issues are whether higher growth in output per worker reduce employment growth and whether higher output per worker is associated with better employment opportunities. These are all priority issues in the search for shared growth strategies.
The problem
A number studies have been done to determine the impact of economic growth, public spending and private sector development on poverty reduction and employment creation in Rwanda. Despite the policy recommendations that are have been generated from these isolated studies, there has been a lack of coherent and concrete policy messages with which to advise the government of Rwanda on sustainable policies on how economic growth, public spending and private sector development can be better translated into poverty reduction and employment creation in Rwanda.
Despite the intuitive notion that employment matters for poverty reduction, there is insufficient empirical research in this area in Rwanda. While there is a broad consensus that not all growth spells have the same impact on poverty, there have been relatively few attempts to systematically unpack the relationship between economic growth, employment and poverty reduction in Rwanda. Part of the problem is attributed to the lack of regular up-to-date indicators to assess progress and impact of the implementation Rwanda’s poverty reduction strategy on poverty reduction given the numerous investment programs, projects and initiatives that have been implemented by donors and governments over time. The other part lies in the lack of panel household data on Rwanda. Due to the deficiencies of panel data in Rwanda it is difficult to ascribe causality to correlative relationships between changes in income and employment. This may go some way to explaining the dearth of labour market research in Rwanda to date.
In this paper, we analyze how employment generation and productivity growth have helped determine the effectiveness of growth in reducing poverty in Rwanda. Here, we seek to determine how growth in Rwanda has been reflected in employment generation and in changes in output per worker. In addition, we seek to determine how is growth reflected in the sectoral pattern of growth and employment generation in Rwanda and what the sources of changes in
4 output per worker are in Rwanda.Answers to these issues will help us to understand whether the pattern or profile of growth observed in Rwanda is conducive to poverty reduction, by pinpointing the sector and factors that should be further and factors that should be further analyzed.
5
Country context Economic growth
Rwanda’s economic growth over the last decade has been remarkable. With a government that is committed to achieving sustainable economic growth coupled with growth in employment opportunities for its people, Rwanda has made impressive progress in rehabilitating and stabilizing its economy to pre‐1994 levels. The overall economy is growing at a significant rate. The average annual growth rate in GDP was 8.8 per cent between 2005 and 2009. Rwanda’s GDP per capita has increased from less than 200US$ in 1994 to 540 US$ in 2010. Exports have grown by approximately 12.5% annually since 2001 with coffee, tourism, and tea accounting for 60% of exports.
In the last five years Rwanda has made dramatic development progress. There has been sustained economic growth and the signs of economic transformation noted in 2005 have been confirmed. There has been a continued growth in non-farm employment and a good performance in all economic sectors. GDP grew by an average of 8.8 per cent between 2006 and 2011, 5.5 per cent in agriculture, 10.1 per cent in manufacturing and 10.5 per cent in services. Between 2005/6 and 2010/11 poverty fell by 11.8 percentage points. This compares with a fall of 1.2 per cent between 2000 and 2005/6. Poverty has fallen faster in Rwanda than in most other successful countries in Sub-Saharan Africa including Ghana (11% between 1998/9 and 2005/6), Senegal (8.5% between 2001 and 2006) and nearly as fast as in Uganda (14.3% between 2002/3 and 2009/10) (NISR 2012b).
The strong economic growth (Figure 1) has transferred into benefitting all the population, with those in the lower wealth quintiles benefitting more than those in the higher ones. The Gini Coefficient fell from 52 per cent in 2005/6 to 49 per cent in 2010/11and the ratio of the 90th percentile of consumption to the 10th also fell. Poverty fell significantly in all the provinces, as did inequalities, except in the Northern Province where it remained unchanged. At the district level, however, poverty reduced in only 13 of the 30 districts (NISR 2012b).
6
Figure 1: GDP per Capita 1999- 2010
(Source: GDP National Account 20092; NISR 2011)
Employment Trends in Rwanda (2006-2010/11)
There was an increase in non-farm employment and a consequence reduction on households’ reliance on agriculture for their income. Nevertheless by 2010/11 still over half of all households (55.9) relied on agriculture for their survival (Figure 3). Interestingly there was also a decrease in households dependent on non-agricultural income from 34.3 per cent to 27 per cent, driven by a decline in households dependent on non-farm self-employment, down by 10.8 per cent. However, the latter was more than compensated for by the increase in households with diversified incomes, up from 3.3 per cent to 15.2 per cent - suggesting that there has been an increase in opportunities for farm and non-farm waged employment and in the number of households that are able to benefit from having income from more than one source.
Figure 3: Main Household Activity 2005/6 and 2011/12
242 225 212 206 220 242 289 333 391 480 520 540 1999 2000 2001 2002 2003 2004 2005 2006 2008 2008 2009 2010 56.6 4.3 7.3 27 1.5 1 2.3 52.2 3.6 10.7 16.2 2.2 4.1 11.1
Agricutural Farm Wage Non Farm Wage Non-Farm Self Employment
Transfers Diversified but Farm Wage 30%
+
Diversified but Farm Wage Less
30% 2005/6 2010/11
7 (Source: NISR 2012b)
8
Poverty , Inequality and Employment in Rwanda
Current poverty in Rwanda is estimated to be 44.9% nationally, with 22.1% poor in urban areas and 48.7% poor in rural areas. The current poverty rate of 44.9% shows a reduction in poverty a by 12 percentage points between 2005/06 and 2010/11 which contrast the limited poverty reduction experienced over the period 2000/01 to 2005/06, partly due to an increase in inequality. In both 2006 and 2011, poverty levels were highest by far among those reliant mainly or heavily on farm wage labour, followed by those working in agriculture.
Poverty fell in almost all categories, but particularly among those reliant on non-farm wage or self-employment work, or transfers. Poverty falls to a lesser extent among those reliant on agriculture or farm wage work. This suggests that non-farm activities, especially for wages, could have played an important role in poverty reduction in Rwanda.
Table 1: Poverty status by job type
Poverty status 2005/6 Poverty status 2010/11
Usual work status
Extremely poor
Poor Non-poor Total Extremely poor
Poor Non-poor Total
Wage farm 13.3% 7.2% 5.1% 8.2% 17.9% 11.8% 6.4% 9.9% Wage non-farm 4.4% 5.3% 17.9% 10.9% 9.1% 9.3% 22.2% 16.9% Small-scale farmer 77.0% 80.0 % 63.5% 71.3% 67.0% 71.4% 56.7% 61.8% Independent non-farm 4.5% 6.8% 11.1% 8.1% 4.8% 6.5% 12.5% 9.7%
Other and n.i .8% .7% 2.5% 1.6% 1.3% .9% 2.3% 1.8%
Total 100.0% 100.0
%
100.0% 100.0% 100.0% 100.0% 100.0% 100.0%
According to table 1, the poorest people are likely to be in waged agricultural work as their main job, while the better off are likely to be working in paid non-farm jobs or as self-employed in non-agricultural businesses. Small-scale farmers and their family workers are slightly more likely to be poor, while persons in paid work in the public sector are more likely to be in the richest quintile. An analysis of people’s main work status by consumption quintile further
9 supports the finding above . 19% of adults living in the poorest quintile worked as waged agricultural workers as compared with just 4% of adults living in the 20% of households who consume the most (quintile 5). This is more pronounced in those working in non-farm waged work, with just 9% of those in the poorest quintile having paid work compared with 38% of those living in the richest households
Income Diversification
Most working people in Rwanda have more than one job; in fact, just 37% have one job, 42% have two jobs and the remaining 21% have three or more jobs. Iit appears that multiple job working is now more common than it was five years ago. Urban dwellers are more likely than rural workers to have just one job.
Number of jobs per person
Urban Rural Total
1 57.60% 33.80% 37.30%
2 29.80% 43.90% 41.90%
3 or more 12.60% 22.30% 20.90%
All 100.0% 100.0% 100.1%
Changes in Occupational Groups (2006-2011)
Agricultural occupations dominate the workforce. However, there has also been a move away from agricultural occupations to those in the professions, commerce and sales and in semi-skilled occupations including driving and machine operators. This move out of agriculture has affected both sexes; however, men have been able to do so more effectively than women, with 9% fewer men working in agriculture and 4% of women doing so. 82% of women currently work in agriculture compared with 61% of men.
10 Table 3; Occupational changes 2005/6 to 2010/11
EICV2 EICV3
Job Type Male Female Male Female
Professionals 2.60% 1.50% 3.50% 2.00%
Senior officials and managers 0.10% 0.00% 0.10% 0.10%
Office clerks 0.60% 0.50% 1.00% 0.80%
Commercial and sales 6.50% 5.40% 7.70% 7.40% Skilled service sector 7.20% 4.10% 7.00% 4.00% Agricultural and fishery workers 71.20% 86.30% 61.30% 81.90% Semi-skilled operatives 8.60% 1.90% 12.90% 2.80% Drivers and machine operators 1.20% 0.00% 5.20% 0.30% Unskilled labourers 2.00% 0.20% 0.60% 0.10%
Gender Gap in Employment in Rwanda
The gap in access to non-farm employment between men and women has widened between 2005/6 and 2010/11 as it had done between 2000 and 2005/6. In 2005/6, 13.7 per cent of women were employed in remunerated non-agricultural work (waged and independent), compared with 28.8 per cent of men, three quarters of the additional paid non-farm jobs created between 2001 and 2006 were taken by men, and men were responsible for 60 per cent of small business start-ups (Strode et al, P10). In 2011/12 38.7 per cent of male workers were in non-farm jobs compared with 18.1 per cent of female workers, an increase of 9.9 percentage points for men compared with 4.4 percentage points for women (NISR 2012b) (Figure 9).
11 Figure9: Trends in Non-Farm Employment for Men and Women 2000 to 2010/11
(Source: NISR 2006; NISR2012b)
Research Issues
Analyzing the links between growth, employment, and productivity in Rwanda
In this paper we address dynamic diagnostics to analyze the links between economic growth and labor market changes in Rwanda. The aim is to better understand to what extent and in what way labor markets channel the benefits of economic growth through jobs and earnings i.e. productivity. In a dynamic setting, mere job creation will not be sufficient to reduce poverty. The creation of jobs with higher quality in the form of higher earnings are also needed. It is widely acknowledged that economic growth is a main driving force behind poverty reduction over the long run, and that labor markets provide the main transmission channel for this process. Higher earnings are likely to be correlated with labor productivity, at least over the longer term.
Research questions
The paper will therefore addresses the following specific questions.
i) Is growth in per capita value added in Rwanda due to demographic changes or changes in the level of growth over time? If growth per capita has increased, is this linked to slower population growth or higher output growth?
ii) Is growth correlated with increases in the quantity of jobs (job creation) or in the quality of jobs (increased productivity of existing jobs)?
7.6 13.7 18.1 16.5 28.8 38.7 2000 2005/6 2010/11 Women Men
12 iii) Are the changes in output per worker due to changes within sectors, or due to shifts of workers from low productivity to higher productivity sectors, i.e. changing employment structure in Rwanda?
iv) What are the sources of any increases in output per worker within sectors in Rwanda? Are they related to increases in Total Factor Productivity (TFP) due to innovation (better use of existing resources)? Or are they due to increases in the ratio of capital to labor in firms(adding more resources)?
Methodology
The methodology uses Shapley decompositions, which is a simple additive method that links changes in a particular component to changes in total per capita GDP, by taking into account the relative size of the sector or component, as well as the magnitude of the change. The methodology decomposes GDP growth using several consecutive steps, each step goes
further in answering the above questions.
In a first step growth in per capita GDP (proxied by per capita Value Added) is decomposed into employment rate changes, changes in output per worker and demographic changes2. In the second step employment changes are further decomposed into changes in employment by sectors. The third step decomposes changes in output per worker into changes linked to variations in output per worker within sectors and changes linked to sectoral relocation of workers between sectors. A fourth step goes further in understanding the role played by each sector on the aggregate effect of employment relocation across sectors while the fifth step looks at the role of capital and TFP as sources of changes in output per worker at the aggregate level. A sixth step puts all the elements together, to see how each factor affected total per capita growth.
13 The Model
1. Understanding the aggregate employment and productivity profile of growth
To understand how growth has translated into increases in productivity and employment at the aggregate level and by sectors (or regions), per capita GDP, Y/N=y can be expressed as:
...(1)
OR
where Yi is total Value Added, E is total employment, A is the total population of working age and N is total population. In this way Y/E=ω is total output per worker, E/A is the share of
14 working age population (i.e. the labor force) employed and A/N is the labor force as a fraction of total population. We will refer to e, as the employment rate.
Per capita GDP growth can be decomposed into growth associated with changes in output per worker, growth associated with changes in employment rates and growth associated with changes in the size of the working age population. The methodology uses Shapley decompositions which have the advantage of being additive. This means that the total change in per capita GDP will be the sum of the growth attributed to each of its components ω, e, and a. Thus if we let ω , e and a denote the fraction of growth linked to each component then the growth rate of an economy can be expressed as:
ω *Δy will reflect the amount of growth that would be consistent with a scenario in which the employment rate e, had changed as observed but output per worker and the share of population of working age a had remained constant. There are in fact several ways in which these two components can stay unchanged. They can both remain in the level observed in the initial year, they can both stay at the level observed in the final year, or one of them can stay in the level observed in the initial year and the other stay in the level observed in the final year. Some decompositions only consider the case where both components stay in the level observed in the initial or final year, and thus end up with a residual. What the Shapley decomposition does is that it considers all possible alternatives, and then makes a weighted average of each ,eliminating the residual which can be very large. Each component thus has the interpretation of a counterfactual scenario.
15 In the same way e *Δy will be the amount of growth consistent with a scenario in which
output per worker ω, and the share of population of working age a, had remained ‘unchanged’. The amount of per capita growth linked to demographic changes will be
a *Δy .
2. Understanding the role of each sector in employment generation
To understand the way in which sectors contributed to employment generation and to total per capita growth we further decompose employment (rate) growth (Δe ) by sectors.
Here, total growth in employment is expressed as the sum of employment generation in each sector.
where is the change in employment in sector i as a share of total working age population. This gives a measure of which sector contributed more to changes in the employment rate. Once we understand the sources of growth in the employment rate, we can understand the link of employment growth in sector i, to the observed change in per capita output by combining the results on per capita value added and employment decomposition by sector . The total contribution of sector i, is its contribution to changes in total employment times the contribution of employment rate changes to total growth calculated in step 1. This is interpreted as the per capita growth consistent with a counterfactual scenario, in which all else (productivity, demographics, and employment in the remaining sectors) had all remained unchanged, and the only change had been the observed employment growth in sector i.
16
Data sources
To perform the decomposition we use data on Rwanda's GDP output for two periods i.e. 2006 and 2011. This data is obtained from Rwanda National Accounts from the National Institute of statistics Rwanda (NISR). We measure growth by value added in order to perform sectoral decompositions 3
We use data on employment per sector from the 2005/6 and 2010/11 Rwanda national household surveys (i.e. EICV2 and EICV3) . We also ensure that that employment numbers refer to all the economy and not just to the formal sector or the urban economy. Data on population by ages comes from population census and population projections. Since we want to decompose real growth, data on output is expressed in constant 2006 Rwandan Francs.
3
National Accounts disaggregated by sectors refers to Value Added rather than GDP, so using Value Added rather than GDP will make sure that sectoral disaggregations are consistent with the aggregate ones
17
RESULTS AND FINDINGS
Table 1 shows the change in Value Added per capita and its components. Rwanda registered a growth rate of 28% in per capital Value Added, for the 2006 to 2011 period. This is equivalent to an average per capita value added growth of 5.7% per annum. This growth was accompanied by increases in output per worker (productivity of 29.2%), increases in the population of working age (15%) and a small increase in employment rates (0.22%). In addition, the share of the working age saw a slight decrease of 0.38% due to increased in school enrollment of young adults over the period.
Table 1: Employment, Output, Productivity and Population. Rwanda 2006-2011
2006 2011 % change
GDP (value added) (in 1000000000's) 1,548 2,306 49.0
Total population 9,441,406 10,942,950 15.9
Total population of working age 5,115,571 5,887,514 15.1
Total number of employed 4,301,000 4,961,000 15.3
GDP (value added) per capita 163,959 210,729 28.53
Output per worker 359,916 464,826 29.15
Employment rate 84.08 84.26 0.22
Share of population of working age 54.18 53.80 -0.38
Table 2 shows the results for the Shapley decomposition of per capita growth into its main components at the aggregate level. It shows that a meagre 0.8% of the change in per capita value added can be linked to changes in employment implying that the observed growth in Rwanda over the 2006-2011 period was not followed by sufficient job creation. Changes in the structure of the population had a negative impact on per capita value added over the 2006-2011 period in Rwanda implying that there more minors and elderly depending on each working adult. Given that the dependency ratio has a negative effect on per capita growth, reproductive health measures to curb population growth in Rwanda are still key to Rwanda's development. Although the decrease of 2.8% in the share of working age population is small, it has a disproportionately larger effect on total growth when compared the 0.9 increase in employment rate. This implies that changes in the share of the working population cause bigger changes to productivity when compared to a proportionate changes in employment rate. This again highlights the importance
18 of population control measures to reduce dependency ratios so as to increase the share of the working age population.
Table 2: Decomposition of Growth in per capita Value Added, Rwanda 2006-2011
2006 Rwf
Percent of total change in per capita value added
growth Total Growth in per capita GDP (value added) 46,770.65 100
Growth linked to output per worker 47,676.06 101.94
Growth linked to changes employment rate 415.03 0.89
Growth linked to changes in the share of population of working Age -1,320.44 -2.82
According to table 2 and figure 1, output per worker (productivity) was the dominant driver of the per capita value added over the 2006-2011 period and accounts for an increase of 47,676 2006 RWF or( 102% of the observed change.) However, compared to period 2000 to 2006 , there has been a 20% decline in the contribution of worker productivity on per capita value added in the 2006 to 2011 period.(World bank ,2009). Given this dominant productivity component, it is important to understand what has driven the increased productivity over the 2006-2011. There are a number of reasons why output per worker might have increased. It could be due to increased productivity in a particular sector or due to movements to higher productivity sectors.
19 After decomposing aggregate employment growth we now go further and decompose changes in the overall employment rate e, to understand the role played by different sectors . In addition, we decompose total output per worker ω, in order to understand the role of capital, Total Factor Productivity and inter-sectoral employment shifts. This step is undertaken both at the aggregate level and by sectors.
Employment Generation and Productivity by Sector over time.
According to table 3,total employment in Rwanda grew by 15.4%, but as a result of the simultaneous growth in the working age population, the employment rate grew by only 0.22%.Although agriculture is still the dominant source of employment with over 60% of the working age population engaged in it, its contribution to employment has grown by a small margin of 6% between 2006 and 2011. However, the small employment growth margin in
47676.1 415.0
-1320.4
-10000.0 0.0 10000.0 20000.0 30000.0 40000.0 50000.0 60000.0
Contribution to Growth in Value Added per capita, in 1000000000's of 2006 Rwf Figure 1: Aggregate Employment, Productivity, and Demographic Profile
of GrowthRwanda 2006-2011
Output per worker Y/E Employment rate E/A Demographic change A/N
20 agriculture has been affected by population growth. The ratio of persons employed in agriculture to the total population has declined by 8% implying that although the absolute numbers of workers in agriculture increeased marginally, the proportion of people working in agriculture is reducing over time. This shift from agriculture is a good development given Rwanda Vision 2020 objective of becoming a middle income service based economy. The challenge that comes with this shift is how to absorb the excess labour in agriculture coupled with about 400,000 youths who are new entrants coming onto the labour market annually.
Mining and construction are the sectors that have registered rapid growth in employment in both absolute numbers and as a proportion of the working age population.The employment to working age population in the mining and construction industries have increased by of 130 and 92 percent respectively despite starting from low bases where they accounted for 0.43 and 1.3% of total employment in 2006. Although the proportion of the working age population employed in the manufacturing sector has increased by 22% over the 2006-2011 period , it still lags behind comparable sectors like commerce, transport and services all of which show increases of over 30% in proportions employed by sector.(See table 3)
Table 3: Employment by Sectors of Economic Activity, Rwanda 2006-2011
Total employment Employment/pop. of working age
2006 2011 % change 2006 2011 % change
Agriculture 3,389,000 3,596,000 6.11 66.25 61.08 -7.80
Mining & Utilities 22,000 58,000 163.64 0.43 0.99 129.07
Manufacturing 80,000 112,000 40.00 1.56 1.90 21.64 Construction 66,000 146,000 121.21 1.29 2.48 92.21 Commerce 296,000 444,000 50.00 5.79 7.54 30.33 Transport 56,000 91,000 62.50 1.09 1.55 41.19 Govt Services 141,000 211,000 49.65 2.76 3.58 30.02 Finance Services 13,000 20,000 53.85 0.25 0.34 33.67 Other 238,000 283,000 18.91 4.65 4.81 3.32 Total 4,301,000 4,961,000 15.35 84.08 84.26 0.22
21
Analyzing the contribution of employment changes to overall employment rate in Rwanda between 2006 and 2011.
Table 4a andfigure 3a below show the results of the decomposition of the 0.19 percentage points of growth in employment rate was distributed among the different sectors in Rwanda. Agriculture had the largest negative contribution of 5.17 percentage points to the change in total employment rate in Rwanda over the 2006-2011 period. This decline was compensated for by positive contributions from commerce, construction ,government services, mining and transport which contributed 1.8%,1.2%, 0.8%, 0.6% and 0.5% to the overall change in Rwanda's employment rate between 2006and 2011.
Note that although the share of working age employed in manufacturing only grew 22% compared to the growth in mining and utilities of 130%, the contribution of manufacturing to changes in employment rate is only slightly less than that of Mining and utilities. This is explained by the fact that manufacturing is more than twice the size of Mining and utilities. The decomposition highlights how, small percent changes in a sector or component, can have big impacts if its relative size is large.
0 1,000,000 2,000,000 3,000,000 4,000,000
Agriculture Mining & Utilities Manufacturing Construction Commerce Transport Govt Services Finance Services Other
nu m be r of w ork ers
Figure 2a: Employment by Sectors Rwanda 2006-2011
22 Table 4a: Contribution of employment changes
to overall change in employment rate, Rwanda 2006-2011
Contribution to change in total employment rate
(percent points)
Percent contribution of the sector to total employment
rate growth
Agriculture -5.17 -2773.3
Mining & Utilities 0.56 297.7
Manufacturing 0.34 181.6 Construction 1.19 638.1 Commerce 1.76 941.4 Transport 0.45 241.9 Govt Services 0.83 443.9 Finance Services 0.09 45.9 Other 0.15 82.8
Total employment rate 0.19 100.0
Monetary values are 2006 Rwf
Agriculture
Mining & Utilities Manufacturing Construction Commerce Transport Govt Services Finance Services Other -6.00 -5.00 -4.00 -3.00 -2.00 -1.00 0.00 1.00 2.00 3.00 Percent points
Figure 3a: Contribution of each Sector to Change in Employment-to-Population Ratio Rwanda 2006-2011
23 Tables 4b and Figure 3b, show the contribution of sectoral employment changes to growth in total per capita output. The reduction in the proportion working age adults to the total population in agriculture had a large negative contribution on per capita GDP growth of Rwf 11,509 . Despite the contraction of the agriculture sector, other sectors expanded leading to positive contributions to GDP growth. The commerce, construction and the government services sectors had the highest increases of 3907, 2448 and 1842 2006RWF of per capita valued added respectively.
The contribution of the manufacturing sector is interpreted as the growth which would have resulted in the counterfactual scenario in which the share of working age population, total output per worker, and employment in all sectors other than manufacturing had remained unchanged, but employment in manufacturing had grown as observed. If this had been the case total per capita output would have increased by 753.5 RWF of 2006.
Table 4b: Contribution of employment changes to overall change in per capita GDP (value added), Rwanda 2006-2011
Contribution to change in per capita GDP
(value added)
Percent of total change in per capita GDP
(value added)
Agriculture -11509.8 -24.6
Mining & Utilities 1235.7 2.6
Manufacturing 753.5 1.6 Construction 2648.3 5.7 Commerce 3907.2 8.4 Transport 1003.9 2.1 Government Services 1842.3 3.9 Finance Services 190.5 0.4 Other 343.5 0.7 Total contribution 415.0 0.9
24
Agriculture
Mining & Utilities Manufacturing Construction Commerce Transport Govt Services Finance Services Other -14000.0 -12000.0 -10000.0 -8000.0 -6000.0 -4000.0 -2000.0 0.0 2000.0 4000.0 6000.0 1000000000's of 2006 Rwf
Figure 3b: Contribution of Change in Employment-to-Population Ratio to Change in GDP (value added) per capita, by Sector
25
LABOUR PRODUCTIVITY
Analyzing Labour Productivity changes by Sector over the 2006-2011 period
Table 5 and figure 2 below show the changes in sector productivity in Rwanda over the 2006-2011 period. Overall productivity in Rwanda grew by 29% between 2006 and 2006-2011.Of all the sectors, agriculture, commerce and government services accounted for the largest contribution to the observed productivity growth with of 22, 14 and 13 per cent growth in productivity respectively. Given that land is constraint and yet not many people are shifting from agriculture to other sectors, it is plausible to say that the large increases in Rwanda's agricultural productivity are attributed the land consolidation and the Crop Intensification Programs(CIP) which entail fertiliser distribution to rural farmers. The CIP is program which has been implemented by government in order to transform agriculture under Rwanda's Plan for Structural Transformation Agriculture program(PSTA). Productivity increases in the government services are attributed to the reforms that have been implemented in the public sector. These included streamlining employment within the public service and capacity building efforts over the 2006-2011 period.
Table 5: Changes in Output per Worker by Sectors. Rwanda 2006-2011
2006 2011 % change
Agriculture 194,748 237,208 21.80
Mining & Utilities 636,364 344,828 -45.81
Manufacturing 612,500 544,643 -11.08 Construction 1,590,909 1,445,205 -9.16 Commerce 648,649 738,739 13.89 Transport 2,089,286 2,285,714 9.40 Government Services 1,319,149 1,488,152 12.81 Finance Services 3,769,231 4,000,000 6.12 Other 739,496 816,254 10.38
Total output per worker 359,916 464,826 29.15
Monetary values are 2006 Rwf
Despite showing considerable increase in employment numbers over the 2006-2011 period, productivity in terms of output per worker has declined in the mining, and construction sectors by 46% and 9% respectively. The reduction in the mining sector could have been due to the halting of mining of some minerals in Rwanda pending environmental impact assessments.
26 Manufacturing is sector of main concern because it has the potential to employ large numbers of people and yet it shows decreases in both numbers employed and in productivity as well. Efforts should be made to revamp Rwanda's manufacturing industries in order to increase employment and productivity in the sector.
Decomposition of Labour productivity into Within Sector Changes in Output per Worker and Inter-sectoral Shifts. Rwanda 2006-2011
Table 6a and figure 4a below show the results of decomposing output per worker using Rwanda data. The first column of the table shows the contribution of each sector as well as of inter-sectoral employment shifts to the observed growth in total output per worker. The 104,909 Rwf increase in productivity per worker for the period is accounted for by a 2449Rwf reduction in mining and utilities, a 1397 RWF reduction from manufacturing, a 3262RWf reduction construction and an increase of Rwf 32,117 in Agriculture , an increase of Rwf 6364 in government services and an increase of Rwf 4313 in other sectors, and an positive effect of sectoral labor relocation of Rwf 58,198. Generally productivity increases due to inter-sectoral shifts and agriculture had the largest contributions of 55.5% and 31% to the labour productivity growth in Rwanda over the 2006-2011 period. The large growth the agriculture
0 500,000 1,000,000 1,500,000 2,000,000 2,500,000 3,000,000 3,500,000 4,000,000 4,500,000
Agriculture Mining & Utilities Manufacturing Construction Commerce Transport Govt Services Finance Services Other
1000000000' s of 2006 Rw f
Figure 2b: Output per Worker by Sectors, Rwanda 2006-2011
2006 2011
27 sector productivity is attributed the implementation of the Crop Intensification Program (CIP) which resulted into increase land consolidation, increased fertilizer use and commercialization among Rwandan farmers
The fact that inter-sectoral shifts had the largest positive contribution means that on average labor movements from lower than average productivity sectors to above average productivity sectors contributed the largest gains to labour productivity over the 2006-2011 period.
As we saw on page 18, the rest of the sectors increased their employment shares except agriculture whose share of working age population contracted over the 2006-2011 period. Thus we can conclude that an important share of growth in output per worker was due to movements of the labor force from agriculture into other sectors of the economy.
Table 6a: Decomposition of Output per Worker into Within Sector Changes in Output per Worker and Inter-sectoral Shifts. Rwanda 2006-2011
Contribution to Change in Total Output per Worker Contribution to Change in Total Output per Worker (%)
Agriculture 32,117.2 30.6
Mining & Utilities -2,449.8 -2.3
Manufacturing -1,397.1 -1.3 Construction -3,261.9 -3.1 Commerce 7,131.5 6.8 Transport 3,080.3 2.9 Govt Services 6,364.2 6.1 Finance Services 813.9 0.8 Other 4,313.1 4.1 Inter-sectoral shift 58,197.9 55.5
Total change in output per worker 104,909.3 100.0
28
Understanding the sources of changes in total output per worker (net of inter-sectoral shifts) at the aggregate level
Aggregate and sectoral changes in output per worker (Δω), capture changes in output per worker, but its interpretation is not so straight forward. Increases in output per worker can come from three different sources: i) increases in capital labor ratio ii) increases in Total Factor Productivity (TFP) and iii) relocation of jobs between bad jobs sectors (low productivity) to good jobs sector (high productivity), which is the between-component of growth in output per worker or the labor relocation effects .
In this section, we decompose within productivity changes (or productivity changes net of inter-sectoralshifts) into changes due to increases in the capital-labor ratio and the residual, which can be interpreted cautiously, as Total Factor Productivity (TFP) growth.
Findings
Figure 5 show the decomposition of changes in aggregate output per worker for Rwanda between 2006-2011. It illustrates the role of capital and TFP, as well as the inter-sectoral shifts .Total output per worker increased by 29% . Of this increase inter-sectoral employment shifts exerted a positive effect on output per worker contributing with 58,198 Rwf of total productivity growth. The capital labor ratio also increased, contributing an additional 63,383Rwf. Although TFP
-10,000.0 0.0 10,000.0 20,000.0 30,000.0 40,000.0 50,000.0 60,000.0 70,000.0
Agriculture Mining & Utilities Manufacturing Construction Commerce Transport Govt Services Finance Services Other Inter-sectoral shift 1000000000's of 2006 Rwf
Figure 4a: Decomposition of Growth in Output per Worker: Inter-Sectoral Shifts and Within Sectoral Output Growth
29 suffered a reduction of Rwf 16671.5, this decrease was more than offset by the increases in output per worker and capital labour ratio.
Table 7: Data used for Decomposition of Output per Worker, Capital Stocks, Capital Labor Ratio and Share of Capital in Total Income. Rwanda 2006-2011
2006 2011 % change
Share of Capital in Total Income (%) 4% 6% 0.00
Capital 71 140 97.18
Total output per worker 359,91
6
464,82
6 29.15
Output per worker net of inter-sectoral shifts 359,91 6
406,62
8 12.98
Capital Labor Ratio 16,508 28,220 70.95
TFP residual net of inter-sectoral shifts 241,70 0
231,42
8 -4.25
Monetary values are 2006 Rwf
58197.9 -16671.5
63383.0
-30000.0 -20000.0 -10000.0 0.0 10000.0 20000.0 30000.0 40000.0 50000.0 60000.0 70000.0 1000000000's of 2006 Rwf
Figure 5: Decomposition of Changes in Output per worker Rwanda 2006-2011
Inter-sectoral shift Total factor productivity Capital labor ratio
30 Understanding the Role of Inter-Sectoral Employment Shifts in Rwanda over 2006-2011.
In the previous section, we decomposed changes in output per worker into changes in output per worker within sectors and changes in output per worker due to movements of labor between sectors with different levels of productivity. In this section we present details about contribution of structural change or inter-sectoral shift component to per capita growth or output per worker in Rwanda. A number of studies have found that structural change, which is movements of labor force shares from low productivity sectors to high productivity sectors, is an important factor behind growth. Increases in the share of employment in sectors with above average productivity will increase overall productivity and contribute positively to the inter-sectoral shift term. On the contrary, movements out of sectors with above average productivity will have the opposite effect. By the same token, increases in the share employment in sectors with below average productivity should reduce growth, while reductions in their share should contribute positively to growth.
Thus if sector i has productivity below the average productivity, and decreases its share si, its contribution will be positive, that is outflows from this low productivity sector have contributed to the increase in output per worker. If on the other hand, the sector sees an increase in its share, these inflows into this low productivity sector will decrease output per worker and thus have a negative effect on the inter-sectoral shift term. The magnitude of the effect will be proportional to: i) the difference in the sector’s productivity with respect to the average and ii) the magnitude of the employment shift.(World Bank, 2012)
Findings on structural shifts in Rwanda
The first column of table 8 below shows average output per worker between 2006 and 2011, with the last row of the column showing average output per worker for the whole economy. The second column shows the change in employment shares in each sector. The final column shows the contribution of each sector to the Rwf 58,198 contributed by inter-sectoral employment shifts to total growth in output per worker.
Movements out of agriculture (a lower than average productivity sector) had a positive effect on productivity and overall growth in Rwanda over the 2006-2011 period. In addition labour
31 movements into high productivity sectors with above average output per worker (like construction, transport and services) had relatively high contributions highly to the between component of productivity changes, thus raising aggregate productivity. On the contrary shifts or movements into below average productivity sectors like mining and utilities had lower positive contributions to aggregate productivity.
Table 8: Understanding the Role of Inter-Sectoral Employment Shifts. Rwanda 2006-2011 Average Output per Worker Change in employment share (percent points) Sectoral contribution to inter-sectoral shift component Agriculture 215,978 -0.063 12392.88
Mining & Utilities 490,596 0.007 514.41
Manufacturing 578,571 0.004 660.77 Construction 1,518,057 0.014 15572.80 Commerce 693,694 0.021 5816.88 Transport 2,187,500 0.005 9448.74 Govt Services 1,403,650 0.010 9663.66 Finance Services 3,884,615 0.001 3503.12 Other 777,875 0.002 624.64 Aggregate 412,371 58197.91
Monetary values are 2006 Rwf
Table 9 shows the percent contribution of each term to the inter-sectoral shift or between
components of productivity changes. Labour movements out of agriculture and labour movements into construction, transport and government services have had an important role in contributing to the growth in Rwanda over the 2006-2011 period. The most outstanding issue is the low contribution of manufacturing(i.e. 1%) to the inter-sectoral shifts despite the fact that manufacturing has a high potential to productively employ a substantial proportion of the 125,000 unemployed youths who enter Rwanda's labour markets annually.
32
Table 9: Decomposition of Inter-sectoral Shifts. Rwanda 2006-2011 Direction of Employment Share shift Contribution to Inter-sectoral Shifts (%) Sectoral contributions Agriculture - 21.29
Mining & Utilities + 0.88
Manufacturing + 1.14 Construction + 26.76 Commerce + 9.99 Transport + 16.24 Govt Services + 16.60 Finance Services + 6.02 Other + 1.07
Total Contribution of inter-sectoral shifts 100
Summary of the growth decomposition in Rwanda (2006-2011)
In this section , we determine the percent contribution of each factor to total changes in GDP per capita.
Table 10a and Table 10b below illustrates the results for Rwanda, in Percentage contribution and in Rwf of 2006, respectively. The demographic component accounts for a small negative proportion of 2.8% of all the change. The other 102.8% increase is explained by an increase in output per worker within sectors (45%), a small increase in the share of working age population employed (0.9%), and a large positive effect of labor relocation (56.5%), which on average moved from low productivity sectors to high productivity sectors.
When looking across sectors the biggest contribution to output per capita was made in the sectors of commerce, government services agriculture and construction at 20.9%, 19.5% 18.6% and 17.6% respectively. For agriculture, the effect was mostly due to increases in output per worker. On the contrary, construction had a large part of its effect through labour re-allocation or inter-sectoral shifts.
33 Sectoral contributions are decomposed into: i) contribution of within changes in output per worker (first column), ii) contribution of changes in employment (second column), and iii) contributions of the sector to the inter-sectoral employment shifts. The final column is the
total effect of the sector. Overall all sectors contributed positively to growth with ,commerce, construction, agriculture and government services contributing the most. Manufacturing and mining and utilities, had a relatively low contribution to Rwanda growth implying the urgent need for improvements in energy to spur the manufacturing sector in Rwanda.
Despite the decrease in employment growth in agriculture, the increase in output per
worker was so large that it more than offset the reduction in the share of employment in agriculture.
Table 10a: Growth Decomposition. Percent Contribution to Total Growth in GDP (value added) per capita, Rwanda 2006-2011 Contribution of within sector changes in output per worker (%) Contribution of changes in Employment (%) Contributions of Inter-sectoral Shifts (%) Total (%) Sectoral contributions Agriculture 31.21 -24.61 12.04 18.64
Mining & Utilities -2.38 2.64 0.50 0.76
Manufacturing -1.36 1.61 0.64 0.90 Construction -3.17 5.66 15.13 17.62 Commerce 6.93 8.35 5.65 20.94 Transport 2.99 2.15 9.18 14.32 Govt Services 6.18 3.94 9.39 19.51 Finance Services 0.79 0.41 3.40 4.60 Other 4.19 0.73 0.61 5.53 Subtotals 45.39 0.89 56.55 102.82 Demographic component - - -2.82 Total 100.00
34 Table 10b: Growth Decomposition. Contribution to Total Growth in GDP (value added) per capita, Rwanda
2006-2011 Contribution of within sector changes in output per worker Contribution of changes in Employment Contributions of Inter-sectoral Shifts Total Sectoral contributions Agriculture 14595.66 -11509.84 5631.95 8717.76
Mining & Utilities -1113.32 1235.68 233.78 356.13
Manufacturing -634.89 753.50 300.29 418.90 Construction -1482.38 2648.32 7077.06 8243.00 Commerce 3240.91 3907.17 2643.48 9791.56 Transport 1399.85 1003.87 4293.98 6697.71 Govt Services 2892.21 1842.28 4391.65 9126.14 Finance Services 369.89 190.50 1591.99 2152.39 Other 1960.09 343.54 283.87 2587.50 Subtotals 21228.02 415.03 26448.04 48091.09 Demographic component - - -1320.44
Total change in value added per capita 46770.65
Monetary values are 2006
Rwf
CONCLUSIONS
The pattern of growth in Rwanda has potential effects on poverty reduction over time.
Identifying which sectors and factors are most linked to per capita GDP growth is a first step in identifying where we should look into to understand poverty reduction.
The results from Rwanda suggest that there are three important drivers in
poverty changes. First, inter-sectoral shifts in which workers are moving from low productivity sectors like agriculture to higher productivity non-agricultural sectors the inter-sectoral shifts are generating a window of opportunity to raise per capita income and thus reduce poverty: The incraesed productivity in agriculture accompanying the governments crop intensification program coupled with labour movements out of agriculture are improving productivity and driving growth.
Second, one of the main risks reducing Rwanda's progress is population growth. Changes in the structure of the population had a negative impact on per capita value added over the 2006-2011 period in Rwanda implying that there more minors and elderly depending on each working
35 adult. Given that the dependency ratio has a negative effect on per capita growth, reproductive health measures to curb population growth in Rwanda are still key to Rwanda's development.
Third, the manufacturing sector seems to have lagged behind other sectors in tersm of output and productivity growth over the last five years. Although the share of employment in agriculture as part of the working age population has gone down over the last 5 years, agriculture has become more productive sector contributing an annual average of about 4 per cent of the productivity growth that has been achieved, thanks to the governments Crop Intensification Program(CIP). Sectors like mining, commerce and services have shown improvements in both the proportions of adults they employ and productivity in terms of value added per capita.
However, despite the above impressive achievements, development of the manufacturing sector seems to have lagged a little behind. Although the proportion of working age population employed in the manufacturing sector has increased by an average of 4% per year, the increase is smaller when compared to other sectors like construction, commerce and transport whose increases are well above 6%. In addition the productivity of the sector in terms of output per worker has gone down by 2% per annum over the last 5 years .
In addition manufacturing has also displayed a decrease in output per worker. This leads tothe following questions. How wellpaying are these new manufacturing jobs? Do the poor have access to them? What types of jobs are being created? Where is this growth in manufacturing employment come from? Is it sustainable? How can Rwanda prop up its manufacturing sector in order to make the country self-reliant with decent jobs from the manufacturing sector?
Part of the answers lies in promoting and empowering medium to large scale manufacturers, entrepreneurs and industrialists in Rwanda. Rwanda can steer local funds like the ''Agaciro funds'' into some sort of investment vehicles to have revolving funds or loans for local industrialists to invest ,more so in the agro-processing industries in the rural districts in order to put our people to work, while at the same time investing in the utilities like electricity to run the manufacturig sector
36
Bibliography
Ansoms , A.,Mckay.A. A quantitative analysis of poverty and livelihood profiles: the case of rural Rwanda. Food policy (2010)
Ansoms, A., 2005. Resurrection after civil war and genocide: growth, poverty and inequality in post-conflict Rwanda. European Journal of Development Research 17 (3), 495–508.
Ansoms, A., 2007. How successful is the Rwandan PRSP? Recent evolutions of growth, poverty and inequality briefing. Review of African Political Economy 111, 371–379.
Ansoms, A., 2009. Faces of Rural Poverty in Contemporary Rwanda: Linking Livelihood Profiles and Institutional Processes. PhD in Applied Economics. University of Antwerp.
Chris(2009) 'Agricultural Policies and Local Grievances in Rural Rwanda', Peace Review, 21:3, 296 — 30
P. Cichello and A. Sienarert (2009); Making work pay in Rwanda; Employment, growth and Poverty reduction. A World Bank Labor Market Study (The World Bank,)
P. Cahu and C. Sanchez (2010); Economic Growth, Poverty Reduction and Distributional Impact of Fiscal Policy in Rwanda; Background paper to PER (Rwanda PER, 2010)
Education development Center (2009); Rwanda Youth Employment Assessment Report; USAID publication for the Rwanda Youth Opportunity Program (RYOP)
Government of Rwanda (2006); Sources of Informal Economic Activity in Rwanda, FIAS International Finance Cooperation (IFC)
Government of Rwanda (2007). Economic Development and Poverty Reduction Strategy(EDPRS), 2008-2012
Government of Rwanda, 2002. A Profile of Poverty in Rwanda: An Analysis based on the Results of the Household Living Condition Survey 1999–2001. MINECOFIN, Kigali.
Government of Rwanda, 2007. Economic Development and Poverty Reduction Strategy 2008– 2012, Preliminary Draft June 2007. MINECOFIN, Kigali.
Howe, G., McKay, A., 2007. Combining quantitative and qualitative methods in assessing chronic poverty: the case of Rwanda. World Development 35 (2), 197–211.
37 Pamela, Abbot et al 2010, Diagnostic Report for the Study of Household enterprises Rwanda; Report written for the World Bank Study on household enterprises
Louis-Marie Asselin; VUP Targeting and Poverty Surveys 2009: Draft report
Mckay, A., Loveridge, S., 2005. Exploring the Paradox of Rwandan agricultural Household Income and Nutritional Outcomes in 1990 and 2000. Staff paper 2005-06. Department of Agricultural Economics Michigan State University, Michigan.
National Institute of Statistics (2012a) The Evaluation of Poverty in Rwanda from 2000 to 2011: Results from the Household surveys (EICV). Kigali: Republic of Rwanda.
National Institute of Statistics of Rwanda (2012b). The Third Integrated Household Living Conditions Survey (EICV3): Main Indicators Report. Kigali: Republic of Rwanda.
National Institute of Statistics (2011) Rwanda Statistics Year Book 2011. Kigali: Republic of Rwanda.
National Institue of Statistics of Rwanda. (2009). Statistical Yearbook 2009 Edition. Kigali: National Institute of Sstatistics of Rwanda.
National Institue of Statistics of Rwanda. (2007b). Millennium Development Goals: Towards a Sustainable Future, Country Report 2007. Kigali: National Institute of Sstatistics of Rwanda. National Institute of Statistics (2006).Preliminary Poverty Update Report: Integrated Living Conditions Survey 2005/6. Kigali: Republic of Rwanda.
National Institute of Statistics Rwanda (2006b) Labor Market and economic Activity Trends in Rwanda
The World bank 2007: Rwanda Toward Sustained Growth and Competitiveness
P. Ronnås(ILO)Karl Backéus (Sida)* ElinaScheja (Sida ) 2010; Rwanda forging ahead. The challenge of getting everybody on board. ILO working paper No.62
