Download How Solid Is Economic Growth in the East African Community?

WP/14/150
How Solid Is Economic Growth in the East
African Community?
Nikoloz Gigineishvili, Paolo Mauro, Ke Wang
WP/14/150
© 2014 International Monetary Fund
IMF Working Paper
African Department
How Solid Is Economic Growth in the East African Community?
Prepared by Nikoloz Gigineishvili, Paolo Mauro, Ke Wang
August 2014
This Working Paper should not be reported as representing the views of the IMF.
The views expressed in this Working Paper are those of the author(s) and do not necessarily
represent those of the IMF or IMF policy. Working Papers describe research in progress by the
author(s) and are published to elicit comments and to further debate.
Abstract
Is rapid economic growth experienced by the East African Community during the past
decade built on solid foundations? To gain some clues, we use a variety of newly-collected
and existing data sources to analyze the structural transformation of output and exports, as
well as indicators of their quality and sophistication. The move from agriculture to a wide
range of other sectors—bodes well for continued growth, as do gradual improvements in
quality. Yet, no clear winners on the production side seem to have emerged, to embed a
durable comparative advantage in international markets. These observations may instill a
note of caution against projecting rapid growth into the distant future.
JEL Classification Numbers: O11, O14
Keywords: Structural transformation, Burundi, Kenya, Rwanda, Tanzania, Uganda.
Authors’ E-Mail Addresses: [email protected]; [email protected]; [email protected]
2
Content
Page
I. Introduction ..........................................................................................................................3 II. Overall Growth Performance and Reversion to the Mean ..................................................4 III. Structural Transformation, Diversification, and Rising Sophistication ...............................8 IV. Summary and Conclusions ................................................................................................17
References ................................................................................................................................22 Figures
1. Real GDP Growth Rates ......................................................................................................4
2. Average Real Growth Rate ..................................................................................................5
3. Electricity Consumption ......................................................................................................6
4. Health Outcomes ..................................................................................................................6
5. Fiscal Revenues and Private Sector Credit ..........................................................................6
6. Frequency Distribution of Economic Growth Rates ............................................................8
7. Average Output by Sectors, 1970–2010 ..............................................................................9
8. Sector Shares and Changes, 2000 vs. 2010........................................................................10
9. Exports, 1990–2010 ...........................................................................................................11
10..Partner Diversification .......................................................................................................11
11. Structure of Exports ...........................................................................................................12
12. Export Product Diversification ..........................................................................................13
13. Number of Export Products ...............................................................................................13
14. Sophistication Index, 1990 vs. 2010 ..................................................................................14
15. Quality of Top Ten Products (four-digit level) ..................................................................16
Appendix
Methodology and Data .............................................................................................................18 Appendix Figures
A1. Changes in Composition of Output by Sector, Individual Countries ...............................20
A2. Changes in Sectoral Structure of Exports, Individual Countries ......................................21
3
I. INTRODUCTION
The East African Community (EAC) countries’ economic growth performance during the
past decade has been impressive:1 at 6.2 percent, the EAC’s (unweighted) average growth
rate in 2004–13 is in the top one-fifth of the distribution of 10-year growth rate episodes
experienced by all countries worldwide since 1960. Such performance is even more
remarkable taking into account that the past decade encompasses the global economic and
financial crisis that began in 2007. Will this prove to be an isolated episode, with growth
returning to lower levels in the years ahead, or is strong growth going to persist?
Prediction is very difficult, especially if it’s about the future, as physicist Nils Bohr famously
remarked. The economics profession’s ability to forecast economic growth beyond the next
year or so is particularly weak; and it is abysmal at forecasting turning points in the economic
cycle (Loungani and Juhn, 2012). This said, informed guesses about medium term economic
growth are simply indispensable to policymaking. For example, economic growth is perhaps
the most important determinant of whether a country’s fiscal policy is sustainable, and
growth slowdowns have caused many public debt crises in the past (Easterly, 1991). Thus,
while it would be unrealistic to expect to come to firm conclusions regarding the likely
persistence of strong growth in the EAC, it is necessary to attempt to glean some clues from
the nature of the recent growth in order to inform one’s views on future growth in the region.
Previous work (McAuliffe, Saxena, and Yabara, 2012) has gauged the sustainability of the
EAC’s growth “acceleration” using an econometric model applied to total GDP growth for a
panel of countries, finding that prospects for sustained growth in the region are good if
macroeconomic stability is maintained, and further progress is made in deepening the
financial sectors and improving the business climate, infrastructure, and human capital.
The present paper takes a complementary approach by looking at more disaggregated
information on the sources of growth in the past in order to help inform judgments about the
sustainability of growth in the future. In particular, we analyze changes in the composition of
output and exports by economic sector, and various indicators of improvements in the quality
of goods produced and exported by countries in the region. The extent to which
developments in the EAC countries seem to reflect a move toward more modern economic
processes and outputs could be interpreted as an indicator of whether growth is likely to be
longer lasting. While we believe the information reported in this paper provides some clues
as to the future evolution of output in the region, many other factors likely to affect future
growth are not analyzed here. For example, the analysis presented in this paper relies on past
growth and thus does not take into account the implications of possible exploitation of recent
natural resource finds in the region.
1
The East African Community member countries are Burundi, Kenya, Rwanda, Tanzania, and Uganda. For a
general introduction to the EAC and its regional integration process, see Davoodi, ed. (2012).
4
The remainder of the paper is structured as follows. In the next section, to motivate our
analysis, we validate the information from official statistics on GDP by showing that other
indicators corroborate the evidence of rapid economic growth; and we document that
economic growth at a pace experienced in the EAC over the last decade is a relatively rare
event. In Section III, we report our main findings obtained by analyzing a host of data
sources on the detailed composition of output and exports: we document the structural
transformation of the economies, and the improving diversification and quality of their
output and exports. Section IV distills the main patterns identified and concludes.
II. OVERALL GROWTH PERFORMANCE AND REVERSION TO THE MEAN
In view of potential concerns about the quality of the GDP data, we present corroborating
evidence that the growth observed in the EAC during the past decade was indeed strong. We
then calculate the likelihood that such growth would persist, based on past empirical
regularities using overall GDP growth data from a worldwide sample.
Economic growth during the past decade was strong and corroborated by other indicators
During the past decade and indeed since the year 2000, economic growth in the EAC
(average for the five countries) was consistently strong, and considerably above that
experienced in previous decades (Figure 1).
Figure 1. EAC: Real GDP Growth Rates
(in percent)
10
EAC Weighted Average
8
EAC Un-weighted Average
6
4
2
0
Source: World Economic Outlook, International Monetary Fund.
2012
2009
2006
2003
2000
1997
1994
1991
1988
1985
1982
1979
1976
1973
1970
1967
1964
-2
5
Indeed, all five EAC member countries displayed average real growth of 4 percent or higher,
in most cases substantially above that of the previous decade (Figure 2).
Figure 2. EAC Average Real Growth Rate
(in percent)
9
8
7
1994-2003
2004-2013
6
5
4
3
2
1
0
Burundi
Kenya
Rwanda
Tanzania
Uganda
EAC
Source: World Economic Outlook, International Monetary Fund.
Notes: For Burundi the growth series starts in 1997 to exclude the
period of rapid contraction in the first years of the civil war. For
Rwanda, the series starts in 1998 to exclude the genocide-related
contraction in 1994 and the subsequent sharp rebound in 1995-97.
To validate the information from official statistics on output growth, it is helpful to consider
developments in the correlates of production, consumption, and economic development, such
as consumption of electricity, the mortality rate, credit, and fiscal revenues (Easterly, 1999).
On the whole, these corroborate the overall picture of healthy economic growth. For
example, electricity consumption (available as a long time series for Kenya and Tanzania) is
estimated to have grown rapidly over the past decades, with an acceleration during the past
ten years (Figure 3). Health outcomes such as infant mortality rates and life expectancy
improved considerably in most countries in the region (Figure 4). And both credit and
revenues expressed in domestic currency at constant prices rose rapidly (Figure 5).
6
Figure 3. Electricity Consumption
GDP and Electricity, Tanzania
GDP and Electricity, Kenya
600
400
350
500
Index, 1980=100
Index, 1980=100
300
250
200
150
100
400
300
200
100
50
0
0
Gross domestic product, constant prices, 1980=100
Gross domestic product, constant prices, 1980=100
Electric Power Consumption, in billions of kWh, 1980=100
Electric Power Consumption, in billions of kWh, 1980=100
Sources: World Economic Outlook, International Monetary Fund; World Development Indicators, World Bank.
Figure 4. Health Outcomes
infant (per 1,000 live births)
140
Mortality Rate, 1980 vs. 2010
Life Expectancy, 1980 vs. 2010
60
120
55
100
80
50
60
40
45
20
0
40
1980 2010 1980 2010 1980 2010 1980 2010 1980 2010 1980 2010
BDI
KEN
RWA
TZA
UGA
1980 2010 1980 2010 1980 2010 1980 2010 1980 2010 1980 2010
SSA
BDI
KEN
RWA
TZA
UGA
Source: World Development Indicators, World Bank.
Figure 5. EAC: Fiscal Revenues and Private Sector Credit
(Year 2000=100)
600
500
400
Revenue
Credit
300
GDP
200
100
2013
2012
2011
2010
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
0
Sources: Country authorities and IMF staff calculations.
Notes: The chart reports fiscal revenues, credit to the private sector, and GDP, all in
domestic currency in constant prices. Each variable is set to 100 in the year 2000.
SSA
7
Economic growth averaging more than 6 percent over a decade is a rare event
The EAC countries’ economic growth performance over the past decade is a relatively rare
event by historical standards, when considering a large sample of growth episodes worldwide.
Consider the frequency distribution of all 10-year growth rate episodes experienced by all
countries worldwide since 1950: this has a mean of 3.8 percent and a standard deviation of
2.6 percent (Figure 6, blue line). Thus, at 6.2 percent, the (unweighted) average real GDP
growth rate experienced by the EAC countries in 2004–13 is within the top one-fifth percent
of the worldwide distribution of all decade-long growth episodes since 1950.
Based on purely historical statistical patterns, sustaining such strong performance for the next
decade would be a challenge. Indeed, previous studies have shown that the correlation of a
given country’s growth rate in a given decade with the next is low, ranging between 0.1–0.3
(Easterly and others, 1993; Pritchett and Summers, 2013; Ho and Mauro, 2014). In that vein,
it is possible to simulate the distribution of economic growth episodes over the next decade
for a country or region that experienced 6.2 percent growth over the past decade, as follows.
Estimating a regression of decade-long economic growth on the previous decade’s economic
growth for a panel of 162 countries over 1960-2010 (with decade fixed effects to control for
fluctuations in average world growth):2
, ,
,
,
,
where ~
0,
The regression yields = 0.23, = 2.8, and = 2.54.3 Using the estimated coefficients and
simulating the distribution of growth rates over the decade ahead by plugging in the past
decade’s average growth rate for the EAC region,
6.2, one obtains the red line
in Figure 6:
2.8
0.23
4.2
,
where ~
0, 2.54
In that case, an average growth rate of 6.2 percent would be in the top 23 percent of the
distribution. In other words, taking into account the estimated historical patterns in the extent
to which countries’ growth performance correlates across decades, there is a little less than a
one in four chance of another decade of growth above 6 percent for the EAC. In the remainder
of the paper, we look at various aspects of the composition of past growth to glean clues about
whether it is likely to be long lasting.
2
3
Data on real GDP growth from the Penn World Table version 8.0.
The standard deviation is obtained from the RMSE of the above regression.
8
.1
0
.05
Frequency
.15
.2
Figure 6. Frequency Distribution of Economic Growth Rates
-10
-5
0
5
10
15
Decadal average growth, percent per annum
Data 1950-2010
Model, conditional on past growth
Sources: Penn World Tables 8.0 and IMF staff estimates.
Note: The sample excludes Equatorial Guinea (an outlier of small economic size) and
countries with fewer than 25 years of data.
III. STRUCTURAL TRANSFORMATION, DIVERSIFICATION, AND RISING SOPHISTICATION
To analyze the nature of past economic growth performance, in this section, we track the
structural transformation and increasing diversification of output and exports by reporting
developments in their composition by sector; we then show the evolution of various
indicators of the sophistication and quality of output and exports.
Declining trend in agriculture’s output share is reflected in gains by several other sectors
Considering the composition of output since 1970, drawing on United Nations data for seven
broad economic sectors (agriculture; manufacturing; construction; wholesale trade;
transportation; mining and utilities; and other), a trend decline in the share of agriculture is
apparent in the EAC: such share declined from almost one half of output (unweighted
average) in 1970 to about a third in 2010. As in many other instances of structural
transformation, a decline in the share of agriculture implies gains in productivity, other things
equal, given that agriculture tends to have lower value-added than the secondary and tertiary
sectors. The decline in the share of agriculture in the EAC was steeper than in the rest of
Sub-Saharan Africa or in low-income countries worldwide (Figure 7). The most pronounced
gains in shares were observed in transportation and construction, with a more visible
acceleration since the 1990s in the EAC than in comparator groups; the increase in the share
9
of manufacturing is limited, and less pronounced than in other groups; the increase in mining,
a relatively small sector in the EAC, is barely visible.4
Figure 7. EAC: Average Output by Sectors, 1970-2010
(in percent of GDP)
Agriculture1
50
45
40
35
30
25
20
15
10
5
0
Manufacturing
14
13
12
11
10
9
8
7
1970
1980
1990
2000
Mining and utilities
15
1970
2010
1980
1990
2000
2010
2000
2010
2000
2010
Construction
10
8
10
6
5
4
0
2
1970
1980
1990
2000
2010
Wholesale2
20
1970
1980
1990
Transport3
10
18
8
16
14
6
12
4
10
8
2
1970
1980
1990
2000
2010
1970
1980
1990
Other
35
30
LIC
SSA
World
EAC
25
20
15
1970
1980
1990
2000
2010
Sources: United Nations National Accounts Main Aggregates Database; authors’ calculations.
1
Agriculture, hunting, forestry, fishing
Wholesale, retail trade, restaurants and hotels
3
Transport, storage, and communications
2
4
Thus, although some commentators have argued that natural resources have been a critical factor underlying
strong growth in recent years in at least parts of Africa, this has clearly not been the case for the EAC.
10
Considering the period since 2000—characterized by rapid growth and accelerated structural
transformation—in greater detail, we draw on a more refined sectoral decomposition of
output from a database assembled by IMF staff.5 The data cover 2000–11 for 12 sectors
(agriculture; mining; manufacturing; utilities; construction; wholesale trade; hotels and
restaurants; transport and communications; financial services; real estate and business
services; public administration; and other). Figure 8 compares 2000 and 2011 for the EAC
unweighted average. Similar charts for individual countries are reported in Appendix
Figure A1.
The decline in the share of agriculture was mirrored by relatively evenly distributed gains in
a broad range of other sectors. The biggest winners were construction, transport and
communications, wholesale trade, and public administration. The shares of manufacturing,
mining and utilities displayed smaller changes.
Analyzing the composition of output by sector over the past decade shows that economic
growth has been generally broad based: all sectors experienced positive growth rates on
average during the period, ranging from 3.2 percent in agriculture to 10.8 percent in transport
and communications. The share of agriculture in total output declined from 36 percent in the
year 2000 to 28 percent in 2010. This decline is mirrored by small gains in shares by several
other sectors—especially subcomponents of the service sector (Figure 8).
Figure 8. EAC: Sector Shares and Changes, 2000 vs. 2010
(average; in percent)
Public
admin
6%
Real
estate 1/
7%
2000
Other
9%
Fin
services
2%
Transp 2/
6%
Hotels and
rest
5%
Public
admin
Real
7%
estate 1/
7%
Agriculture
36%
Other
9%
Agriculture
28%
Fin
services
3%
Manufact
9%
Constr
6%
Utilities
2%
Mining
1%
Hotels and
rest
5%
Constr
2.0
Transp 2/
1.7
Whls. trade
1.5
Public admin.
1.2
Manufact
1.1
Mining
0.8
Fin. services
Transp 2/
9%
Whls.
trade
11%
Sector Change, 2000-10
2010
Manufact
9%
Whls.
trade
12%
Constr
8%
Mining
1%
Utilities
2%
0.5
Utilities
-0.2
Real estate 1/
-0.5
other
-0.5
Hotels and rest.
Agriculture
-0.7
-7.1
-10.0
-5.0
0.0
In Percent
5.0
1/
Real estate & business services. 2/ Transport and communications.
Sources: Data assembled by IMF desk economists input; authors’ calculations.
Exports are rising as a share of GDP and are becoming more geographically diversified
Evidence on the increasing role of exports in GDP in the EAC is similar to the trend
observed elsewhere and is consistent with the economies becoming more internationally
integrated and thus more modern in terms of their production processes and outputs. Exports
as a share of GDP in the EAC rose from 12 percent in 1990 to 19 percent in 2010, but the
5
An effort coordinated by N. Gigineishvili. See Data Appendix for sources and methodology.
11
increase was particularly pronounced in Tanzania and Uganda, and to a lesser extent in
Kenya and Rwanda.
Figure 9. Exports, 1990-2010
(percent of GDP)
LIC
SSA
World
EAC
40
35
45
Burundi
Kenya
Rwanda
Tanzania
Uganda
40
35
30
30
25
25
20
20
15
10
15
5
0
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
10
Sources: World Development Indicators, World Bank; authors’ calculations.
Exports have also become more diversified by geographic destination, as evidenced by the
declining Theil index (a widely used measure of diversification—the lower the index, the
higher diversification; for the definition and method of calculation, see Appendix) computed
with the shares of trade partner countries in total exports and the declining share of the top
ten destination countries in total exports (Figure 10).
Figure 10. Partner Diversification
Theil Index
5.5
5.0
4.5
Share of Top 10 Partners
1.1
SSA
Kenya
Tanzania
Burundi
Rwanda
Uganda
4.0
1.0
0.9
3.5
0.8
3.0
0.7
2.5
0.6
2010
2007
2004
2001
1995
1998
1992
1989
1986
Kenya
Tanzania
1983
1977
1980
1974
1971
1968
Burundi
Rwanda
Uganda
1965
2007
2004
2001
1998
1995
1992
1989
1986
1983
1980
1977
1974
1971
1968
0.4
1965
1.0
1962
1.5
0.5
1962
2.0
Sources: United Nations, COMTRADE ; authors’ calculations.
Notes: In the left panel, a declining Theil index is indicative of increasing diversification. In the right panel, for
each country, the line depicts the share of exports going to the top ten export destinations.
12
Exports Products Have Become More Diversified
Reflecting rapid growth and structural transformation of the EAC economies, exports have
also expanded in a wide range of goods and services outside the traditional agricultural
sector. The share of agriculture in exports fell from 4/5 of all exports in the 1970s-90s to 2/3
in 2010, with mirroring gains in manufacturing. (Figure 11 for the EAC unweighted average;
Appendix Figure A2 provides individual country information)
Figure 11. Structure of Exports
(in percent of total exports of goods)
1970
1990
2009
3%
14%
6%
11%
12%
EAC
23%
80%
20%
SSA
Agriculture
Manufacturing
Mining
Agriculture
Manufacturing
Mining
Agriculture
Manufacturing
Mining
66%
85%
22%
27%
44%
15%
51%
65%
27%
29%
Agriculture
Manufacturing
Mining
Agriculture
Manufacturing
Mining
Agriculture
Manufacturing
Mining
Considering data since the early 1960s, exported products have become more diversified in
all EAC countries at a more rapid pace than for the Sub-Saharan average, with a steeper trend
since the early 1990s. The Theil index shows that by mid-2000s export products were at least
as diversified in each EAC country as in the average country in Sub-Saharan Africa, with
Kenya and Tanzania being substantially ahead of their peers (Figure 12). The results are
confirmed using alternative measures of diversification, such as the share of the top ten
products in total exports of goods.
13
Figure 12. Export Product Diversification
Theil Index
7.0
Share of Top 10 Products
1.1
6.0
1.0
5.0
0.9
0.8
4.0
0.7
3.0
0.6
SSA
Kenya
Rwanda
0.5
Burundi
Tanzania
Uganda
2007
2004
2001
1998
1995
1992
1989
1986
1983
1980
1977
1974
1971
1968
1965
1962
2007
2004
2001
1998
1995
0.4
1992
1989
1986
1983
1980
Burundi
Tanzania
Uganda
1977
1974
1968
1965
1962
1.0
1971
SSA
Kenya
Rwanda
2.0
Source: UN COMTRADE, authors’ calculations.
The sheer number of export products has also increased substantially, especially since the
early 1990s, and especially in manufacturing: the number of distinct products exported by
Uganda, for example, rose from about one hundred in 1980 to more than five hundred in
2010, with more than three hundred products in manufacturing. To minimize potential bias
from a large number of small exports, the same analysis has been undertaken for the distinct
products for which export revenues exceed US$1 million (or US$10 million). As shown in
Figure 13, the results confirm that that the increase in the number of products has been
sizable, particularly in the larger economies (Kenya, Tanzania, and Uganda).
Figure 13. Number of Export Products
60
Burundi
50
50
>USD 10m
>USD 1m
40
150
30
30
20
20
10
10
0
0
50
0
160
Tanzania
50
30
>USD 10m
>USD 1m
100
>USD 10m
80
>USD 10m
50
20
20
10
0
0
1962
1965
1968
1971
1974
1977
1980
1983
1986
1989
1992
1995
1998
2001
2004
2007
1962
1965
1968
1971
1974
1977
1980
1983
1986
1989
1992
1995
1998
2001
2004
2007
>USD 1m
60
30
40
0
70
40
60
20
10
Uganda
80
120
>USD 1m
100
90
140
40
100
1962
1965
1968
1971
1974
1977
1980
1983
1986
1989
1992
1995
1998
2001
2004
2007
1962
1965
1968
1971
1974
1977
1980
1983
1986
1989
1992
1995
1998
2001
2004
2007
Rwanda
>USD 10m
>USD 1m
>USD 10m
60
Kenya
200
>USD 1m
40
250
1962
1965
1968
1971
1974
1977
1980
1983
1986
1989
1992
1995
1998
2001
2004
2007
SSA Average
Source: UN COMTRADE, authors’ calculations.
1962
1965
1968
1971
1974
1977
1980
1983
1986
1989
1992
1995
1998
2001
2004
2007
60
14
The sophistication of export products has increased
The EAC countries have generally moved toward exporting more “sophisticated” products,
defined as those predominantly produced by advanced economies. Specifically, a
“sophistication” index is calculated from COMTRADE data at the 4-digit SITC level, using a
method similar to Hausmann and others (2007). The construction of the index proceeds in
two steps. The first step is to construct a “productivity” index for each product (worldwide)
each year, by considering the weighted average of per capita GDP of the countries that export
it, with the weights being given by the share of the given product in each country’s total
exports. Intuitively, more productive or more sophisticated products worldwide are those that
feature more prominently in the export structure of more advanced economies. The second
step is to compute a country’s sophistication index in each given year as the weighted
average of the “productivity” index of its exports, where the weights are the value shares of
the products in the country’s total exports. As shown in Figure 14, the sophistication index of
each of the EAC countries rose visibly between 1990 and 2010. Indeed, the gains for most
EAC countries were noticeably greater than the average for sub-Saharan Africa and other
regions. The smallest gain was experienced by Kenya, which started out with the highest
sophistication index within the EAC in 1990. In the other four countries, “convergence”
clearly seems to have been at play. At the end of the period, not only Kenya but also Uganda
were slightly above the sub-Saharan average, and Tanzania was close to it.
Figure 14. Sophistication Index, 1990 vs. 2010
All Regions
EAC
9.5
9.5
9.0
9.0
8.5
8.5
8.0
8.0
7.5
7.5
7.0
7.0
6.5
6.5
6.0
6.0
1990
2010
BDI
1990
2010
KEN
1990
2010
RWA
1990
2010
TZA
1990
2010
UGA
1990 2010 1990 2010 1990 2010 1990 2010 1990 2010 1990 2010 1990 2010
East Asia & Europe &
Latin
Middle East North
Pacific Central Asia America & & North
America
Caribbean
Africa
South Asia
SubSaharan
Africa
Source: UN COMTRADE, Penn World Table, authors’ calculations.
The Quality of Individual Items Being Exported Has Improved
Summary measures of the quality of exports (calculated by Henn and others, 2013, see
Appendix) within each export category also indicate a trend toward improved quality for
each of the EAC countries. For each category of items exported by a given country, quality is
measured as the ratio of the value of such items exported by the country in question to the
average value of such item exported by all other countries. The results are reported in
Figure 15, where the red diamonds represent the quality of a given export category by the
country in question (right hand side scale, as a ratio of the quality of the same category for all
15
other countries), and the blue bars represent the share of the same goods export category in
the country’s total exports (in percent, on the left hand side scale). For example, in 2009,
Burundi’s coffee was considered to be of a relatively high quality (0.8, on a scale from
0 to 1) and coffee represented 80 percent of Burundi’s goods exports as estimated in the
database. Focusing on changes in the quality of the various export items, there is no clear
rising trend, perhaps not too surprising given that most of the products are primary
commodities or goods that involve limited processing. It is also worth recalling that the
quality is measured relative to trading partners, suggesting that the countries in the region are
broadly keeping pace with their competitors. Focusing on the shares of the top ten products
in total exports and on a more positive note, the figures reveal greater diversification in 2009
than in 1980, consistent with previous results on diversification using a different data source
and methodology.
16
Figure 15. Quality of Top 10 products (four-digit level)
1980
100
80
2009
Burundi
60
40
20
0
40
35
30
25
20
15
10
5
0
100
80
Kenya
1.2
1.0
0.8
0.6
0.4
0.2
0.0
100
1.2
25
1.0
20
0.8
0.6
Rwanda
60
40
20
0
Burundi
80
60
40
20
0
Kenya
1.2
1.0
0.8
0.6
0.4
0.2
0.0
1.2
1.0
0.8
15
0.6
0.4
10
0.2
5
0.2
0.0
0
0.0
1.2
1.0
0.8
0.6
0.4
0.2
0.0
35
30
25
20
15
10
5
0
0.4
Rwanda
1.2
1.0
0.8
0.6
0.4
0.2
0.0
35
30
25
20
15
10
5
0
Tanzania
1.2
1.0
0.8
0.6
0.4
0.2
0.0
12
10
8
6
4
2
0
Tanzania
1.2
1.0
0.8
0.6
0.4
0.2
0.0
120
100
80
60
40
20
0
Uganda
1.2
1.0
0.8
0.6
0.4
0.2
0.0
35
30
25
20
15
10
5
0
Uganda
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Export share
Quality (right axis)
Note: Gold is not included in the export products here.
Source: UN COMTRADE and authors’ calculations.
17
IV. SUMMARY AND CONCLUSIONS
In this paper, we have reported several stylized facts about economic growth experienced by
the EAC member countries. The pace of overall economic growth as well as the associated
structural transformation of the economies accelerated during the past decade. Both output
and exports have become more diversified: the share of agriculture has fallen substantially;
the gains in share have been broadly distributed, with the largest gains going to construction,
transportation, and wholesale trade; manufacturing and mining posted modest gains. Rapid
growth has thus not been driven by a narrow range of products (as might have been the case
with natural resources in some other low-income developing countries). Moreover, the
sophistication and quality of items exported by the EAC countries have improved over time,
and more noticeably during the past decade than previously.
When attempting a possible interpretation of these facts, the picture that seems to emerge is
that recent diversification and structural transformation bode well for continued economic
growth. Yet, the kind of growth observed seems to be one in which consumer and investment
demands for more sophisticated goods and services are beginning to be met, as one would
expect as per capita incomes rise in the region. There do not yet seem to be any clear winners
on the production side that are likely to embed a clear and durable comparative advantage in
international markets, particularly beyond the region. Nor are there major quality
improvements vis-à-vis competitor countries, where progress is also being made. These
observations may instill a further note of caution against projecting continued rapid growth
into the distant future.
As emphasized in the introduction, one cannot expect definitive answers about future growth.
To the extent that the observations reported above can provide valuable clues, they are
consistent with the view that, going forward, economic growth is likely to remain healthy but
to slow down to a more moderate pace than during the past decade, consistent with the
statistical pattern for “reversion to the mean” reported in Section II of this paper.
18
APPENDIX
Methodology and Data
Sectoral Output Database
The new IMF in-house database was compiled using information on the sectoral composition
of GDP provided by the country authorities to IMF country desks. The dataset contains value
added series in constant prices for 12 economy-wide sectors in Sub-Saharan Africa for the
period of 2000–2011. The 12 sectors are: agriculture (including forestry and fishing); mining
and quarrying; manufacturing; utilities (electricity and water); construction; trade and repairs;
hotels and restaurants; transportation and communication; financial intermediation; real
estate and business activities; public administration; and other. FISIM and net taxes are
excluded. The share of each sector is calculated as a ratio of the value added of a given sector
to the sum of all 12 sectors. For some countries, data were not available for all 12 sectors.
Measure of Diversification of Export Products/Partners
The main export data we use is from UN Comtrade database, using the SITC1 classification
of products at the four-digit level. The time period covers 1962 to 2010.
The Theil index of export diversification may be expressed as follows:
∑
∑
, where
For the export product level Theil index, xk is the export value of product k and n is the
number of products; μ is therefore the products’ average dollar value. For the partner level
Theil index, xk is the export value to partner k and n is the number of partners for each
exporter.
Measure of Sophistication of Export Products
To measure the sophistication of a country’s export products, we use an approach developed
by Hausmann and others (2007), which we summarize here for ease of reference. The first
step is to construct a “productivity” index for each product (worldwide) each year, by
considering the weighted average of the per capita GDP of the countries that export it, with
the weights being given by the share of the given product in each country’s total exports.
Specifically, the productivity level associated with product k may be expressed as follows:
⁄
∑
⁄
where denotes per capita GDP of country j and denotes total exports by country j. The
numerator of the weight is the value-share of export k in total exports by country j. The
denominator of the weight is the sum-total of the value shares of export k by all countries.
Intuitively, more productive or more sophisticated products worldwide are those that feature
19
more prominently in the export structure of more advanced economies. The second step is to
compute a country’s sophistication index in each given year as the weighted average of the
“productivity” index of its export products, where the weights are the value shares of each
,
product in the country’s total exports. The sophistication of country j’s export basket,
may thus be expressed as follows:
1
where export products are indexed as l=1…N.
Measure of Quality of Export Products
Export quality is measured using a unit value, or trade price, which is defined as a ratio of
value of exports over quantity of exports for any given product category. Quality index data
were obtained from Henn et al (2013), who using UN Comtrade data calculate quality by
adjusting unit values for differences in production costs and for the selection bias stemming
from relative distance between exporter and importer countries:
′ ln ′ ln ′ ln
Where subscripts m, x, and t denote importer, exporter, and time period; pmxt is trade price; yxt
is exporter income per capita, and Distmx is the distance between importer and exporter.
Coefficients are estimated separately using the methodology by Hallak and Schott (2011).
To enable cross-product comparisons, all quality estimates are first normalized by their 90th
percentile in the relevant product-year combination. The quality estimates are then
aggregated, using current trade values as weights, to higher level sectors and to country-level
totals. The normalization to the 90th percentile is repeated at each level of aggregation.
20
Figure A1. Changes in Composition of Output by Sector, Individual Countries
Burundi, 2000
Financial
services
2%
2011
Real estate Public
1/ administration
0%
5%
Sector Change, 2000-11
Public administration
Other
5%
Transport 2/
3%
Hotels and
restaurants
16%
3%
2.6
Wholesale trade
Agriculture
34%
1.2
Utilities
0.4
Transport 2/
Transport 2/
3%
0.3
Real estate 1/
Manufacturing
11%
Utilities
1%
3.4
financial services
Hotels and
restaurants
13%
Wholesale
trade
7% Construction
4.2
Construction
Real
estate
Public
1/ administration
Financial 0%
11%
services
4%
Agriculture
46%
6.0
other
Other
9%
Kenya, 2000
-1.9
Hotels and restaurants
Manufacturing
10%
Utilities
1%
Construction
7%
-0.5
Manufacturing
Mining
0%
Wholesale
trade
8%
Mining
1%
0.0
Mining
2010
-2.9
Agriculture
-12.7
-20.0
-10.0
0.0
In Percent
10.0
Sector Change, 2000-10
Transport 2/
Public
administration
4%
Other
15%
Public
administration
6%
Agriculture
32%
Real estate 1/
6%
Agriculture
28%
Real estate 1/
6%
Financial
services
3%
Financial
services
4%
Wholesale
trade
10%
Hotels and
restaurants
1% Construction
3%
Manufacturing
11%
Manufacturing
11%
Other
5%
Hotels and
restaurants
2%
Public
administration
7%
Mining
0%
Manufacturing
6%
-0.4
-0.6
-0.8
-1.8
-1.0
In Percent
Sector Change, 2000-11
Transport 2/
3.5
Construction
3.3
Other
2.3
1.3
Hotels and restaurants
Agriculture
38%
1.2
Manufacturing
0.4
Mining
0.3
Financial services
0.1
Wholesale
trade
14%
Real estate 1/
11%
Mining
3%
Manufacturing
10%
Wholesale
trade
16%
Utilities
3%
Construction
8%
2011
2.0
Transport 2/
1.7
Wholesale trade
1.5
Public administration
1.2
Manufacturing
1.1
Mining
0.8
financial services
0.5
Utilities
-0.2
Real estate 1/
-0.5
Mining
0%
Real estate 1/
10%
Utilities
3%
Financial
services
4%
Hotels and
restaurants
5%
Construction
16%
Wholesale
trade
14%
Real estate & business services. 2/ Transport and communications.
Sources: IMF desk economists; authors calculations.
-0.5
Hotels and restaurants
Agriculture
-0.7
-7.1
-10.0
-5.0
0.0
In Percent
5.0
5.7
Construction
Agriculture
15%
Manufacturing
8%
Utilities
4%
5.0
Sector Change, 2000-11
Transport 2/
10%
Construction
12%
-5.0
0.0
In Percent
Transport 2/
Other
11%
Agriculture
25%
Mining
0%
-10.0
other
Hotels and
restaurants
2%
Public
administration
4%
Manufacturing
8%
Agriculture
Construction
Transport 2/
9%
Utilities
3%
Other
15%
-2.3
-8.9
Sector Change, 2000-11
Other
4%
Uganda, 2000
Wholesale
trade
14%
Mining
1%
Manufacturing
7%
Agriculture
23%
Manufacturing
9%
Construction
6%
Real estate 1/
9%
Utilities
0%
Agriculture
30%
Mining
2%
-0.7
Real estate 1/
2011
Financial
services
2%
-0.5
Public administration
Wholesale
trade
12%
Construction
9%
Other
5%
4.0
Wholesale trade
Public
administration
9%
Real estate 1/
12%
Financial
services
2%
Transport 2/
7%
Hotels and
restaurants
5%
0.0
Real estate 1/
financial services
-6.0
Other
8%
Tanzania, 2000
Financial
services
0%
Transport 2/
4%
0.0
Utilities
HotelsConstruction
and
restaurants6%
Utilities
1%
1%
Public
administration
4%
0.2
Manufacturing
Agriculture -4.5
Utilities
Transport 2/
8%
Wholesale
trade
10%
Hotels and
restaurants
3%
Hotels and restaurants
Construction 3%
4%
Real estate 1/
6%
Financial
services
3%
Agriculture
47%
Transport 2/
5%
1/
Mining
0%
2011
Public
administration
6%
Financial
services
3%
0.5
0.5
Public administration
Wholesale
trade
12%
Hotels and
restaurants
1%
Utilities
2%
Real
estate 1/
9%
Utilities
Construction
other
Mining
1%
Rwanda, 2000
Public
administration
7%
2.0
Mining
Transport 2/
14%
Transport 2/
9%
4.9
Wholesale trade
Other
14%
4.6
financial services
3.9
Real estate 1/
0.6
Wholesale trade
0.3
Mining
0.2
Public administration
0.0
Hotels and restaurants
-0.1
Manufacturing
-0.3
Utilities
-1.3
other
Agriculture
-20.0
-4.0
-9.5
-10.0
0.0
In Percent
10.0
21
Figure A2. Changes in Sectoral Structure of Exports, Individual Countries.
1990
1970
2010
1%
6% 2%
2% 3%
Agriculture
Manufacture
Mining
11%
Burundi
88%
92%
95%
1970
1990
2010
3%
3%
10%
17%
Kenya
34%
28%
62%
80%
63%
2010
1990
1970
3% 5%
37%
39%
Rwanda
Agriculture
Manufacture
Mining
Agriculture
Manufacture
Mining
41%
60%
92%
22%
1%
1970
1990
2010
0%
7% 3%
11%
21%
Agriculture
Manufacture
Mining
24%
Tanzania
1970
1%
0%
65%
79%
90%
2010
1990
1%
0%
0%
27%
Uganda
99%
99%
73%
Agriculture
Manufacture
Mining
22
References
Davoodi, Hamid, ed., 2012, The East African Community After 10 Years—Deepening
Integration, International Monetary Fund.
Easterly, William, 1999, “Life During Growth,” Journal of Economic Growth, Vol. 4, No. 3,
pp. 239–275.
Easterly, William, 2001, “Growth Implosions and Debt Explosions: Do Growth Slowdowns
Explain Public Debt Crises?” Contributions to Macroeconomics, Vol. 1, No. 1,
Berkeley Electronic Press Journals.
Easterly, William, Michael Kremer, Lant Pritchett, Lawrence H. Summers, 1993, “Good
policy or good luck? Country growth performance and temporary shocks,” Journal of
Monetary Economics, Vol. 32, No. 3, pp. 459–483.
Hausmann, R., J. Hwang, and D. Rodrik, 2007, “What You Export Matters,” Journal of
Economic Growth, Vol. 12, pp. 1–25.
Henn, Christian, Chris Papageorgiou, and Nikola Spatafora, 2013, “Export Quality in
Developing Countries,” IMF Working Paper No. 13/108.
Ho, Giang, and Paolo Mauro, 2014, “Growth: Now and Forever?” IMF Working Paper
No. 14/117.
International Monetary Fund, 2013, “Drivers of Growth in Nonresource-Rich Sub-Saharan
African Countries,” Chapter 2 in Regional Economic Outlook: Sub-Saharan Africa—
Keeping the Pace, October.
Loungani, Prakash, and Grace Juhn, 2012, “Further Cross-Country Evidence on the
Accuracy of the Private Sector’s Output Forecasts,” IMF Staff Papers, Vol. 49, pp.
49–64.
McAuliffe, Catherine, Sweta C. Saxena, and Masafumi Yabara, 2012, “The East African
Community: Prospects for Sustained Growth,” IMF Working Paper No. 12/272.
Pritchett, Lant, and Lawrence H. Summers, 2013, “Asiaphoria,” Harvard University.