CHAPTER 4

Botswana’s Success: Good Governance,

Good Policies, and Good Luck
Michael Lewin

ver the past 60 years, Botswana’s economy has

O
Annual growth in per capita income averaged 7.0 percent
been one of the most successful in the world. The between 1966 and 1999 (table 4.1 and figure 4.1). The coun-
country’s achievement is remarkable, because at try’s performance is particularly impressive compared with
independence, in 1966, its prospects were not encouraging. that of other African economies (figure 4.2).
Critics have argued that the social gains from this
growth have been somewhat limited. In fact, in addition to
ECONOMIC AND SOCIAL INDICATORS
the gains in health and life expectancy noted above, there
Botswana is a sparsely populated, arid, landlocked country; have been gains in poverty reduction and education. The
at independence it was also one of the poorest countries in proportion of poor people fell from about 50 percent in
the world, with per capita income of just $70 a year. In the 1985 to 33 percent in 1994, and the proportion of people
first few years of independence, about 60 percent of current completing at least primary school rose from less than
government expenditure consisted of international devel- 2 percent at independence to about 35 percent in 1994
opment assistance. There were only 12 kilometers of paved (Leith 2005).
roads, and agriculture (mostly cattle farming for beef pro- Not all indicators are as positive: income distribution in
duction) accounted for 40 percent of gross domestic Botswana remains very unequal (the Gini coefficient was
product (GDP). about 0.55 in 1994).2 Unemployment remains high, reflect-
By 2007 Botswana had 7,000 kilometers of paved roads, ing to a large extent rural to urban migration, although it
and per capita income had risen to about $6,100 ($12,000 too has fallen, dropping from about 21 percent in the 1990s
at purchasing power parity), making Botswana an upper- to about 17 percent in 2008.
middle-income country comparable to Chile or Argentina.
Its success is also evident in other measures of human
DIAMONDS AND DEVELOPMENT
development. At independence, life expectancy at birth was
37 years. By 1990 it was 60, 10 years above the African aver- Botswana’s extraordinary growth was fueled by minerals,
age. Under-five mortality had fallen to about 45 per 1,000 particularly diamonds. At independence, beef, the country’s
live births in 1990, compared with 180 for Africa as a whole. main export and largest sector, contributed 39 percent of
Development assistance has shrunk to less than 3 percent of GDP. From independence until the 1970s, international aid
the government budget, and agriculture currently accounts dominated the government budget and was the main source
for only about 2.5 percent of GDP. Major strides have also of foreign exchange. At that time the mineral sector, mainly
been made in infrastructure and education.1 diamonds, began to take off and soon became the dominant

81
 Figure 4.2 Average per Capita Income in Africa and
Table 4.1 Annual Growth in per Capita Income in
Botswana, 2001–09
Selected Economies, 1966–99
(purchasing power parity at current
Economy Average growth rate dollar prices)
Botswana 7.0
Chile 2.1
16,000
Hong Kong SAR, China 4.6
Indonesia 3.8 14,000
Ireland 4.1
12,000
Korea, Rep. of 6.1
Singapore 6.2 10,000

Dollars
Thailand 4.6 8,000
Source: Adapted from Leith 2005. 6,000
4,000

Figure 4.1 Average Annual Growth in per Capita Income 2,000

in Botswana, 1980–2008 0
2001 2002 2003 2004 2005 2006 2007 2008 2009
8 Africa Botswana
7 Source: OECD n.d.
6
5
Percent

4 “good economics, bad politics.” In almost all developing

3 countries, the government owns the mineral resources and
2
is therefore the main recipient of the revenues from their
extraction. This concentration of revenues with the govern-
1
ment as the conduit of benefits to the rest of the economy
0
1980–90 1990–2000 2000–08
can lead to a host of problems, including rent-seeking, cor-
ruption, and the efficiency losses that result from them. It is
Source: Leith 2005.
not surprising that, with easily accessible wealth concen-
trated in the hands of the government, malignant dictator-
sector. Income growth and the growth of the mining sector ships and predatory regimes often thrive. There also seems
accelerated in tandem from about 1974/75 until recently. to be a greater tendency for armed civil conflict in mineral-
Luck—the discovery of diamonds as well as other based economies (Collier and Hoeffler 2004). These “bad
important minerals—was clearly an element of Botswana’s politics” are extremely costly.
success. However, there is considerable disagreement over Botswana has avoided these manmade disasters. Part of
whether the discovery of minerals in a developing economy its success may reflect luck: Botswana’s relatively homoge-
generally brings good luck or bad (box 4.1).3 Indeed, many neous population has less potential for ethnic polariza-
countries in Africa, including Zambia, Nigeria, the Demo- tion, which when combined with mineral rents can be
cratic Republic of Congo, Sierra Leone, and others, have particularly combustible. Still, most of the credit must be
squandered vast amounts of their natural wealth. One does given to the leadership, which, since independence, has
not have to be persuaded that natural resources are neces- designed and fostered the conditions of governance that
sarily a curse to conclude that they certainly are not sufficient have ensured stability and social and economic progress.
for economic success or even a good predictor of it. In the The government established respect for property rights
case of Botswana, minerals turned out to be lucky, but other and the rule of law. It maintained a high degree of trans-
key ingredients in the recipe for success were present, includ- parency, which was reinforced by continuing the Tswana
ing good governance and good economic management. tribal tradition of consultation. These consultative insti-
tutions, known as kgotla, created a degree of trust in the
government—the sense that government exists to serve
Avoiding the bad governance curse
the people and promote development and is not the
Mineral-based countries seem to be prone to bad govern- instrument of one group or individuals for the purpose of
ment, a phenomenon Acemoglu and others have termed getting hold of the wealth. Tswana tradition also

82 CHAPTER 4: BOTSWANA’S SUCCESS: GOOD GOVERNANCE, GOOD POLICIES, AND GOOD LUCK
Box 4.1 The Resource Curse

The counterintuitive notion that the endowment of nat- can be financed by the revenue from the mineral
ural resources is a curse rather than a blessing has exports.
become received wisdom. Sachs and Warner (1995) Because the entire process is initiated by an increase
report regression results showing that being a natural in income, expenditure, and consumption, it is not clear
resource or mineral exporter reduces a country’s devel- why it should be thought of as a disease. One reason
opment prospects. Many other researchers find evidence given is the presence of positive externalities in the
to support this claim (see for example, Auty 2004 and declining sector. The declining traded goods sector is
the references therein). The fact that most of the East thought to have spillover effects that contribute to long-
Asian tigers (the Republic of Korea; Taiwan, China; run growth and industrialization. Hence, the real
Hong Kong SAR, China; and Singapore) plus Japan are appreciation caused by mineral exports is said to cause
resource poor has added to the perception that endowed “deindustrialization,” contributing to poor long-run
wealth is an obstacle to growth and development. performance. Of course, in Africa and Botswana in
There is certainly evidence that a disproportionate particular, the declining traded goods sector was agri-
number of the poorest countries are dependent on culture, not industry (there was no significant industry
mineral exports. The direction of causality, however, is to deindustrialize). Some might argue that mineral
far from clear: does mineral dependence retard devel- exports prevent the real depreciation needed to indus-
opment, or does retarded development create mineral trialize. Additionally, it is argued, the mineral sector
dependence? fails to stimulate the nonmineral sector through link-
Accounting for the endogeneity of mineral depen- ages to the rest of the economy.
dence overturns the Sachs-Warner conclusions. Coun- Against this effect, one must weigh the advantages
tries that fail to develop remain resource dependent, but that minerals bring in terms of increased saving and
resource abundance may be correlated with higher hence investment, which should offset any loss of com-
growth (see, for example, Brunnschweiler and Bulte petitiveness caused by the real exchange rate. Moreover,
2008; Alexeev and Conrad 2009). Many rich countries— because government is usually the chief beneficiary of
including Canada, Australia, Norway, the United States, the resource revenues, the increase usually eases con-
and Latin America’s top performer, Chile—were or are straints on public goods, such as infrastructure, educa-
mineral dependent. In Africa the two richest countries, tion, and health, which should contribute to growth
South Africa and Botswana, owe much of their wealth and industrialization. If private investment or public
to gold and diamonds. goods expenditure fails to materialize, the problem lies
In short, resources are not necessarily a curse, but not in the real exchange rate but rather in the failure of
many countries have squandered their resources, and in financial markets to allocate the saving and investment
some cases, the presence of mineral resources may have productively or of government to invest wisely.
contributed to economic stagnation or decline. What is
The Volatility Curse
it about mineral endowments that can make things go
wrong? Mineral prices tend to be volatile, which, it is argued,
destabilizes the economy (Hausmann and Rigobon
Dutch Disease
2003). This volatility affects the economy through the
Dutch disease refers to the deleterious effects that pur- exchange rate, which appreciates in the boom and
portedly result from the real appreciation of the cur- depreciates in the bust, and through fiscal policy, which
rency caused by a booming resource export sector tends to be procyclical, leading to inefficient “stop-go”
(Corden and Neary 1982). The discovery of minerals provision of government services and infrastructure
or an increase in their international price leads to an projects.a It is argued that the perceived riskiness that
increase in income and expenditure. As long as some of this volatility engenders leads to lower investment, par-
the increased expenditure goes to domestically pro- ticularly foreign direct investment, which seeks out
duced goods, there will be a real appreciation of the safer markets. In addition, governments tend not only
currency, which causes productive resources to move to increase spending when current revenues rise but often
from nonmineral traded goods, the output of which also to believe that revenues will continue to rise; they
declines, to nontraded goods, the output of which therefore borrow against future revenues. Instead of
increases. The resulting excess demand for tradables being able to use accumulated assets to help in the bust,
(continued next page)

CHAPTER 4: BOTSWANA’S SUCCESS: GOOD GOVERNANCE, GOOD POLICIES, AND GOOD LUCK 83
 Box 4.1 (continued)

economies find themselves contending with debt Exhaustibility

overhang.
Minerals are a windfall, even in the long run, in the
It is possible to insulate the domestic market from
sense that they are finite and that when they are
the volatility of the commodity markets. Because the
depleted an economic slump will occur. That a blessing
government is the conduit of the cycle to the economy,
does not last forever hardly makes it a curse, however. If
it is also the solution. The key is to delink expenditure
at least some of the windfall is saved, in the form of
and revenue, essentially by saving all or part of the rev-
physical and human capital or financial assets, future
enues, thereby replacing a volatile income stream with a
generations will benefit from it. Moreover, the deple-
more stable one. Expenditure can then be based on this
tion usually does not occur suddenly. With some fore-
more stable, permanent income. In this way, consump-
sight, the adjustment and transition can be cushioned
tion can be smoothed over time, in some cases over
by accumulated saving.b
generations. Many countries, including Botswana,
Chile, Kuwait, Norway, Qatar, and Saudi Arabia, have How Has Botswana Fared?
successfully applied such an expenditure-smoothing
To a large extent, Botswana overcame the threat of the
policy.
resource curse:
The Governance Curse
1. Dutch disease was avoided, most significantly by
In most developing countries the government claims government investment in public goods and infra-
ownership of all minerals. If government institutions structure. The government also took measures to
are weak to begin with, there is a strong likelihood that help boost productivity, by limiting parastatals and
the resources will be squandered. In addition, the pres- avoiding import substitution policies.
ence of the resources may weaken and corrupt govern- 2. The volatility curse was overcome by unlinking pub-
ment. The relative ease with which mineral revenues are lic expenditure from revenue. The government
collected can lead to a lack of transparency and established savings funds and avoided typical pro-
accountability. The resulting interdependence of gov- cyclical behavior and real exchange rate volatility.
ernment and the mining industry reinforces this ten- 3. The governance curse was avoided, through a series
dency. Government itself is seen as the means to of actions described in the chapter. Botswana’s suc-
acquiring wealth, which typically leads to rent seeking cess in this regard is exceptional.
and corruption. There is also evidence that in many 4. The question of whether the exhaustibility chal-
cases mineral wealth provokes or fuels internal conflicts lenge has been met remains unclear. If current pre-
(Collier and Hoeffler 2004). dictions are correct, diamond revenues will begin
In an ironic metaphor, underground minerals are declining in 2016 and could be exhausted by 2029.
often referred to as one of the “commanding heights” of Botswana has adopted many of the appropriate
the economy. Postindependence governments claimed policies to prepare for the depletion of its mineral
the need to control these commanding heights for the base, by accumulating funds for the future, building
benefit of all the governed. One of the unforeseen con- infrastructure, and investing in health and educa-
sequences of this policy was the perception that control tion. However, the growth of the nonmineral traded
of government is the path to wealth, which in turn, goods sector, which has remained flat at about
unsurprisingly, has often led to regional and ethnic con- 5 percent of GDP, will need to accelerate to avoid
flict for control of these resource rents (Collier and balance of payments problems as diamond revenues
Hoeffler 2004). begin to decline.

a. Rather than being procyclical, in most developing countries, the government initiates the cycle because it is the recipient of the
rents. The boom therefore begins with government expansion and the bust with its contraction. Hence, real exchange rate vari-
ability is also the result of volatile government expenditure.
b. A full discussion of mineral resource management in the context of intertemporal and intergenerational allocations goes
beyond the objectives of this chapter. Essentially, the optimal management of resource rents requires expenditure smoothing
not only to avoid cyclical volatility but also to spread the benefits of the windfall over time. Like a rational household, an opti-
mizing government will attempt to base its long-run expenditure on some calculation or estimate of its “permanent” or stable
revenue rather than on its current or transitory revenue. For some calculations of permanent income in the case of Botswana,
see Basdevant (2008).

84 CHAPTER 4: BOTSWANA’S SUCCESS: GOOD GOVERNANCE, GOOD POLICIES, AND GOOD LUCK
respected private property; the fact that many of the tribal economies seem to spawn the predatory type of dictatorship
leaders who helped usher in modern government were rather than the relatively benevolent ones of East Asia’s past.
also large cattle owners may have reinforced this respect. Although this notion is largely speculative, it seems that a
Acemoglu and Robinson (1999) emphasize that property democratic government dependent on mineral wealth is
rights and the rule of law are the key factors explaining more likely than a dictatorial one to be responsive to devel-
development success. They also emphasize the importance opment needs, to settle disputes peacefully, and to respect
of the preindependence colonial regime in determining the the rule of law. Very few, if any, mineral-rich countries in
adoption of these institutions after independence. More Africa have been ruled as peacefully and productively as
heavily settled colonies tended to establish institutions to Botswana; it is hard to escape the conclusion that the dem-
maintain the rule of law and enforce property rights; in ocratic institutions established at independence are an
sparsely settled colonies, the colonial regimes tended to be important part of the explanation.
more “extractive” or “exploitative,” unconstrained by law Social science cannot rigorously assess the relative
and respect for property. Although Botswana’s colonial past importance or contribution of leadership in the evolution
is not typical of the kind of regime that usually led to a of successful institutions. However, in the case of Botswana,
postindependence government constrained by law and leadership, particularly that of its first president, Seretse
property rights, its unique colonial history had the same Khama, may have been crucial, especially in the areas of
effect.4 Before independence, Botswana was a British pro- mineral exploitation and the rights of the state versus those
tectorate, the colonial power having been “invited” in. of the tribes. The discovery of minerals can easily lead to
Because Botswana was not colonized for economic or civil war and the dissolution of the state. To prevent this
strategic advantage, the colonial rulers, it is argued, did not from happening, even before independence, Khama’s party,
impose the extractive-type regime often found in other the Botswana Democratic Party (BDP), wrote into its plat-
sparsely populated areas. Thus the regime that evolved after form its intention to assert the state’s rights to all mineral
independence was one that respected the law and property resources. After independence, the government reached
and was dedicated to development. agreement on ownership of mineral resources with the
More controversial is the role of democracy. Botswana tribal authorities. Although the largest diamond deposits
has maintained a parliamentary democracy since indepen- were discovered in Khama’s own district of Bamangwato, he
dence. To be sure, its democracy is not perfect: the country chose the country over his tribal land, thus helping limit the
has had one-party rule since independence, women’s repre- possibility of conflict.
sentation is limited, and there has been some criticism that
minorities (particularly the San people) are not treated
Implementing good policies
equally. Nevertheless, the government functions in a demo-
cratic manner, elections are “free and fair,” and the govern- All mineral-based economies face the issue of an appreciat-
ment is responsive to the electorate and transparent in its ing real exchange rate (box 4.2). The well-known possible
dealings. harmful consequences of this are often referred to as Dutch
Is democracy a positive factor for economic growth and disease (see box 4.1). Prevention of Dutch disease in
development? None of the miracle East Asian economies or Botswana consisted of three key components: fiscal saving, a
Latin America’s high performer (Chile) was democratic surplus on the current account of the balance of payments,
during the first years of rapid growth and development, and heavy government investment in infrastructure and
leading many commentators to conclude that property human capital. Together these policies limited the erosion of
rights and the rule of law are more important than democ- domestic productivity and competitiveness that can result
racy (Acemoglu, Johnson, and Robinson 2003). For some from the appreciation of the real exchange rate. High fiscal
observers, democracy is actually a hindrance to economic saving limits current consumption, reducing pressure on
development. Although this conclusion is probably unwar- domestic price inflation, a typical problem in natural
ranted, the fact that many democracies (such as India) have resource booms. Some of the government saving took the
not fared well economically undoubtedly means that form of offshore investments, which directly limited real
democracy is neither sufficient nor necessary for growth exchange rate depreciation and diversified the sources of
and development.5 future foreign exchange revenues. The accumulation of
In the case of mineral-rich countries like Botswana, reserves is a form of self-insurance against short-run
democracy may be an important catalyst. Mineral-dependent declines in mineral revenues as well as long-run reductions

CHAPTER 4: BOTSWANA’S SUCCESS: GOOD GOVERNANCE, GOOD POLICIES, AND GOOD LUCK 85
 Box 4.2 Windfall Economics 101

The mechanism by which mineral resource flows affect money into the economy, however, resulting in domestic
the real exchange rate depends on the nominal price inflation, which will eventually erode local compet-
exchange rate regime maintained by the country. itiveness until the balance of payments surplus is elimi-
Botswana has a fixed exchange rate system. Initially, it nated. This is the effect of real appreciation when the
pegged its currency to the U.S. dollar. When the South nominal rate is fixed. To prevent this from happening,
African rand depreciated relative to the dollar, the government has to sterilize the effects of the resource
Botswana suffered from the appreciation of the pula rents on the domestic money supply. In the medium to
(Botswana’s currency) relative to the rand (the currency long run, the fiscal authority can do so by absorbing the
of its main trading partner).a The pula is therefore now excess domestic money through fiscal budget surpluses.
pegged to the rand. This policy of twin surpluses—a balance of payments
In a fixed-rate regime, the first effect of a resource surplus in tandem with a fiscal budget surplus—is essen-
windfall is the creation of an excess supply of foreign tially the policy that Botswana has followed. (Some ana-
exchange. To maintain the peg, the monetary authority lysts have argued that China has followed a similar policy
must accumulate reserves. Doing so will inject domestic to maintain an undervalued currency.)

a. In contrast, Argentina maintained a dollar peg after the devaluation of the Brazilian real. Many analysts consider this a key ele-
ment leading to the Argentine crisis.

in these revenues as a result of mineral depletion. The instru- seems to have served Botswana well. As noted above, the
ments of public saving took the form of the Public Service number of paved roads increased from around 20 kilome-
Debt Management Fund and the Revenue Stabilization ters in 1970 to 2,300 in 1990; 90 percent of the population
Fund. These funds have proved to be successful mechanisms had access to safe water (compared with 29 percent in
for managing fiscal saving; income from the Revenue 1970); and the number of telephone connections rose from
Stabilization Fund is now a stable source of government around 5,000 in 1970 to 136,000 in 2001.8 Similarly, in edu-
revenue. cational achievement Botswana leads Africa. For example,
Another part of the fiscal saving was channeled to adult literacy, male and female, is around 80 percent
domestic assets, combating the effect of the loss of compet- compared with 69 percent and 50 percent, respectively, for
itiveness by raising productivity. When this investment is the rest of sub-Saharan Africa.9
focused on public goods (for example, infrastructure, As a result of its prudent fiscal policy, Botswana has
health, and human capital), it will contribute to growth enjoyed relative macroeconomic stability and avoided the
without crowding out private sector investment and devel- boom-slump cycles that characterize many mineral-based
opment. Public sector saving was positive in every year economies. Monetary policy was also restrained: for most of
between 1975 and 1996, fluctuating between 10 and 40 per- the postindependence period, inflation was moderate, aver-
cent of GDP. Public sector investment was fairly constant at aging about 10 percent (Maipose n.d.). The periodic slow-
about 10 percent of GDP.6 However, if one counts expendi- downs in the diamond industry have thus by and large not
ture on health and human capital as investment, government been passed on to the rest of the economy. By withholding
investment has consistently remained about 20 percent of some of the benefits from the economy during the booms,
GDP. The capital budget focused on basic infrastructure the government has, to some extent, been able to insulate it
with about 30 percent devoted to water, electricity, roads, from the busts.10
communication, and transportation. Twenty percent of the
capital budget, on average, went to education and around
Shunning import substitution and parastatals
30 percent of recurrent expenditures was devoted to educa-
tion and health.7 While quantifying the effectiveness of Two things Botswana did not do are also significant. Unlike
public expenditure on growth and development is notori- many African countries, it did not adopt a policy of import
ously difficult, particularly in resource-based economies substitution, and it did not expand the extent of state-
where soft budget constraints may lead to overinvestment, owned productive entities, which employ only about 5 per-
the emphasis on infrastructure, health, and education cent of the workforce in Botswana (figure 4.3).

86 CHAPTER 4: BOTSWANA’S SUCCESS: GOOD GOVERNANCE, GOOD POLICIES, AND GOOD LUCK
Figure 4.3 Percentage of GDP Accounted for by Botswana’s combination of policy and governance evidently
State-Owned Enterprises in Selected helped it avoid the worst effects of the resource curse.
African Countries, Early 1990s

30 PLANNING FOR A FUTURE

WITHOUT MINERALS
25
Botswana is still a mineral-based economy: it has not suc-
Percentage of GDP

20 ceeded in significantly diversifying its economy away from

diamonds. Government remains the largest employer
15
(employing 30 percent of the active workforce); other than
10 minerals (which employ a relatively small share of workers),
most production is in nontraded goods. This pattern is typ-
5 ical of mineral-dependent economies. Real GDP and mining
income have moved in tandem since 2000 (figure 4.4), and
0
mining consistently accounts for about 40 percent of GDP,
a

na

ia

go

e
an

ny

bi

bw
oo

an

completely dwarfing the contributions of manufacturing

ha

To

m
w

Ke

nz
er

ba
G

Za
ts

am

Ta

m
Bo

and agriculture (figure 4.5).11 Minerals, mostly diamonds,

Zi
C

Source: Leith 2005.

Figure 4.4 Mining Revenues and Real GDP in Botswana,
2000–09
Being part of the South African Customs Union (which 30,000
required cooperation with the then odious regime in South
Pula (millions) at constant

25,000
Africa) meant maintaining a fairly low tariff regime and
1993/94 prices

20,000
provided a steady stream of government revenue. Avoiding
an activist import-substituting policy and maintaining lim- 15,000
ited government involvement in production seems to have 10,000
paid off for Botswana by allowing it to avoid many of the
5,000
inefficiencies and structural deficits that so often arise from
0
such policies.
00

01

02

03

04

05

06

07

08

09
Trade policy is also an area where good governance and
20

20

20

20

20

20

20

20

20

20
good policies reinforce one another. A government rich with Mining Real GDP
mineral revenues is an inviting target for rent seekers and
worse; restricting the avenues for rent seeking and corrup- Source: Bank of Botswana 2010.

tion thus helps preserve the efficiency and integrity of the

government. Even if the theoretical merits of import substi- Figure 4.5 Contributions of Mining, Manufacturing, and
tution or the existence of state-owned enterprises seem per- Agriculture to Botswana’s GDP, 2000–09
suasive, in practice both often result in inefficiency and
45
drain fiscal resources.
40
Good fiscal policies by themselves may not be sufficient
35
for success. Many mineral-based economies with high rates 30
of investment have not enjoyed the positive results that
Percent

25
Botswana has. The quality of investment is evidently as 20
important as the quantity. Moreover, in the hands of the 15
venal and corrupt, government savings funds can easily turn 10
into slush funds for the favored elites. There may be a lesson 5
in this from Botswana: good governance will aid the effec- 0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
tiveness of good policy, and good policy encourages better
Agriculture Mining Manufacturing
government. Policy formulators should therefore not ignore
the political economy consequences of economic policy. Source: Bank of Botswana 2010.

CHAPTER 4: BOTSWANA’S SUCCESS: GOOD GOVERNANCE, GOOD POLICIES, AND GOOD LUCK 87
 comprise about 85 percent of exports, and mineral revenues ships this may cause, will be necessary to create the compet-
account for about 50 percent of government revenues (Bank itiveness needed to replace falling mineral exports.
of Botswana 2008).12 In 2009 manufacturing contributed less than 5 percent
The diamond industry is expected to decline in the near of GDP (see figure 4.5). This share will have to rise if
future, with revenues projected to begin falling in 2016 and Botswana is to avoid decline. The fact that manufacturing
to be depleted by 2029 (Basdevant 2008). If these projections has been flat as a proportion of income conceals its growth,
are correct, adjustment is inevitable and unlikely to be pain- because just keeping pace with Botswana’s growing econ-
less, although Botswana’s prudent policies—particularly omy has required rapid growth.13 However, manufacturing
regarding saving and investment—have left it in a relatively growth will have to accelerate if Botswana is to raise or even
strong position to facilitate a soft landing. For most years, maintain its standard of living in the long run.
domestic saving has been above 40 percent and investment Key to such growth will be Botswana’s ability to attract
about 35 percent of GDP. (The difference is the surplus on foreign direct investment (FDI). Although FDI fell between
the current account of the balance of payments.) This 2002 and 2007 (Bank of Botswana 2010), Botswana’s repu-
implies that from the perspective of the economy as a whole, tation as a stable country and successful economy should
all revenue from minerals is being saved rather than con- help it attract FDI in the future.14 Policy makers should
sumed, because total saving from national income is roughly examine the country’s overall investment environment with
equal to mineral income. When mineral revenues begin to a view to attracting desirable investment.15
decline, some of the adjustment could thus be absorbed by Clouding Botswana’s smooth transition to a more diver-
saving rather than consumption. Thus, even if growth slows, sified economy is HIV/AIDS, which hit Botswana harder
consumption growth can be maintained as the proportion of than any country in Africa (with the possible exception of
consumption to income rises. South Africa). The epidemic has reversed many of the
The government has also been a big saver. Central bank impressive gains in health indicators achieved over the past
profits and other income from accumulated assets now 100 years and reduced productivity. The number of deaths
make up about 30 percent of government revenue, most of from AIDS began declining in about 2003 (World Bank
it income from the accumulated saving of mineral revenues. 2009), but the costs and consequences of the epidemic will
At times, the overall fiscal surplus rose to more than 30 per- persist for many years.
cent of GDP (Maipose n.d.), and public sector investment
(as noted) has been consistently about 10 percent of GDP. If
CONCLUSION
investment is defined more broadly to include human cap-
ital and health, public investment rises to about 20 percent Landlocked Botswana seems to have defied the odds by
of GDP, which meets the IMF benchmark for sustainability creating a successful economy. Poverty has been reduced,
(Basdevant 2008). The government has used this accumu- education has become more widespread, and health indi-
lated savings to smooth expenditure over the business cators had improved before the HIV/AIDS epidemic undid
cycle, thus providing the economy with a short-run shock some of that progress.
absorber. However, as argued in box 4.1, the accumulated The country’s vast natural resources played a key role in
saving is also a long-run shock absorber, because it has this accomplishment, but the mere endowment of resources
built up human and physical capital and financial assets is clearly not the whole story. In much of Africa—and in
that should serve the economy well during the adjustment other parts of the world—natural resources have not always
period. In addition, the accumulated foreign assets man- been conducive to growth and development; in many cases
aged by the Bank of Botswana now amount to about they seem to have brought out the worst in countries, in the
40 months of imports. form of conflict and predatory governments.
These advantages notwithstanding, Botswana faces seri- Studying the effect of mineral wealth on economic out-
ous challenges. Dependence on mineral exports is the key comes is timely, because the increase in the prices of natu-
weakness of the current economic outlook and diversifica- ral resources as a result of the rise of China and India is
tion of the traded goods sector is the most important policy likely to result in windfalls for many African countries.
objective. Accumulated foreign assets and the income from How can countries turn these windfalls into long-run
them can ease the welfare cost of adjustment, but it cannot growth and development? In Botswana’s case, the key to
eliminate it. Barring a miraculous jump in productivity, successfully harnessing natural resources lay in good gover-
some real depreciation of the pula and the consequent hard- nance and good policies. Governance has not been perfect

88 CHAPTER 4: BOTSWANA’S SUCCESS: GOOD GOVERNANCE, GOOD POLICIES, AND GOOD LUCK
in Botswana, but it has been good. Botswana has been was defense expenditure, which average about 11 percent of
largely free of kleptocracy and civil conflict; it has main- total expenditure. While this is high relative to most coun-
tained a transparent, law-abiding government; and it has tries Botswana arguably had good reason to do this.
implemented good policies, including a hyper-prudent fis- 8. World Bank cited in Lange and Wright, 2002, 32.
cal policy, which has done much to diversify foreign 9. World Bank 2009. Botswana, however, lags other upper
exchange earnings and prevent the volatility that typifies middle income countries in these categories.
many resource-based economies. Investments in human 10. Botswana has not been able to escape the effects of the
and physical capital and vast improvements in infrastruc- current global crisis: GDP declined 4 percent in 2009, and
ture have also raised Botswana’s productivity, which, for the first time in many years, Botswana had fiscal and
together with its substantial financial reserve in the form of balance of payments deficits. However, the economy
foreign assets, should help ease the transition to a more appears to have weathered the worst, with the Bank of
Botswana projecting real growth of more than 3 percent for
diversified economy.
2010 and 2011.
11. Mining’s share fell in 2009 because of the slump in
NOTES world trade, not because of the declining importance of the
1. For discussions of Botswana’s success, see Maipose sector.
(n.d.), Acemoglu and Johnson (2003), and Leith (2005). 12. Between 2005 and 2008, before the effects of the global
2. Leith (2005) notes that when measured accurately, tak- slump set in, exports grew by about 40 percent. The share
ing into account social, health, and educational services pro- of diamonds fell from about 75 percent in 2005 to about
vided by the government to the poor, this measure falls to 65 percent in 2008. Total mining remained more or less
about 0.53. No measure of inequality is without serious ana- constant, however, at about 85 percent of total exports. The
lytical problems; the Gini coefficient is probably the best share of diamonds fell because of rising copper and nickel
available and most widely used index. The coefficient ranges exports (Botswana Central Statistical Office 2009).
from 0 (perfect equality) to 1 (perfect inequality). The 13. Other areas of interest are services, including finan-
higher the index, the more unequal the society. An index cial services; downstream diamond processing and trad-
above 0.5 is thought to denote an unequal distribution ing; and tourism, which represented about 10 percent of
(Leith 2005). GDP in 2008 and is a potential growth sector for export
3. See, for example, Auty (2004) and Sachs and Warner earnings.
(1995). For a contrary view, see Stevens (2003) and 14. FDI rose again in 2008. It is too early to tell whether this
Brunnschweiler and Bulte (2008). See also Lederman and is a trend.
Maloney (2007). 15. Botswana had the best ranking in Africa on Trans-
4. The analysis is based on the idea that some colonial parency International’s corruption perception index in 2011
regimes were mostly “extractive”—that is, the regime and ranked 33rd worldwide. (See http://www.transparency
existed to reap the maximum out of the colony’s economy. .org/policy_research/surveys_indices/cpi/2010.) On the
Other regimes, usually ones in which there were more colo- World Bank’s 2010 Ease of Doing Business index, Botswana
nial settlers, were less extractive. They often established ranked 3rd regionally but only 52nd worldwide, indicating
institutions that put more constraints on the extractive and some room for improvement (see http://www.doingbusiness
arbitrary powers of government. These constraining institu- .org/rankings).
tions often carried over into independence.
5. All of the high-performing developing countries men-
REFERENCES
tioned have established strong democratic institutions. The
question therefore arises whether the causality runs the Acemoglu, D., S. Johnson, and J. A. Robinson. 2003. “An
other way, from growth to democracy. It can also be argued African Success: Botswana.” In In Search of Prosperity:
that democracy is necessary to sustain a high-income econ- Analytical Narratives on Economic Growth, ed. D. Rodrik.
omy (see Barro 1997). Princeton, NJ: Princeton University Press.
6. Isham and Kaufmann (1999) estimate this proportion Acemoglu, D., and J. A. Robinson. 1999. “On the Political
(10 percent of GDP) to be the likely ceiling for public invest- Economy of Institutions and Development.” American
ment to remain productive. Beyond this they find public Economic Review 91 (4): 938–63.
sector investment is likely to be detrimental to growth. Alexeev, Michael, and Robert Conrad. 2009. “The Elusive
7. Lange and Wright (2002) point out that the only single Curse of Oil.” Review of Economics and Statistics 91 (3):
“nonproductive” item of comparable weight in the budget 586–98.

CHAPTER 4: BOTSWANA’S SUCCESS: GOOD GOVERNANCE, GOOD POLICIES, AND GOOD LUCK 89
 Auty, R. (ed.) 2001. “Resource Abundance and Economic Implications.” NBER Working Paper 9424, National
Development.” Wider Studies in Development. Oxford Bureau of Economic Research, Cambridge, MA.
University Press, Oxford, UK. Isham, J., and D. Kaufmann. 1999. “The Forgotten Rationale
Bank of Botswana. 2008. Annual Report. Gabarone. of Policy Reform: The Productivity of Investment Pro-
——. 2010. Botswana Financial Statistics. Gaborone. http:// jects.” Quarterly Journal of Economics 116 (1).
www.bankofbotswana.bw/index.php/content/2009 Lange, G., and M. Wright. 2002. “Sustainable Development
110615035-botswana-financial-statistics. in Mineral Economies: The Example of Botswana.”
Barro, R. 1997. Determinants of Economic Growth: A Cross- Ceepa Discussion Paper Series, University of Pretoria,
Country Empirical Study. Cambridge, MA: MIT Press. Pretoria, South Africa.
Basdevant, O. 2008. “Are Diamonds Forever? Using the Per- Lederman, D., and W. F. Maloney, eds. 2007. Natural
manent Income Hypothesis to Analyze Botswana’s Resources: Neither Curse nor Destiny. Washington, DC:
Reliance on Diamond Revenue.” IMF Working Paper, World Bank.
International Monetary Fund, Washington, DC. Leith, J. C. 2005. Why Botswana Prospered. Montreal:
Botswana Central Statistical Office. 2009. External Monthly McGill-Queen’s University Press.
Trade Digest January. Gaborone. http://www.cso.gov.bw/. Maipose, G. n.d. Policy and Institutional Dynamics in Sus-
Brunnschweiler, C. N. and E. H. Bulte. 2008. “The Resource tained Growth in Botswana. University of Botswana,
Curse Revisited and Revised: A Tale of Paradoxes and Gaborone.
Red Herrings.” Journal of Environmental Economics and OECD (Organisation for Economic Co-operation and Devel-
Management 55: 248–64. opment). n.d. Africa Economic Outlook. http:// www
Collier, P., and A. Hoeffler. 2004. “Greed and Grievance in .africaneconomicoutlook.org/en/countries/southern-
Civil War.” Oxford Economic Papers 56, Oxford Univer- africa/botswana/.
sity, Oxford, UK. Sachs, Jeffrey D., and Andrew M. Warner. 1995. “Natural
———. 2005. “Democracy and Resource Rents.” Oxford Resource Abundance and Economic Growth.” NBER
University, Oxford, UK. Working Paper W5392, National Bureau of Economic
Corden, W. M., and P. J. Neary. 1982. “Booming Sector and Research, Cambridge, MA.
Deindustrialization in a Small Open Economy.” Eco- Stevens, P. 2003. “Resource Impact: Curse or Blessing? A Lit-
nomic Journal 92: 825–48. erature Survey.” Journal of Energy Literature 11 (1): 3–42.
Hausmann, R., and R. Rigobon. 2003. “An Alternative Inter- World Bank. 2009. Country Partnership Strategy for the
pretation of the ‘Resource Curse’: Theory and Policy Republic of Botswana. Washington, DC.

90 CHAPTER 4: BOTSWANA’S SUCCESS: GOOD GOVERNANCE, GOOD POLICIES, AND GOOD LUCK