ArticlePDF AvailableLiterature Review

Abstract and Figures

Malaria and poverty are intimately connected. Controlling for factors such as tropical location, colonial history, and geographical isolation, countries with intensive malaria had income levels in 1995 of only 33% that of countries without malaria, whether or not the countries were in Africa. The high levels of malaria in poor countries are not mainly a consequence of poverty. Malaria is geographically specific. The ecological conditions that support the more efficient malaria mosquito vectors primarily determine the distribution and intensity of the disease. Intensive efforts to eliminate malaria in the most severely affected tropical countries have been largely ineffective. Countries that have eliminated malaria in the past half century have all been either subtropical or islands. These countries' economic growth in the 5 years after eliminating malaria has usually been substantially higher than growth in the neighboring countries. Cross-country regressions for the 1965-1990 period confirm the relationship between malaria and economic growth. Taking into account initial poverty, economic policy, tropical location, and life expectancy, among other factors, countries with intensive malaria grew 1.3% less per person per year, and a 10% reduction in malaria was associated with 0.3% higher growth. Controlling for many other tropical diseases does not change the correlation of malaria with economic growth, and these diseases are not themselves significantly negatively correlated with economic growth. A second independent measure of malaria has a slightly higher correlation with economic growth in the 1980-1996 period. We speculate about the mechanisms that could cause malaria to have such a large impact on the economy, such as foreign investment and economic networks within the country.
Content may be subject to copyright.
The Economic Burden of Malaria
John Luke Gallup and Jeffrey D. Sachs
CID Working Paper No. 52
July 2000
Copyright 2000 John Luke Gallup and Jeffrey D. Sachs and the
President and Fellows of Harvard College
at Harvard Universit
Center for International Develo
Working Papers
John Luke Gallup and Jeffrey D. Sachs
CID Working Paper No. 52
July 2000
Malaria and poverty are intimately connected. Controlling for factors such as tropical location, colonial
history, and geographical isolation, countries with intensive malaria had income levels in 1995 only 33% of
countries without malaria, whether or not the countries were in Africa.
The high levels of malaria in poor countries are not mainly a consequence of poverty. Malaria is very
geographically specific. The ecological conditions that support the more efficient malaria mosquito vectors
primarily determine the distribution and intensity of the disease. Intensive efforts to eliminate malaria in
the most severely affected countries in the tropics have been largely ineffective. Countries that have
eliminated malaria in the past half century have all been either subtropical or islands. These countries’
economic growth in the five years after eliminating malaria has usually been substantially higher than
growth in the neighboring countries.
Regressions using cross-country data for the 1965-90 period confirm the relationship between malaria and
economic growth. Taking into account initial poverty, economic policy, tropical location, and life
expectancy among other factors, countries with intensive malaria grew 1.3% less per person per year, and a
10% reduction in malaria was associated with 0.3% higher growth. Controlling for many other tropical
diseases does not change the correlation of malaria with economic growth, and these diseases are not
themselves significantly negatively correlated with economic growth. A second independent measure of
malaria has a slightly higher correlation with economic growth in the 1980-1996 period.
The paper concludes with speculation about the mechanisms that could cause malaria to have such a large
impact on the economy, such as foreign investment and economic networks within the country.
Keywords: malaria, economic cost of disease, economic growth, burden of disease, tropical disease
JEL Codes: I12, O11, O40
John Luke Gallup is a Research Fellow at the Center for International Development, an Institute Associate at the
Harvard Institute for International Development, and a Lecturer in the Department of Economics at Harvard
University. His recent research focuses on economic geography, malaria, and the relationship between economic
growth and poverty alleviation.
Jeffrey D. Sachs is the Director of the Center for International Development and Galen L. Stone Professor of
International Trade at the Department of Economics at Harvard University.
Poverty and malaria
Malaria and poverty are intimately connected. As Nobel Laureate in Medicine T.H. Weller (1958: 497) noted, “It
has long been recognized that a malarious community is an impoverished community.” Weller could have said the
same for malarious countries. Malaria is most intractable for countries in the poorest continent, Africa. The only
parts of Africa free of malaria are the northern and southern extremes, which have the richest countries on the
continent. India, the country with the greatest number of poor people in the world, has a serious malaria problem.
Haiti has the worst malaria in the Western Hemisphere, and it is the poorest country in the hemisphere.
Malaria risk has always been very geographically specific, as shown in Figure 1. Widespread malaria is confined to
the tropical and subtropical zone. Poverty is also geographically specific. As shown in Figure 2, poor countries
predominate in the same regions as malaria. Almost all of the rich countries are outside the bounds of malaria risk.
A basic problem when studying the macroeconomic impact of malaria is the lack high quality data on malaria
incidence or prevalence in the most severely affected countries. This study uses an index of malaria prevalence
derived from historical maps of the geographical extent of high malaria risk shown in Figure 1 (digitized from
Pampana and Russell, 1955, WHO 1966, and WHO 1997). Combined with detailed data on the world population
distribution (Tobler and others, 1996), one can estimate the fraction of the population in high malaria risk areas in
each country. Since most malaria mortality and severe morbidity is due to one of the four malaria species, the
malignant Plasmodium falciparum, the index of malaria intensity used in this paper is the fraction of the population
at risk of malaria multiplied by the fraction of malaria cases that are falciparum malaria (from WHO, 1990).
the comparative statistics in this section, intensive malaria is defined as having a malaria index greater than 0.5.
Since we completed an earlier version of this paper, McCarthy and others (1999) have used recently released World
Health Organization (WHO) data for malaria incidence data to estimate the impact of malaria on economic growth.
The WHO data are typically least accurate in countries where a large proportion of the population is infected with
malaria, and because of doubts about the comparability of the data across countries, we have not used the data in this
study. McCarthy and others (1999) use these data with a methodology similar to ours, but a different time period,
and they find somewhat smaller estimated effects of malaria than we find here.
Take the 150 countries with populations over one million in 1995, which account for over 99 percent of the world’s
population. Forty-four of the 150 countries, or 29 percent, have intensive malaria. Thirty-five of these 44 countries
are in Africa. The average purchasing-power parity gross domestic product (GDP) per capita in 1995 for the
malarial countries was $1,526, compared to an average income of $8,268 in the countries without intensive malaria,
more than five times higher.
Ranking the 150 countries by income per capita, all but three of the 44 countries with
intensive malaria are in the bottom half of the ranking. The exceptions are Oman and Gabon, ranked 34
and 41
which owe their wealth to oil, and the Philippines, which is barely in the top half with a rank of 74
out of 150. Of
the 119 poorest countries, all but twelve have some incidence of malaria or are recovering from socialism. The
richest 31 countries are entirely free of malaria.
Not only are malarial countries poor, but economic growth in malarial countries over the past quarter century has
been dismal. Growth of income per capita from 1965 to 1990 for countries with intensive malaria has been 0.4%
per year, while average growth for other countries has been 2.3%, over five times higher.
More than a third of the
countries with intensive malaria (11 out of 29) had negative growth from 1965 to 1990.
The question is whether these dramatic correlations mean that malaria causes poverty and low growth. We will
address this question in three ways. First, we consider the correlation of malaria with income levels after controlling
A second index of malaria derived from completely different data is described and used below.
According to the WHO (1999, p. 265) the data “permit only very limited comparison between countries or even
between various periods for the same country.” In other words, the data contain substantial measurement error
which may explain the smaller estimated impact of malaria in McCarthy and others (1999) than we find.
The 1995 PPP GDP data are from World Bank (1998), supplemented by CIA (1996, 1997) for countries without
World Bank data.
The data for GDP growth from 1965 to 1990 for 95 countries are from the Penn World Tables, Mark 5.6 (Summers
and Heston, 1994).
for other factors that are likely to affect the world distribution of income, such as geography, history, and policy.
Second, we discuss the determinants of malaria risk. Unlike other important diseases in poor countries caused by
deficient living conditions, like diarrhea, tuberculosis, and schistosomiasis, malaria is not a direct consequence of
poverty; its extent and severity are largely determined by the climate and ecology. Third, we explore the impact of
malaria on subsequent economic growth. This provides the most direct evidence on the continuing importance of
malaria as a cause of poverty.
Malaria and Income Levels
The coincidence of intensive malaria and low incomes could be due to many factors besides malaria itself. It could
be a general effect of the tropics, caused by poor soils, low agricultural productivity, or tropical diseases other than
malaria. It may capture geographical trade barriers, since many malarial countries are landlocked and far from the
major centers of world trade. It could be an accident of history. Many malarial countries were colonies until
recently, and the misfortunes of colonization may linger, keeping incomes low. Malaria could simply be a proxy
for Africa, which may be poor for other reasons such as weak institutions, poor economic policies, or ethnic conflict.
There are strong geographical patterns to income levels around the world.
As shown in Regression 1 of Table 1,
just four geographical variables account for almost half the variation in the log of GDP per capita income levels in
A country’s accessibility to the coast, measured by the share of the population within 100 kilometers of the
coast, is an important indicator of success in foreign trade and integration into the global economy, and hence
related to high income levels. Another measure of accessibility, the minimum distance to the core world markets
(New York, Rotterdam, and Tokyo), is inversely related to higher incomes. Thirdly, resource deposits, proxied by
the log of hydrocarbon reserves per person, are higher in wealthier countries. Lastly, the tropics, measured by the
percentage of a country’s land area in the geographical tropics, is much poorer than the rest of the world. The
“penalty” for being tropical was –0.68, signifying that tropical areas had only 51% (=exp(-0.68)) of the per capita
income of non-tropical areas controlling for other factors.
The next three regressions in Table 1 add the malaria index to the geographical correlates of income per capita in
two different years, 1950 and 1995.
The malaria index has a strong negative association with income levels after
controlling for the other four geographical factors. Malaria’s coefficient increases slightly from 1950 to 1995,
suggesting that if anything, malaria has become more important for explaining income levels over time. The
association of income with malaria dominates the association with the tropics, which loses statistical significance in
the regressions. The coefficient on malaria of -1.22 in 1995 implies that malarial countries have a per capita income
only 30% as high as non-malarial countries.
Regression 4 of Table 1 includes indicators for former colonies, and socialist countries in the post-World War II era.
These new explanatory variables are strongly associated with lower income levels, but taking them into account
does not substantially alter the correlation of malaria with low incomes. Regression 5 adds a measure of economic
policy, trade openness in the 1965 to 1990 period, and an index of the quality of government institutions. Malaria’s
association is still unaffected, but the socialist and colony variables lose their significance. If malaria is excluded
from this regression, income levels are significantly lower in countries that have been colonized, which suggests that
the economic weakness of countries with malaria might have been a factor in their colonial subjugation.
The final regression in Table 1 excludes the sub-Saharan African countries. Malaria has just as strong association
with poverty outside of Africa as for the whole world. Malaria is clearly distinguishable from other problems faced
by Africa.
See Gallup and Sachs (1998) for a wider investigation of the role of geography in economic development, and
more detailed explanation of the variables used in this section.
Regression analysis is explained in any econometrics textbook, such as Greene (1997).
The data for PPP GDP per capita in 1950 in Table 1 are from Maddison, 1995. The PPP GDP per capita data for
1995 are as in footnote 2.
Geography, history, and policy all have clear correlations with income levels, but taking them into account does not
alter the pattern of lower incomes in malarial countries. The association of malaria with poverty seems to be more
than just a mask for other plausible causes of low income.
Cause or Effect?
Malaria is prevalent in the poorest countries. Could this be a consequence, rather than a cause, of poverty? Many
other serious diseases predominantly found in poor countries clearly are a direct consequence of poverty, caused by
inadequate sewage treatment, unsafe drinking water, poor hygiene, or substandard housing. Malaria, though, does
not follow this pattern – it’s severity, and the difficulty in controlling it, are determined mainly by climate and
ecology. Personal behavior, such as use of screens and bednets, and the general level of development, especially
urbanization, also affect malaria prevalence, but they are not the main determinants.
Certain countries with very high incomes still face serious malaria problems due to their geographical location.
Oman, with an income per capita of almost $10,000, has intensive malaria throughout the country except in remote
areas of high altitude and desert. United Arab Emirates, next door with one of the highest income levels in the
world, also has been unable to eliminate malaria.
Successful elimination of malaria through vector control requires a well-run organization and financial resources.
However, the determining factor in where malaria has been eliminated in the post-war era has not been institutional
or financial. It has been the susceptibility of malaria and the vector to control. Figure 1 shows that since 1946
malaria has only been eliminated in nontropical regions and certain islands where it foothold is much weaker.
Coluzzi (1999) writes “Above all, it should be stressed that malaria eradication [in temperate areas in the late nine-
teen forties and early 1950s] was [only] achieved within more or less marginal ecoepidemiological zones,
particularly for P. faliparum” due to seaonality of malaria transmission, low nighttime outdoor temperatures, and
less efficient malaria vectors in temperate regions.
The large differences in the difficulty of controlling malaria in various climatic zones is supported by Table 7.
Those regions with the worst malaria in 1965 had the least reduction in malaria in the following three decades.
Countries with a predominantly humid tropical climate actually saw an increase in the malaria index. Although the
absolute reduction in the malaria index in temperate countries was lower than in other climatic zones, that is because
the malaria level is bounded by zero – malaria as measured by the malaria index was completely eliminated in
temperate countries.
Some of the most effective control efforts historically in the worst affected areas have used very few material
resources other than labor, so they are not constrained by poverty per se. The elimination of breeding sites for
malarial mosquitoes in parts of Panama by Gorgas at the time the canal was built (McCullough, 1977), the control of
the outbreak of Anopheles gambiae mosqitoes in northeastern Brazil in the 1930s (Soper and Wilson, 1943), and the
malaria-free enclaves around some African mines (Watson, 1953) show what is possible with a combination of
complete monitoring of all open water sources inside and outside households, drainage of wetlands, and a military
precision in all operations. Unfortunately, such control efforts have never been sustained in more than small areas
or for more than short periods of time.
In addition to differences in malaria intensity due to climate, the world distribution of Anopheles mosquitoes, the
malaria vectors, have a major impact on malaria prevalence and severity. Vectorial capacity is a measure of the
efficiency with which mosquitoes carry malaria from one human to another, an estimate of the number of secondary
cases of malaria generated by one primary case. The vectorial capacity of different species of Anopheles varies by
The major efforts devoted to malaria control in the building of the Panama canal and at African mines
demonstrates the economic impact of malaria on workers. Malaria mortality was a major factor in the French failure
to complete the canal (at least 20,000 people lost in nine years), and the American efforts were not effective until
malaria and yellow fever were brought under control (McCullough, 1977). Some tropical African mines created a
cordon sanitaire around their operations where African workers could not regularly leave or enter. The large
investments in monitoring, drainage and housing could only have been justified by higher worker productivity in the
malaria-free mines.
orders of magnitude. By far the most efficient vector, Anopheles gambiae, is exclusively found in sub-Saharan
Vectorial capacity has a major impact on the feasibility of controlling or eradicating malaria in a region.
Consequently, malaria eradication through vector control has been orders of magnitude more difficult in sub-
Saharan Africa. According to a recent expert committee report: “The epidemiology of malaria is driven by the
dynamics of the mosquito vectors. Thus, 90% of the world’s malaria is in Africa because it is home to the three
most effective vectors.” (NIH, 1997, p.12) Not only do the mosquito species determine the intensity of
transmission, but they also affect the mix of malaria between the malignant P. falciparum, and the less serious P.
vivax, P. malariae, and P. ovale. Africa is also the only major region of the world where falciparum malaria
Malaria control in sub-Saharan Africa has been a non-starter. There has been no successful malaria control of large
regions outside of the temperate southern tip, the controlled environment of some mining camps, and a few islands.
In response to the failure of WHO vector control projects in Cameroon, Nigeria, and elsewhere in Africa in the
1960s, the WHO sponsored an intensive malaria control and research project in the district of Garki, Nigeria
(Molineaux and Gramiccia, 1980). No resources, manpower, or institutional support were spared. Over the course
of seven years, WHO and the Nigerian government spent more than $6 million to try to eliminate malaria in 164
villages and compare the changes to control villages. Insecticide spraying of every hut at least every 10 weeks
during the course of the study had an average coverage of 99%. A third of the villages were also given mass drug
administration as a prophylaxis against malaria.
The intensity of malaria transmission in Garki was “very high indeed”. During the wet season, a person in this
district would be bitten on average 174 times per night by the Anopheles gambiae s.l.malaria vector and 94 times
per night by the Anopheles funestus vector.
The vectorial capacity, or the transmission rate of malaria between
persons through the vectors, reached 2000 times the critical value required to maintain endemic malaria, with a
range of between 18 and 145 malaria-transmitting bites per person per year in the eight villages studied (Molineaux
and Gramiccia, 1980, p. 107). In lay terms, everyone was constantly reinfected with malaria.
The vector control efforts reduced the man-biting rate of mosquitoes in the Garki villages by 90% from their pre-
study level, but despite this huge reduction in mosquito density, there was no significant change in the parasite rate
among the villagers. The control efforts were defeated by the vectorial capacity of the mosquitoes, which vastly
exceed what was required to maintain transmission of malaria. The conclusions of the study show that the failure to
control malaria in similar environments was not the consequence of poverty or lack of institutional capacity. This
was summed up in a conference paper summarizing the Garki study: “The malaria control measures employed in the
Garki Project failed to have a significant overall impact on malaria transmission, suggesting that these measures are
unlikely to be of long-term use in the African dry savannah belt.” This failure occurred despite the fact that “at all
times during this study, it was known that the strategies employed were much too detailed and expensive for long-
term use in the study area.” (Loutan and others, 1981, p.15).
At least two biological factors explain the exceptional severity of malaria in Africa. The most efficient mosquito
vector and the most serious malaria strain both most likely came from Africa. The vector Anopheles gambiae s.s.
coevolved with humans in the Afrotropical rain forest. The development of African agriculture in forest clearings
resulted in the vector’s most important characteristic for malaria transmission: it almost exclusively bites humans
(Coluzzi, 1999). The explosive potential of the Anopheles gambiae vector for transmitting malaria in similar
climates elsewhere was shown by the introduction of the mosquito into Brazil in the late 1920s, which was luckily
brought under control soon enough to eliminate it (Soper and Wilson, 1943). The most pathogenic human malaria
species, P. falciparum, most likely originated in Africa, probably in the past 5000-10000 years with the onset of
agriculture (Coluzzi, 1999).
With no proven method of controlling malaria in sub-Saharan Africa and other areas of intense transmission, it is
difficult to argue that poverty effectively causes malaria, or determines the success of control efforts. A recent NIH
Such high biting rate estimates are not unusual. Robert and others (1991) estimates that a person in the Kou Valley
in Burkina Faso sleeping without mosquito protection (as most do) receives 158 bites by Anopheles gambiae per
night, with total mosquito bites of 35,000 per year.
(1997) report notes the intractable nature of malaria Africa: “The availability of anti-malaria measures, when
correctly integrated and applied without financial constraints, can probably cope successfully with the malaria
problem everwhere in the Tropics except in the Afrotropical region.” (emphasis added)
A different sort of evidence that malaria is a cause of poverty comes from evolution. In areas with the most severe
malaria today, sub-Saharan Africa and parts of the Middle East and India, many ethnic groups have developed a
partial genetic defense against the ravages of malaria: sickle cell trait. In some parts of Africa, this red blood cell
abnormality is carried by 25-30% of the population (Weatherall, 1984). The value of sickle cell’s protection against
malaria must be great because it comes at a very high cost: all children (in developing countries) who inherit the trait
from both mother and father die before the age of childbearing. The Garki project confirmed this cruel equilibrium
(Molineaux and Gramiccia, 1980, p.229). Sickle cell trait in Garki adults was much higher than in children due to
selective survival. The burden of malaria on human wellbeing must have been high indeed for such a mutation to be
Milder congenital blood diseases, such as thalassemia in parts of southern Europe and Asia, confer
some protection against malaria in regions where malaria is correspondingly less severe. These blood diseases
highlight the importance of the burden of falciparum malaria relative to other forms since they protect primarily
against P. falciparum infections (Luzzato, 1984).
The geographical specificity of malaria, the wide biological variation in the capacity of mosquito vectors, the
inability to control malaria in Africa under experimental conditions, and the persistence of fatal blood diseases as a
defense all point to a causation from malaria to poverty, not vice versa. Large-scale vector control projects require
resources, but if they were clearly feasible, the resources would probably be forthcoming from the international
community. Much of the effective malaria control (in subtropical areas) has in fact come from low-technology
drainage and larvaciding,
which could be carried out independently by a poor tropical country if the technique
offered a viable prospect of malaria control.
Anecdotes from countries that have eliminated malaria
A small number of the countries that had intensive malaria earlier this century eliminated the disease. Many other
changes were simultaneously occurring in the economies of these countries before and after eradication, but in
almost all cases for which we have data, the countries experienced an acceleration of growth immediately following
eradication, and faster growth than neighboring countries.
Malaria eradication in southern Europe has been a clear success story in the fight against malaria. Major control
efforts in Greece, Italy, and Spain were started in the 1930s and completed in the late 1940s. Greece up to that time
had been the most malarial country in Europe, in peak years totalling a quarter of the total population (Kamarck,
1976, p. 12). Jones (1909) argues that the spread of faliparum malaria through most of Greece in the first millenium
AD was the main factor in the decline of ancient Greek civilization. Greece was the site of major malaria epidemics
in the nineteenth and early twentieth century, and the famed plain of Marathon became virtually uninhabited due to
malaria despite very fertile soils. The use of DDT starting in 1946 had spectacular results (having a major influence
on the subsequent WHO world eradication campaign), with malaria falling from 1-2 million cases per year in the
early 1930s to only 5,000 in 1951 (Bruce-Chwatt and de Zulueta, 1980). Although complete eradication would take
another twenty years, partly due to vector resistance to DDT, from an economic point of view malaria was under
Sickle cell trait also shows the role of climate in determining the relative burden of disease in different regions of
the world. “The distribution of sickle cell trait in tropical Africa corresponds almost exactly to the areas of tropical
rain forest” (Carlson, 1984, p.29).
Many ethnic groups in Africa also have complete protection from P. vivax malaria due to a blood characteristic
called the Duffy factor, which makes vivax malaria rare in Africa. Although this suggests that vivax malaria is also
burdensome, it does not demonstrate that the human burden is large because the Duffy factor causes no mortality in
people who carry it. Africans could easily maintain the Duffy factor in the face of evolutionary selection even with
a low disease burden from vivax malaria.
See Kriton and Spielman (1989) on the major role of these simple technologies in many of the successful
eradication efforts.
The longstanding problem of malaria in Italy contributed to the major role of Italians in early malaria research. Just
before the control campaign, Italy had over 300,000 cases of malaria per year with about 20,000 deaths (Haworth,
1988). The Pontine Marshes south of Rome were rendered uninhabitable by the disease. P. falciparum was
eliminated by the end of the 1940s, with P. vivax and P. malariae disappearing more slowly.
Spain reported 400,000 cases of malaria with 1700 deaths in 1943 (Haworth, 1988), but had effectively controlled
the disease by the end of the 1940s.
The period immediately before effective control of malaria was wartime and the post-war reconstruction. Due to the
anomalies of the period and the lack of data, we compare growth in the post-control years of 1950-1955 to growth in
the period 1913-1938 in Table 2.
In all three countries, economic growth in the post-control period was much
higher than in the prewar period, and higher than growth in rest of Western Europe 1950-1955. In the prewar
period, Greece and Italy also grew somewhat faster than Western Europe, but the increment in growth over the
European average was also higher in the post-control period than the prewar period.
Portugal was another southern European country with intensive malaria (over 100,000 cases per year in the 1940s –
Haworth, 1988) that controlled malaria later than Greece, Italy and Spain. As shown in Table 3, growth accelerated
after eradication in 1958 compared to the period before eradication, and once again the increment of growth over the
average in the rest of Western Europe increased after eradication.
There are, unfortunately, few success stories for malaria eradication in developing countries, but the islands of
Taiwan and Jamaica are among the few. Tables 4 and 5 show that growth accelerated in the two countries after
eradication, in 1961 for Taiwan and 1958 in Jamaica. In both cases, growth also increased by more than growth in
their respective regions.
The U. S. South was still very malarious before World War II. 135,000 cases of malaria with 4000 deaths were
reported in 1935 (Haworth, 1988). After large-scale drainage projects by the W.P.A. in the 1930s were followed by
insecticide spraying after the war, malaria was brought under control by the end of the 1940s. In the decade of the
1950s, the South had its most dramatic catch-up with the rest of the country, going from 60% of the income per
capita of the rest of the U.S. in 1950 to 68% in 1960 (calculated from Barro and Sala-i-Martin, 1991).
An exception to prove the rule is Mauritius. A small island off the coast of East Africa, Mauritius was first exposed
to malaria in 1865 with castrophic results. In one year, 1867, between an eighth (Verdrager and others, 1964) and a
quarter (Ross, 1910, quoted in Soper and Wilson, 1943) of the total population died in the malaria epidemic.
Malaria was finally eliminated in 1963. Economic growthi n a small, closed, sugar-producing economy continued to
be negative until 1973 when Mauritius opened its economy, built export processing zones, and took off
economically. Countries do not become prosperous by controlling malaria alone, but the dramatic success of
Mauritius in become a manufacturing exporter since 1973 was certainly made easier by eliminating malaria.
Malaria control within regions of some other countries has had dramatic impacts on agricultural output and
settlement patterns. “Until malaria was wiped out [in Corsica], no one farmed [on the eastern plain]. Today this
plain accounts for 60 percent of Corsica’s agricultural production.” (Kamarck, p. 12) The southern plains of Nepal,
the Terai, were virtually uninhabited until the early 1950s because of malaria. It is now the richest and most
agriculturally productive part of the country (Kamarck, p. 12).
These country examples of growth after the control of malaria are merely suggestive. In every country examined
economic growth was higher immediately after the eradication of malaria, but there were surely many other factors
that influenced the economy at the same time. In several of the countries (Greece, Spain, and Jamaica), the rapid
development of the tourism industry was only possible because of malaria eradication. Few tourists thought of
basking on shores of the Aegean when Greece was the most malarial country in Europe.
GDP data for the 1913-1938 period are from Maddison, 1995. All other country GDP data in this section are from
Summers and Heston, 1994.
Malaria and economic growth
We have shown that most malarial countries are poor, and certain countries that managed to completely eliminate
malaria in recent times have had more rapid economic growth than their neighbors. But can we find any general,
statistically convincing evidence that initial malaria prevalence and reductions in malaria affect economic growth?
Would a reduction in malaria significantly improve the economic prospects of poor countries?
The most direct way to assess the causal effect of malaria on country economic performance is to look at the
relationship between economic growth, initial malaria levels and change in malaria over same period. Above we
saw that countries with intensive malaria in 1965 have had much lower economic growth in the subsequent 25 years,
but this did not take into account the initial poverty of countries, nor did it consider the role of human capital levels
(overall health and education), government policies, or geographical variables. After the role of human capital,
policy, and geography are taken into account, it is generally found that poorer countries grow faster than richer
countries (Barro and Sala-i-Martin, 1995), so if malaria were really just a proxy for poverty, one would expect
malarial countries also to grow faster.
Table 6 presents a cross-country empirical growth estimation in the style of Barro (1991). Growth in GDP per
capita over the 1965 to 1990 period is related to initial income levels, initial human capital stock, policy variables,
and geographical variables. Human capital stock is measured by secondary education and life expectancy at birth.
Policy is measured by trade openness over the period and an index of the quality of public institutions. The
geographical variables include an indicator for the geographical tropics and the fraction of the population within 100
kilometers of the coast.
To these well-researched predictors of economic growth we add the malaria index in 1965
in Regression 1 of Table 6.
Countries with intensive malaria in 1965 had much lower economic growth,
amounting to 1.3% lower growth per year, even after other factors like initial income level, overall life expectancy
and tropical location are taken into account.
Reductions in malaria over the 1965-1990 period, in addition to malaria levels in 1965, are associated with much
higher economic growth, as shown in Regression 2 of Table 6. This corresponds to a 0.3% rise in annual economic
growth for a 10% reduction in the malaria index. Over the twenty-five year period the average reduction in the
malaria index was 7% among countries that had malaria in 1965. By extrapolation far outside the observed sample
variation, a country with its whole territory affected by 100% P. falciparum malaria is predicted to raise
permanently its annual growth by 3.2% if it completely eliminates malaria! Unfortunately, no country came near to
accomplishing this. Of the fourteen countries in the sample with a malaria index above 0.9 in 1965, only one
reduced it significantly: the malaria index in Zimbabwe fell by one-third.
Economic growth itself might be a cause of the observed malaria reductions if economic growth made greater
resources available for malaria control, or other factors such as a high institutional capacity could be responsible for
both economic growth and successful malaria control. In this case the estimates of the effect of malaria reduction on
economic growth would be biased. To control for the possible endogeneity of malaria reduction, Regression 3 of
Table 6 uses instrumental variables (see, for example, Greene, 1997, pp. 288-295). The instruments are the
prevalence of 53 different Anopheles mosqito vectors in each country in 1952.
The different Anopheles mosqitoes
vary widely in their efficacy in transmitting human malaria, so that the distribution of Anopheles vectors is strongly
correlated with malaria intensity and its change (the first-stage instrumental variables regression of the change in the
malaria index on Anopheles vectors, which is not shown, has an R
of 0.51). There is no reason to think that the
distribution of malaria mosquito vectors is a cause of economic growth apart from the direct influence of malaria,
In fact, over the 1965 to 1990 period, poor countries on average grew slower than rich countries, but poor
countries also had lower initial human capital, followed less successful economic policies, and were disadvantaged
Gallup and Sachs (1999) give a more detailed description of these variables.
The malaria index for 1965 is constructed similarly to the malaria index for 1994, described above. It is the
product of the fraction of the population living in areas with high malaria risk in 1965 times the fraction of malaria
cases in 1990 that are P. falciparum. This assumes that the relative share of P. falciparum cases did not change
substantially from 1965 to 1990. The malaria data for 1965 were digitized from a WHO (1966) map.
The Anopheles data were digitized from American Geographical Society (1952) and used to calculate the percent
land area in each country affected by each Anopheles species.
making vector prevalence an ideal instrument for malaria change. After correcting for the possible endogeneity of
malaria reduction, the estimated effect on economic growth is essentially unchanged, so it is unlikely that the
changes in malaria prevalence are a consequence of economic growth. A Hausman (1978) test finds no significant
difference the ordinary least squares and instrumental variables estimates, rejecting the endogeneity of the change in
Regression 4 restricts the sample to non-sub-Saharan African countries. The size of the estimates for malaria are
substantially the same. The change of malaria has a statistically significant coefficient, but the estimate for initial
malaria loses statistical significance. Even without including the sub-Saharan African countries with the most
intensive malaria, a reduction in malaria corresponds to much higher economic growth.
Malaria could be a proxy for a range of tropical diseases that are not adequately controlled for by life expectancy.
One disease that is starting to have major economic impacts in many of the same countries with intensive malaria,
AIDS, is not relevant for the time period under study here. By 1990, the end of the period of economic growth
studied, the burden of AIDS was still sufficiently small to cause very minor economic impacts. Other major
diseases prevalent in the tropics that may be correlated with malaria are hookworm, onchocerciasis, schistosomiasis,
filariasis, dengue fever, and trypanosomiasis. We have indentified detailed maps of the geographical extent of all
these diseases except for trypanosomiasis from the 1950s, as well as data for ten other less important tropical
diseases providing measures of the extent of twenty different tropical diseases.
The land area affected by the
disease is weighted by detailed population distribution data to provide an estimate of the fraction of the population at
risk of each disease. Since the data preceed the period of economic growth under study, they show the impact of
initial disease on subsequent economic performance avoiding problems of reverse causation.
The large number of diseases make it impractical to include them all as independent correlates in the economic
growth regression due the limited sample of countries. To assess whether the other diseases were responsible for the
correlation of initial malaria with economic growth, we included each of the twenty diseases as an additional
regressor separately to the regression specification in Regression 1 of Table 6. The estimated impact of malaria was
remarkably stable across these twenty regressions with a point estimate range of just –0.7 to –1.3, and statistically
significant at the ten percent level in 17 of 20 regressions (not shown – available from the authors upon request).
Rather surprisingly, none of the other tropical diseases had a significant negative correlation with economic growth
even at the 10 percent level, after controlling for malaria in these regressions. A second way to combine the other
disease information is to estimate its principal components and include a linear combination of the other disease
variables in the growth regression. As shown in Regression 5 of Table 6, the first principal component of the
tropical diseases has a insignificant positive correlation with subsequent economic growth, and malaria has the same
significantly negative correlation with economic growth as in Regression 1 of Table 6. Controlling for a range of
other tropical diseases does not substantially affect the correlation of initial malaria with subsequent growth.
The malaria index, though it is the best measure of malaria burden we could construct, is admittedly crude. We have
also developed an alternative measure of malaria intensity, which although it is also crude, is derived from
completely different data sources, and covers a different time period. The alternative malaria indicator used
qualitative assessments of the severity of malaria from the WHO’s (1981) country-specific health advice for
travelers. The earliest descriptions of malaria in these advisories date from 1980, and the index is set equal to one
for countries in which malaria affects the whole country or the whole country except for major cities, and zero
otherwise. Using data on economic growth from 1980 to 1995 and corresponding data on covariates in Regression 6
of Table 6, the “WHO Advisory” malaria indicator for 1980 shows a significant negative correlation of initial
malaria with subsequent growth.
Countries with malaria throughout the country except for major cities had 1.6%
The other diseases are dengue fever, yellow fever, helminthiases (Paragoniumus westermani, Fasciolopsis buski,
Opisthorchis elineus, Diphyllobothrium sp., and Clonorchis sinensis), and leischmaniases (oriental sore, kala azar,
and various American forms of the disease). The schistosomiasis data are broken down into Schistosoma
haematobium and S. mansoni, and the filariasis data into Loa loa, Wuchereria Bancrofti, W. malayii,
Acanthocheilonema perstans, and Mansonella ozzardi, giving a total of 20 non-malaria tropical disease variables.
The data sources for the variables in Regression 6 of Table 6 that have not already been documented are the
World Bank (1998) for purchasing power parity GDP per capita 1980-1995, Barro and Lee (1993) for initial
secondary schooling for those aged fifteen and over, initial life expectancy at birth from United Nations (1996)
supplemented with government yearbook estimates for Taiwan.
lower growth in GDP per capita in the 1980-95 period. Using malaria data from a completely independent source
and assessing a different (though overlapping) time period from the other growth regressions, malaria still has a very
large and statistically significant correlation with economic growth.
A recent study of the macroeconomic impact of malaria (McCarthy and others, 1999) uses still different data sources
for measuring malaria, and a different period of economic growth (1983-1998), and different independent control
variables. The study finds a robust correlation between malaria and growth using WHO morbidity data, but of a
somewhat smaller magnitude than we find here: just over one quarter of a percent per year of economic growth for
about a quarter of the sample.
The growth regression results show that countries with intensive malaria in 1965 had dramatically lower economic
growth in the subsequent twenty-five years, after controlling for other factors that likely influenced growth, like
initial poverty, economic policy, initial health and education levels, and tropical location. Countries that managed to
reduce malaria over the period had much higher economic growth. These problems affected sub-Saharan Africa
most severely because malaria levels are highest there, but the same relationship with economic growth holds in the
non-African world. Using an independent malaria measure over a different time period shows a similar correlation
of malaria and economic growth.
Could malaria have such a large impact on economic growth?
We have presented several kinds of evidence suggesting that malaria has large economic effects. What are the
channels through which malaria could be a major drag on the economy?
The traditional medical view of malaria at its most severe, in holoendemic areas, is that malaria contributes
significantly to child mortality and can cause acute disease in pregnant women, but it does not have large effects on
the fitness of other mature adults due to their partial immunity acquired through constant reinfection. McGregor
(1988, p. 754) states this clearly: “in adult life...a host-parasite balance resembling commensualism is achieved.
Despite sustained infectious challenge, adults constitute an economically viable work-force capable of coping with
the strenuous physical activities that are required to maintain essential food supplies in subsistence agricultural
communities.” Though this view may be shared by many in the medical field, it has rarely been the subject of
careful research. One wonders if the medical focus on mortality and acute disease obscures a general debilitation
that could be caused by malaria. At least one article reports that long-term asymptomatic malaria may be the cause
of chronic pains and lassitude among Europeans in East Africa (Wilks and others, 1965).
Formidable methodological and measurement problems confront any assessment of the impact of malaria on
individuals and households in areas of stable malaria. There is not even a clear method for diagnosing which
individuals suffer from malaria. Virtually the whole population carries malaria parasites, and the density of
parasites is not a reliable measure of disease due to a variable immune response, which is still poorly understood.
Fever symptoms are not specific to malaria. If everyone is infected with malaria, there is no comparison group for
measuring the impact of malaria on diseased individuals relative to the healthy population.
If a clear measure of disease burden were available, one still faces the problem of assessing the cost of illness in
extended rural households, accounting for the compensating behavior of other household members. It is hard to
evaluate the cost of lost opportunities of household members who help out a person with malaria. Most attempts to
directly measure the lost work due to malaria (which ignore these problems) find small impacts (Chima and Mills,
1998; Malaney, 2000). Some recent studies have found larger measurable impacts of malaria at the household level
(Cropper and others, 1999, Audibert and others, forthcoming). A recent study of the impact of parental illness on
child health in Tanzania found that children with fathers or especially mothers who were ill with malaria had worse
health, whereas parental AIDS was not significantly related to child health (Dayton, 1999). However the difficulty
in measuring how much individuals actually suffer from malaria in an environment where most people carry malaria
parasites, and the myriad problems of measuring household response to debilitation make all the microeconomic
estimates incomplete.
Malaria has life-long effects on cognitive development and education levels through the impact of chronic malaria-
induced anemia and time lost or wasted in the classroom due to illness. The importance of these effects is
speculative, though, since their impact is virtually unstudied. Iron-deficiency anemia per se has been shown to
affect the cognitive skills of children as well as their cognitive abilities in later life (Pollit and others, 1989, and
Lozoff and others, 1991).
In short, the impact of malaria on the productivity of individuals in areas of stable malaria cannot be assessed with
the current state of research. Whether or not individuals are significantly debilitated by malaria, there are several
other channels through which malaria could have large impacts on the economy. The first is the impact of malaria
on foreign direct investment and tourism. Malaria, unlike diseases resulting from poverty, does not discriminate
between rich and poor victims. As long as malaria protection is imperfect and cumbersome, well-to-do foreign
investors and tourists may stay away from malarial countries. A second channel through which malaria may affect
the economy is limitation on internal movement.
The better educated and the ambitious who move to the largely
malaria-free cities lose their natural protection due to lack of exposure. They may be reluctant to maintain contact
with the countryside for fear of infection. Communities in unstable malaria areas may make people from stable
malaria areas unwelcome. In general, the transmission of ideas, techniques, and development of transportation
systems may all be stunted by malaria.
The location and severity of malaria are mostly determined by climate and ecology, not poverty per se. Areas with
intensive malaria are almost all poor and continue to have low economic growth. The geographically favored
regions that have been able to reduce malaria have grown substantially faster afterwards. The estimated impact of
malaria on economic growth, using two different measures of malaria, is very large, but the mechanisms behind the
impact are not clear.
It might be thought that malaria has a large impact in poor countries because of its interaction with malnutrition.
Malaria, along with other childhood infectious diseases, has been found to exacerbate malnutrition. Surprisingly,
though, malnutrition probably confers some protection against malaria. McGregor (1988, p. 763), in his survey,
finds that “the balance of available evidence indicates that malnutrition in humans is more commonly antagonistic to
We thank Andrew Spielman for suggesting this effect of malaria.
American Geographical Society. 1951. "Distribution of Malaria Vectors," The Geographical
Review 41(4), map.
American Geographical Society. 1952a. "Distribution of Helminthiases," The Geographical Review
42(1), map.
American Geographical Society. 1952b. "Distribution of Dengue and Yellow Fever," The
Geographical Review 42(2), map.
American Geographical Society. 1954. "World Distribution of Leischmaniases." New York:
American Geographical Society, map.
Audibert, Martine, Jacky Mathonnat, Innocent Nzeyimana, and Marie-Claire Henry. Forthcoming.
“Rôle du Capital Humain dans l’Efficience Technique des Producteurs de Coton du Nord
de la Côte-d’Ivoire,” Revue d’Economie du Developpement.
Barro, Robert J. 1991. "Economic Growth in a Cross-Section of Countries,Quarterly Journal of
Economics 106(2): 407-43.
Barro, Robert J. and Xavier Sala-i-Martin. 1991. “Convergence across States and Regions,”
Brookings Papers on Economic Activity 1: 107-182.
Barro, Robert J., and Xavier Sala-i-Martin. 1995. Economic Growth. New York: McGraw Hill.
Barro, Robert J., and Jong-Wha Lee. 1993. “International Comparisons of Educational
Attainment,” Journal of Monetary Economics 32:363-394.
Bruce-Chwatt, Leonard Jan, and Julian de Zulueta. 1980. The Rise and Fall of Malaria in Europe:
A Historico-Epidemiological Study. Oxford: Oxford University Press.
Carlson, Dennis G. African Fever: A Study of British Science, Technology, and Politics in West
Africa, 1787-1864. Canton, MA: Science History Publications.
Chima, Reginald Ikechukwu, and Anne Mills. 1998. “Estimating the Economic Impact of Malaria
in Sub-Saharan Africa: A Review of the Evidence,” mimeo, London School of Hygiene and
Tropical Medicine, June.
CIA. 1996. The World Factbook. Washington, D.C.: Central Intelligence Agency.
CIA. 1997. The World Factbook.
Coluzzi Mario. 1999. “The Clay Feet of the Malaria Giant and its African Roots: Hypotheses and
Inferences about Origin, Spread and Control of Plasmodium falciparum.” Parassitologia
Cropper, Maureen L., Julian A. Lampietti, Mitiku Haile, Christine Poulos, and Dale Whittingdon.
1999. “The Value of Preventing Malaria in Tigray, Ethiopia” manuscript presented at the
International Health Economics Association Conference in Rotterdam, June.
Dayton, Julia. 1999. “Does Parental Morbidity Affect Child Health? Evidence from Rural
Tanzania,” manuscript presented at the International Health Economics Association 2
World Conference, Rotterdam, Netherlands, June 6-9, 1999.
Gallup, John Luke, and Jeffrey D. Sachs. 1999. “Geography and Economic Development.” World
Bank Annual Conference on Development Economics 1998. Boris Pleskovic and Joseph E.
Stiglitz, eds., pp. 127-178. Washington, DC: The World Bank.
Greene, William H. 1997. Econometric Analysis. New York: Macmillan. Third Edition.
Hausman, J. 1978. “Specification Tests in Econometrics,Econometrica 46:1251-1271.
Haworth, J. 1988. “The Global Distribution of Malaria and the Present Control Effort,” in
Wernsdorfer, Walther H., and Ian McGregor, eds., Malaria: Principles and Practice of
Malariology. Edinburgh: Churchill Livingstone.
Jones, W. H. S. 1909. Malaria and Greek History. Manchester: University Press.
Kriton, Uriel, and Andrew Spielman. 1989. “Supression of Transmission of Malaria Through
Source Reduction: Antianopheline Measures Applied in Israel, the United States, and
Italy,” Reviews of Infectious Diseases 11(3):391-406.
Gallup, John Luke, Jeffrey D. Sachs and Andrew D. Mellinger. 1999. “Geography and Economic
Development”(with) in Pleskovic, Boris, and Joseph E. Stiglitz, eds., World Bank Annual
Conference on Development Economics 1998. Washington, DC: The World Bank.
Kamarck, Andrew M. 1976. The Tropics and Economic Development: A Provocative Inquity into
the Poverty of Nations. Washington, D. C.: World Bank.
Loutan, Louis, Letitia Acquah, and Gregg Davies. 1981. “The Garki Project.” Unpublished
conference paper available from the author.
Lozoff, B., E. Jimenez, and A. Wolf. 1991. “Long-term developmental outcome of infants with iron
deficiency,” New England Journal of Medicine 325:687-695.
Luzzato, Lucio. 1984. “Genetic Factors Modifying Tropical Disorders,” in Warren, Kenneth S. and
Adel A. F. Mahmoud, eds., Tropical and Geographical Medicine. New York: McGraw-
Malaney, Pia. 2000. “The Microeconomic Burden of Malaria,” manuscript. Harvard Center for
International Development.
Maddison, Angus. 1995. Monitoring the World Economy, 1820-1992. Paris: OECD.
McCarthy, Desmond F., Holger Wolf, and Yi Wu. 1999. “Malaria.” Unpublished manuscript. The
World Bank.
McCullough, David. 1977. The Path Between the Seas: The Creation of the Panama Canal, 1870-
1914. New York: Simon and Schuster.
McGregor, Ian A. 1988. “Malaria and Nutrition,” in Wernsdorfer, Walther H., and Ian A.
McGregor, eds., Malaria: Principles and Practice of Malariology. Edinburgh: Churchill
Molineaux, L., and G. Gramiccia. 1980. The Garki Project: Research on the Epidemiology and
Control of Malaria in the Sudan Savanna of West Africa. Geneva: World Health
NIH. 1997. Final Report, International Conference On Malaria In Africa: Challenges And
Opportunities For Cooperation. January 6-9, 1997 Dakar, Senegal. Available at
Pampana, E. J., and P. F. Russell. 1955. Malaria: A World Problem. WHO, Geneva.
Pollit, E., P. Hathirat, N. Kotchabharkdi, L. Missel, and A. Valyasevi. 1989. Iron deficiency and
educational achievements in Thailand,” American Journal of Clinical Nutrition
Robert, Vincent, Vincent Ouedraogo, and Pierre Carnevale. 1991. “La Transmission du Paludisme
Humain dans un Village au Centre de la Rizière de la Vallée du Kou, Burkina Faso,” in
Robert, Vincent, Jean-Philippe Chippaux, Louti Diomandé, and collaborators, Le
Paludisme en Afrique de l’Ouest: Etudes Entomologiques et Epidémiologiques en Zone
Rizicole et en Milieu Urbain. Paris: Editions de ORSTOM.
Ross, Ronald. 1910. The Prevention of Malaria. New York: E. P. Dutton.
Soper, Fred L., and D. Bruce Wilson. 1943. Anopheles gambiae in Brazil, 1930 to 1940. New
York: The Rockefeller Foundation.
Summers, Robert, and Alan Heston. 1994. The Penn World Tables, Mark 5.6.
Tobler, Waldo, Uwe Deichmann, Jon Gottsegen, and Kelly Maloy. 1995. The Global Demography
Project. Technical Report TR-95-6, National Center for Geographic Information and
Analysis, April.
United Nations. 1996. World Population Prospects 1950-2050 (The 1996 Revision). New York:
United Nations. Computer diskettes.
Verdrager, J., R. Mamet, S. Roche, and J. P. Klein. 1964. La Campagne d’Éradication du
Paludisme à l’Ile Maurice. Port Louis, Ile Maurice: Imprimery Officiel.
Watson, Sir Malcolm. 1953. African highway; the battle for health in Central Africa. London:
Weatherall, David J. 1984. “Common Genetic Disorders in the Tropics,” in Warren, Kenneth S. and
Adel A. F. Mahmoud, eds., Tropical and Geographical Medicine. New York: McGraw-
Weller, T.H. 1958. “Tropical Medicine,” in Encyclopedia Britannica. Chicago: William Bennet.
pp. 495-497.
WHO. 1966. “Malaria Eradication in 1965,” WHO Chronicle 20(8, August): 286-300.
WHO. 1981. Vaccination Certificate Requirements For International Travel and Health Advice to
Travellers. Geneva: World Health Organization.
WHO. 1992. “World Malaria Situation in 1990, Part II,” Weekly Epidemiological Record 67(23,
June 5): 169-74.
WHO. 1997. “World Malaria Situation in 1994, Part I,” Weekly Epidemiological Record 72(36,
September 5): 269-274.
WHO. 1999. “Malaria, 1982-1997,” Weekly Epidemiological Record 74(32, August 13): 265-270.
Wilks, N. E., P. P. Turner, K. Somers, and O. P. Markandya. 1965. “Chronic Ill-Health from
Unrecognized Malaria,” East African Medical Journal 42(11):580-583.
World Bank. 1998. World Development Indicators 1998 CD-ROM. Washington, D.C.: International
Bank for Reconstruction and Development.
Table 1. Level of GDP per capita
(1) (2) (3) (4) (5) (6)
Log GDP per capita
1995 1950 1995 1995 1995 1995 (non-
Population within 100km of coast (%) 1.26 0.80 0.57 0.65 0.33 0.40
(6.31)** (5.19)** (2.74)** (3.40)** (2.23)* (2.76)**
log Distance to major markets -0.35 -0.12 -0.33
(3.79)** (1.37) (4.03)**
log Hydrocarbons per person 0.01 0.01 0.01 0.01 0.00 0.00
(2.28)* (2.56)* (1.86) (2.13)* (1.36) (1.27)
Tropical land area (%) -0.68 -0.14 -0.23 -0.59 -0.09 -0.10
(3.97)** (0.89) (1.01) (3.04)** (0.59) (0.83)
Falciparum malaria index -1.17 -1.22 -1.16 -1.16 -1.10
(6.28)** (5.67)** (4.73)** (6.41)** (4.34)**
Socialist -0.80 -0.10 -0.05
(5.20)** (0.66) (0.30)
Colony -0.14 -0.05 -0.12
(2.18)* (0.89) (2.24)*
Trade openness (0-1) 0.50 0.43
(2.99)** (2.98)**
Quality of public institutions (0-10) 0.22 0.23
(6.85)** (7.82)**
Constant 10.50 8.54 10.91 8.75 7.15 7.15
(14.10)** (13.54)** (17.36)** (46.40)** (29.27)** (32.30)**
Observations 149 127 127 149 97 66
R-squared 0.47 0.59 0.62 0.62 0.88 0.88
Robust t-statistics in parentheses
* significant at 5% level; ** significant at 1% level
Table 2. GDP per capita growth before and after malaria eradication in southern European countries (late
GDP p.c. growth Difference w/W. Europe
1913-38 1950-55 1913-38 1950-55
Greece 2.1 3.6 1.1 1.3
Italy 1.0 5.3 0.1 3.0
Spain -0.4 6.2 -1.4 4.0
Western Europe 0.9 2.3 0.0 0.0
Table 3. GDP per capita growth before and after malaria eradication in Portugal (1958)
1953-58 1958-63 Change
Portugal 3.0 5.3 +2.3
Western Europe 1.9 3.8 +1.9
Difference +1.1 +1.5 +0.4
Table 4. GDP per capita growth before and after malaria eradication in Taiwan (1961)
1956-61 1961-66 Change
Taiwan 2.8 5.8 +3.0
East Asia 3.4 5.5 +2.1
Difference -0.6 +0.3 +0.9
Table 5. GDP per capita growth before and after malaria eradication in Jamaica (1961)
1956-61 1961-66 Change
Jamaica 3.4 4.1 +0.7
Central America and Caribbean 2.6 3.1 +0.5
Difference +0.8 +1.0 +0.2
Table 6. Growth of GDP
(1) (2) (3) (4) (5) (6)
1965-90 1965-90 1965-90
1965-90 1980-95
Log initial GDP per capita -2.6 -2.6 -2.4 -2.5 -2.3 -3.6
(8.07)** (7.90)** (7.54)** (6.36)** (8.04)** (7.95)**
Log initial secondary schooling 0.1 0.1 0.1 0.1 0.1 -0.2
(1.04) (0.90) (0.60) (0.62) (0.77) (0.62)
Log initial life expectancy 4.4 3.1 3.0 3.8 4.6 9.6
(4.46)** (3.41)** (3.51)** (2.34)* (4.19)** (3.44)**
Trade openness (0-1) 1.8 1.7 1.6 1.7 1.7 3.0
(4.91)** (4.91)** (4.51)** (4.14)** (4.55)** (5.10)**
Quality of public institutions (0-10) 0.4 0.4 0.3 0.4 0.3 0.6
(3.29)** (3.79)** (3.32)** (2.95)** (2.78)** (4.03)**
Tropical land area (%) -0.6 -0.6 -1.0 -0.6 -1.0 -0.6
(1.30) (1.31) (2.55)* (1.28) (2.50)* (1.22)
Population within 100km of coast (%) 0.9 0.7 0.7 0.6 0.8 0.9
(2.85)** (2.64)* (2.41)* (1.66) (2.36)* (1.80)
Initial falciparum malaria index -1.3 -2.1 -1.8 -1.8 -1.3
(2.24)* (3.77)** (3.12)** (1.77) (1.98)*
Change of falciparum malaria index -2.6 -2.5 -2.2
(4.07)** (3.48)** (2.24)*
0.1Tropical disease, 1st principle
component (1.51)
-1.6Initial “WHO Advisory” malaria
index (2.18)*
Constant 1.3 6.1 5.7 3.7 -0.9 -14.8
(0.36) (1.68) (1.58) (0.63) (0.21) (1.42)
Observations 75 75 73 60 73 78
R-squared 0.77 0.80 0.80 0.76 0.77 0.71
Robust t-statistics in parentheses
* significant at 5% level; ** significant at 1% level
Table 7. Level and Changes in Malarial Prevalence between 1965 and 1994 by Climate Zone
Predominant Climate
Malaria Index 1965
Average Change 1965-
0.2 -0.2
27.8 -8.8
61.7 -5.0
64.9 0.5
Note: Countries are classified by their predominant ecozone from the following groupings: Temperate (temperate,
boreal and polar ecozones), Desert (tropical and subtropical deserts), Subtropical (non-desert subtropical), and Tropical
(non-desert tropical). The index and average reduction are unweighted averages over countries. See Gallup and Sachs
(1999) for the sources of climate zone data.
Malaria risk - 1946, 1965, 1994
High risk of Malaria
Tropic of Cancer
Tropic of Capricorn
Figure 1.
GDP per capita 1995
GDP per capita 1995
$450 - 999
$1,000 - 2,699
$2,700 - 3,199
$3,200 - 4,399
$4,400 - 4,499
$4,500 - 6,499
$6,500 - 18,099
$18,100 - 21,999
$22,000 - 31,100
No Data
Tropic of Cancer
Tropic of Capricorn
Figure 2.
Figure 3.
Malaria Index 1965
Tropic of Cancer
Tropic of Capricorn
Malaria index 1965
0.01 - 0.10
0.11 - 0.29
0.30 - 0.47
0.48 - 0.74
0.75 - 1.00
Figure 4.
Malaria Index 1994
Tropic of Cancer
Tropic of Capricorn
Malaria index 1994
0.01 - 0.18
0.19 - 0.42
0.43 - 0.65
0.65 - 0.80
0.81 - 1.00
... In the second strand of the literature, Gallup et al. (1999) and others (Gallup and Sachs 2001;Sachs and Malaney 2002;Carstensen and Gundlach 2005) use geography, in several dimensions, as the main driver of early institutional and economic development, with one of those dimensions being "malaria ecology." However, there has been a lot of discontent over their measure of malaria prevalence, because malaria ecology is not exogenous to modern development, as well as it does not provide enough variation. ...
... Later, following the criticism of Acemoglu et al. (2001Acemoglu et al. ( , 2002, Gallup and Sachs (2001) and Sachs and Malaney (2002) used a malaria risk index, which is based on the 1994 world malaria prevalence map by WHO. Their main finding was that even after controlling for institutions, a higher risk of malaria negatively affects current income per capita, thus supporting the argument of a direct link. ...
... Their main finding was that even after controlling for institutions, a higher risk of malaria negatively affects current income per capita, thus supporting the argument of a direct link. Gallup and Sachs (2001) and Sachs and Malaney (2002) also added that the reason Acemoglu et al. (2001) did not find a direct effect of malaria is because they restricted their data sample to former colonies, which are mainly in the tropics, therefore leading to low variability of the malaria environments. Similar results to those in Gallup and Sachs (2001) and Sachs and Malaney (2002) were reported by Carstensen and Gundlach (2006). ...
Full-text available
This research examines the causal impact of institutional quality on economic development from a novel perspective. At the country level, we exploit variation in the natural malaria prevalence in 1900, just before vector-control methods were developed, to instrument for institutional quality using a two-stage least squares instrumental variables framework. Our instrument is a population-weighted average of malaria endemicity estimates for the year 1900 developed by the World Health Organization scientists in the 1960s. We argue that this measure of historical malaria offers more expansive geographic information about the disease environment, and our baseline IV estimates reveal that greater institutional quality causes greater contemporaneous economic growth. Next, we investigate the robustness of these baseline results to alternative explanations, including the role of geography and the role of early colonizers’ experiences, as the causal link between the early disease environmental, institutional quality, and contemporary growth. As an additional test of the explanatory power of malaria endemicity in 1900 as an instrumental variable, we replace our instrument for settler mortality and replicate the core results from the seminal study on colonial origins of comparative development by Acemoglu et al. (2001). In summary, we propose that malaria endemicity estimated for 1900 holistically explains the legacy of early disease on institutional quality development and contemporary economic development, thus significantly contributing to the resolution of the geography versus institutions debate.
... Past studies revealed that countries in Africa with high level of poverty tend to record greater incident of malaria as compared to richer countries although this seems to be more of an effect and not the cause [6]. According to the Nigeria Bureau of Statistics (2020), the level of poverty in Nigeria is estimated at 40.1% and this high level of poverty is one of the major burdens to prevention and treatment of the disease [7]. ...
... The economic geography hypothesis postulates that not all the areas in the world are equally endowed in terms of geographic attributes such as soil quality, natural resources, topography, climate, and disease environment thus necessitating cross-country transfer of resources. In this framework, each country's production function is intrinsically dependent on the costs of transportation as a response to the need to trade with one another (Acemoglu, 2009;Diamond, 1997;Gallup and Sachs, 2001;Landes, 1998). With resource and output disparities across countries, Harris (1954), argues that the potential demand for goods and services produced in any one location is determined by the distance-weighted GDP of all other locations. ...
Full-text available
This study examines the finance-growth nexus among countries in the Southern African Development Community and East Africa Community regional blocs. We aim to see if small economies can rely on expansionary monetary policies to stimulate economic growth given the illusive nature of such policies. The panel error correction model is applied to analyze the aggregated data, which were obtained from the World Development Indicators and the Comité Européen d'Etudes des Polyphosphates. The results indicate that proximity to trading giants jeopardizes the freedom to use monetary expansionary policies among relatively small countries. If such policies are adopted, it will be a significant impact on countries located in regions where trading powers are evenly distributed.
... Many studies have highlighted malaria with poverty, which is apparent since it is highly prevalent mostly in regions with poor socio-economic conditions that prevent populations from using ITNs optimally [44,45]. A continuous effort is required to ensure the development and implementation of policies that promote rapid and sustained economic growth, as well as programmes to improve the livelihoods of individuals in impoverished malaria communities [46]. This study shows that financial incentives to support household resources and other activities can change people's perceptions of inappropriate ITN use and lead to an increase in proper ITN use. ...
Full-text available
Insecticide-treated net (ITN) is the most applicable and cost-effective malaria intervention measure in sub-Saharan Africa and elsewhere. Although ITNs have been widely distributed to malaria-endemic regions in the past, their success has been threatened by misuses (in fishing, agriculture etc.) and decay in ITN efficacy. Decision-making in using the ITNs depends on multiple coevolving factors: malaria prevalence, mosquito density, ITN availability and its efficacy, and other socio-economic determinants. While ITN misuse increases as the efficacy of ITNs declines, high efficacy also impedes proper use due to free-riding. This irrational usage leads to increased malaria prevalence, thereby worsening malaria control efforts. It also remains unclear if the optimum ITN use for malaria elimination can be achieved under such an adaptive social learning process. Here, we incorporate evolutionary game theory into a disease transmission model to demonstrate these behavioural interactions and their impact on malaria prevalence. We show that social optimum usage is a function of transmission potential, ITN efficacy and mosquito demography. Under specific parameter regimes, our model exhibits patterns of ITN usage similar to observed data from parts of Africa. Our study suggests that the provision of financial incentives as prompt feedback to improper ITN use can reduce misuse and contribute positively towards malaria elimination efforts in Africa and elsewhere.
... Malaria and other infectious diseases, according to Bhattacharyya(2008),have fatal as well as debilitating effects on the human population in Africa. According to scholars (Bloom and Sachs, 1998;Gallup and Sachs 2001;Bhattacharyya, 2008), Malaria reduced the annual growth of the continent by 1.3% and that the suppression of malaria in the 1950s would have led to the doubling of per capita income from what we are experiencing today. Malaria, according to Bhattacharyya(2008), has a huge effect on African per capita income and growth. ...
Full-text available
This study's main objective is to examine the impact of corporate governance mechanisms on financial performance within the banking industry in Nigeria after it went through consolidation in 2005. ROA and Tobin's q are used as the dependent variables, while board size, board composition, CEO shareholding, and ownership concentration as the independent variables. The study employed a correlation research design, and the population comprises the commercial banks in Nigeria listed on the Nigeria Stock Exchange (NSE). Secondary data was used, covering thirteen-year period (2006-2018) using sampled banks' audited reports. The techniques for data analysis used in this work are ordinary least squares-based multiple regression and correlation. The study's findings are that the three variables of interest to the research (board size, board independence, and ownership concentration) are negatively related to the firm performance as measured by Tobin's Q with only board size having a significant relationship. In spite of this, the context analysis shows that boards dominated by independent directors tend to perform better compared to the board with more executives. It is, therefore, recommended that the banks should continue to keep their boards sizeable and manageable not minding pressure of inclusion from their consolidating constituent banks. Infusion of more Non-Executive Directors is also recommended while further research is conducted on the directors' structure to enhance their performance. Finally, it is advised to diversify the Board composition by having representation from the various constituents of the mergers or acquisitions.
We examine the negative child health impacts of soil zinc (Zn) deficiency in Nepal. Soil Zn deficiency limits both crop yields and the Zn concentration in food crops, leading many to speculate that it underlies human Zn deficiency and child stunting, globally and particularly in South Asia. We find strong evidence that soil Zn deficiency does have a causal impact on child stunting in Nepal’s Tarai region, the breadbasket of the country. Using causal bounds, we find that a 1 part per million increase in plant-available soil Zn — achievable with application of Zn-enriched fertilizer — decreases child stunting by between 1 and 7.5 percentage points. Multiple statistical sensitivity tests indicate that this relationship is unlikely to be manufactured by omitted, relevant variables.
Under what circumstances might climate change lead to negative security outcomes? Over the past fifteen years, a rapidly growing applied field and research community on climate security has emerged. While much progress has been made, we still don't have a clear understanding of why climate change might lead to violent conflict or humanitarian emergencies in some places and not others. Busby develops a novel argument – based on the combination of state capacity, political exclusion, and international assistance – to explain why climate leads to especially bad security outcomes in some places but not others. This argument is then demonstrated through application to case studies from sub-Saharan Africa, the Middle East, and South Asia. This book will provide an informative resource for students and scholars of international relations and environmental studies, especially those working on security, conflict and climate change, on the emergent practice and study of this topic, and identifies where policy and research should be headed.
The majority of the world’s population is at risk of one or more mosquito-borne diseases that are transmitted by blood-feeding female mosquitoes, affecting both human health and economic development. Especially Anopheles gambiae, the principal malaria vector, and Aedes aegypti, the vector of dengue and yellow fever, are of primary concern due to their strong specialisation on human hosts, and the high number of casualties caused by the pathogens they transmit. Host seeking and discrimination are crucial for disease transmission, and are predominantly mediated by olfaction. Using a wind tunnel system and a custom analysis pipeline, this thesis confirms that the two mosquito species use volatile host cues, derived from breath and body, differentially, as carbon dioxide on its own drives host seeking in Ae. aegypti, but not in An. gambiae (paper I). To discriminate between host and nonhost species (paper V), Ae. aegypti encode human identity by the relative activation of two glomeruli within the antennal lobe, the primary olfactory centre, of which one is tuned to long-chain aldehydes enriched in human odour. A synthetic blend mimicking the glomerular activation elicited host seeking in Ae. aegypti (paper II). Next to preferring human over non-human hosts, Ae. aegypti also prefer some human individuals to others, which was demonstrated to be affected by the ABO blood type and pregnancy or menstrual cycle phase. Analysis of the volatiles associated with individual volunteers, identified 1-octen-3-ol to be significantly associated with very high attractiveness (paper III). The molecular regulation of host seeking acquisition during An. gambiae female adult maturation was independent of odorant receptor gene AgamOR39 expression (paper IV). The results presented in this thesis contribute to the understanding of mosquito host seeking and discrimination from multiple perspectives, which is a prerequisite to ultimately develop novel tools for mosquito monitoring and control.
Insecticide-treated nets (ITNs) have been useful and effective in mitigating the risk of malaria (one of the oldest, deadliest, and most common human mosquito-borne diseases) globally. However, due to misuse and normal/human-induced physical and chemical wear, etc., the effectiveness of ITNs in combating malaria has been declining. Underlying heterogeneities in the nature of malaria, combined with environmental factors such as temperature lead to complex malaria transmission and control dynamics. In particular, temperature plays a significant role in determining the risk of malaria since it influences the growth and survival of mosquitoes and the malaria parasite. Here, a unifying mechanistic model framework that integrates malaria dynamics with waning ITN-efficacy and temperature change is developed and used to assess the impact of interactions between significant sources of variation (e.g., temperature) and waning ITN-efficacy on the risk of malaria transmission and the success of ITN programs. The model exhibits a backward bifurcation when ITN-efficacy is constant implying that control efforts must be stepped up and sustained a bit longer even when the reproduction number is slightly less than one. The study shows that malaria is more effectively controlled with ITNs that have a longer lifespan and if ITNs are replaced before the end of their expiration period. Also, failing to account for waning ITN-efficacy leads to an underestimation of disease risk, burden, and effort level required to contain the disease. Local and global sensitivity analyses show that control and temperature-related parameters are primary drivers of the reproduction number and the human disease burden, highlighting the significance of temperature in malaria transmission dynamics. Furthermore, results of the study show that the human disease burden is optimal at a temperature of ≈28°C and that high seasonal variations can trigger major malaria outbreaks even in regions with low mean temperatures. Additionally, accounting for both seasonality and decay in ITN-efficacy leads to complex malaria patterns. To sum it up, insights into the sensitivity of malaria dynamics on temperature are useful in assessing the potential impact of changes in temperature on malaria risk. Also, a malaria control program, which ensures that ITNs are replaced regularly and early enough, and that educates at risk populations on proper use and care for ITNs is necessary for reducing the burden of malaria.
Using a massive malaria eradication program in India during the 1950s as a natural experiment, we examine the effects of disease environment on child health outcomes and fertility. We harmonise a rich dataset on malaria endemicity with fertility histories of women to exploit the cohort level variation in exposure to the program. We find that the program leads to a significant decline in infant and neonatal mortality and leads to a significant increase in the probability of birth during adolescent years in high malaria endemic regions. We confirm that a fall in the mother’s age at first birth in the post-eradication period drives the fertility response.
This double-blind clinical trial was conducted in Thailand to assess the impact of iron treatment on the IQ and educational attainment of 1358 9–11-y-old children. The children were classified into one of three groups: iron replete, iron depleted, and iron-deficient anemic. The Raven Progressive Matrices was used to measure IQ. A Thai language and a math test were administered to assess school attainment. A 50-mg/d tablet of ferrous sulphate was given for 2 wk and a 100 mg/d tablet, for 14 wk. An anthelminthic drug was given on the day of the blood test before treatment and 3 mo after the intervention started. There is evidence of a positive association between iron status and IQ and a language school achievement test but there is no support for the internal validity of the hypothesis that this association is causal.