political science 2.5~3pages essay need to be done in 18hours.

Problems 451

in the box “Determinants of Cooperation.” Then tell them to imagine they are the proposer, that the responder is someone from their own economic and social group, and that the sum of money to be divided is $1,000. Ask them what

division they would propose and why. Report the results of your survey, and then draw con- clusions about what explains differences among your respondents regarding their proposed divisions.

For additional exploration and practice using the Online Data Plotter and data sets, please visit www.pearsoninternationaleditions.com/weil.

452

In examining the fundamental determinants of income differences among countries, we have repeatedly grappled with the issue of whether the vari- ables we were measuring were really so fundamental. How can we be sure that the nature of a country’s government determines its level of income, for example, when there is also a compelling case that income affects gov-

ernment? The same holds true for income inequality and for culture. In this chapter we look at a set of potential determinants of income—

geography, climate, and natural resources—that are clearly immune to this problem. As we shall see, however, our difficulties are not completely behind us. To give a preview, Figure 15.1 shows a scatter plot comparing income per capita and a nation’s latitude—that is, its distance from the equator. There is clearly a strong relationship in the data: The farther a country is from the equator, the richer it is, on average. Further, there is certainly no danger of “reverse causa- tion,” that is, no risk that the relationship in the data occurs because becom- ing rich causes a country to move farther from the equator. But what does the relationship between income and latitude tell us? What economic machinery underlies this relationship?

In this chapter we examine data on how geography, climate, and natural re- sources differ among countries. We will see that there are good theoretical reasons why each of these characteristics should affect income. In the case of two of these characteristics—geography and climate—we will also find good empirical evi- dence that the effect is significant. We will also see how these characteristics have been incorporated into theories of why the Eurasian landmass developed before the rest of the world and into two different explanations for why Europe developed before China.

One conclusion from this chapter is that the natural resources available in a given country today are not a great constraint on growth because countries can import resources from abroad. But this finding leaves open the question of whether the availability of resources at the world level may not constrain world- wide growth. We return to this topic in Chapter 16.

GEOGRAPHY, CLIMATE, AND NATURAL RESOURCES

15C H A P T E R

Geography is destiny.

—Napoleon Bonaparte

15.1 Geography 453

15.1 GEOGRAPHY The population of the world passed the 7 billion mark on about October 31, 2011. This population inhabited planet earth’s surface area of 58 million square miles (150 million square kilometers) of land. These numbers imply an average population den- sity of 121 people per square mile (47 people per square kilometer). But as the map of world population density inside this book’s back cover shows, the population of the planet is not spread out evenly. Mountains, deserts, and polar regions have extremely low population densities. In total, 90% of the world’s population lives on only 10% of the land. Population densities of countries range from 5.2 people per square mile in Mongolia and 7.8 people per square mile in Australia up to 3,271 people per square mile in Bangladesh. The United States has 88.1 people per square mile.

That some parts of the world have low population densities is no surprise. It is difficult to live at all—let alone to produce output—in some of the earth’s inhos- pitable regions. Most people live in places where conditions are favorable for pro- duction and for living: where temperatures are moderate, the ground is level, soil is fertile, and there is neither too much nor too little precipitation.

FIGURE 15.1

Relationship between Latitude and Income per Capita

Sources: Heston, Summers, and Aten (2011), Gallup, Mellinger, and Sachs (2001).

China

Congo, Dem. Rep.

India

Luxembourg

Malaysia

Moldova

Norway

Russia

Singapore United States

Zimbabwe

GDP per capita, 2009 (2005 Dollars, ratio scale)

5040 60 70 803020100

10,000

1,000

100,000

100

Latitude (absolute degrees distant from the equator)

F

454 CHAPTER 15 Geography, Climate, and Natural Resources

In this section we consider how the standard of living (as opposed to the density of population) is related to geographic characteristics. We examine effects of geography on economic growth through the channels of international trade, cross-border influences of neighboring countries, and geographic influences on government. We also look at the role that geographic factors played in economic growth in the millennia before the continents were linked by transoceanic travel. Then, in Section 15.2, we explore the effects on economic growth caused by one of geography’s most important aspects: climate.

Location, Trade, and Growth In Chapter 11 we considered how international trade affects economic growth. We saw that openness to international trade raises income per capita primarily through its effect on productivity: Openness facilitates technology transfer, leads to a more efficient organization of the economy, and raises income by allowing a country to specialize in the goods it is best at producing.

Recall that one of the determinants of openness to trade is geography. Unlike other determinants of trade, such as tariffs and quotas, a country’s geography is unchange- able. If geography determines trade, and if trade helps a country grow rich, then some countries (or regions of countries) have a fundamental advantage over others.

The most important geographic determinant of a country’s ability to partici- pate in international trade is its proximity to the ocean. In The Wealth of Nations (1776), Adam Smith wrote:

As by means of water carriage a more extensive market is opened to every sort of industry than what land carriage alone can afford it, so it is upon the sea-coast, and along the banks of navigable rivers that industry of every kind begins to sub-divide and improve itself, and it is frequently not till a long time after that those improve- ments extend themselves to the inland part of the country.

As in Smith’s day, today ocean transport is the cheapest way to ship goods. And geographic evidence bears out the importance of ocean transport in determining where people live and what standard of living they enjoy. Only 17.4% of the world’s landmass is located within 60 miles (100 kilometers) of an ocean or a river that is navigable as far as the sea. However, 49.9% of the world’s population lives on this land, and 67.6% of the world’s gross domestic product (GDP) is produced on it. GDP per capita in areas that are within 60 miles of the sea is, on average, twice as high as GDP per capita in areas farther inland.1

Proximity to the sea can also go far toward explaining large differences in economic success among regions of the world. Figure 15.2 shows a strong positive relationship between the fraction of a region’s population that lives within 60 miles

1Mellinger, Sachs, and Gallup (2000).

15.1 Geography 455

of an ocean or a navigable river and the region’s average level of GDP per capita. Particularly striking in the figure is the low fraction of the population in sub- Saharan Africa, only 21%, that has access to the sea. This is caused by several factors, includ- ing a dearth of natural ports and an absence of navigable rivers (e.g., the enormous Congo River system is blocked by waterfalls only 83 miles inland). In addition, sub-Saharan Africa has a low ratio of coastline to land area (Western Europe has one-eighth the land area of Africa, but its coastline is 50% longer). Furthermore, Africa’s population is concentrated in the interior highlands, where the tropical heat is somewhat reduced.

Besides access to waterways, another determinant of a country’s openness to trade is its location with respect to major centers of economic activity. On aver- age, each 600 miles (1,000 kilometers) of distance from one of the most developed regions of the world (the United States, Western Europe, or Japan) raises transport costs by one percentage point. Similarly, increasing the distance between two coun- tries by 1% lowers the volume of trade between them (relative to GDP) by 0.85%.

Together, access to the sea and distance from major centers of economic activ- ity help account for differences in the cost of transporting goods. The average cost of transporting imports, expressed as a ratio to the total value of imports, is 3.6% for the United States, 4.9% for Western Europe, 9.8% for East Asia, 10.6% for Latin

FIGURE 15.2

Regional Variation in Income and Access to the Sea

Source: Gallup, Sachs, and Mellinger (1999).

10060 8040200

Percentage of population with access to the sea

GDP per capita, 1995

$10,000

$20,000

$25,000

$5,000

$15,000

0

Sub-Saharan Africa South Asia

Latin America Eastern Europe and

former Soviet Union

East Asia

Western Europe

F

456 CHAPTER 15 Geography, Climate, and Natural Resources

America, and 19.5% for sub-Saharan Africa. These differences in transport costs correlate well with differences in trade volume and income per capita.2

Access to trade explains differences in income not only among countries but also among the regions of a single country. The case of China provides an interesting

what was the source of this discrepancy? Why had the Americas, Australia, and sub-Saharan Africa not developed to the same level as the civilizations of Europe and Asia by the time that they all came into close contact?

In Guns, Germs, and Steel: The Fates of Human Societies, biologist Jared Diamond argues that it was geography that determined this out- come.* According to Diamond, Eurasia, the land- mass composed of the continents of Europe and Asia, had several key advantages over the rest of the world that allowed it eventually to dominate.

Eurasia’s most important advantage was its good fortune in having numerous species of plants and animals that could be domesti- cated. The large mammals that formed the basis for premodern agricultural economies—cows, horses, pigs, sheep, and goats—were all native to Eurasia. By contrast, in the Americas, the only mammals that could be domesticated were llamas and alpacas, both of which were localized in their habitats and of limited economic use- fulness. Similarly, of the 56 large-seeded grass species that could potentially be domesticated as food grains, 39 were native to Eurasia, 11 to the Americas, 4 to sub-Saharan Africa, and only 2 to Australia.

This advantage in the range of species that could be domesticated stemmed partially from plain luck. It also resulted from a second advan- tage that Eurasia had: its size. Including North Africa, to which it was both culturally and eco- nomically linked, the Eurasian landmass is 50%

S ome 500 years ago, there began one of the most pivotal events in human history. Over three centuries, starting with Christopher Columbus’s voyage to the Americas in 1492 and ending with James Cook’s discovery of Australia in 1770, human populations that had been geographically isolated from each other for millennia came into violent contact. The impetus for this collision of civilizations was the spread of Europeans—explorers, traders, colo- nists, and conquerors—throughout the world.

The most dramatic impact of this clash of civilizations occurred in the Americas and in Australia, where much of the native cul- ture, language, and population were wiped out and replaced with European imports. Most of Asia, by contrast, either was never colonized by Europeans at all or, if colonized, managed to maintain its precolonial civilization, popula- tion, and language. Sub-Saharan Africa’s fate lay between these two extremes: European control was more complete than in Asia, but it never resulted in the wholesale displacement of the native population, as it had in the Americas.

Many of the proximate causes of Europe’s domination are fairly obvious: Europeans of the 16th century had better weaponry and more sophisticated social organizations than did the American, Australian, and African natives with whom they came into contact. The considerably smaller technological gap between Europeans and Asians explains why Europeans failed to dominate Asia as they did other regions. But

FGUNS, GERMS, AND GEOGRAPHY

2Bloom and Sachs (1998), Gallup, Sachs, and Mellinger (1999), Frankel and Romer (1999).

15.1 Geography 457

example. Before 1978, government policy closed the Chinese economy to most foreign trade. In the 25 years following the liberalization of trade policy, the ratio of trade to GDP quadrupled. The fastest growth in trade occurred in China’s coastal regions, which also experienced much faster income growth than did the country’s interior provinces. Thus, once the artificial barrier of government trade policy was removed, geographic differences in the ease of trading with the rest of the world led to an increase in income inequality among different regions in China.

larger than the Americas, two-and-a-half times as large as sub-Saharan Africa, and eight times as large as Australia. A larger area is likely to contain more useful plant and animal species. If the employment of these species then spreads throughout the landmass, residents of a larger area ultimately have the benefit of a higher number of useful species.

Another advantage of Eurasia was in its geographic orientation along an east-west axis. This orientation allowed for the spread of ag- ricultural techniques and of useful plant and animal species throughout a zone of relatively similar climates. Thus, the chicken, domesti- cated in China, could spread to Europe, whereas grains first domesticated in the Fertile Crescent of southwest Asia spread as far east as Japan. By contrast, the north-south orientation of the Americas meant that climactically similar zones, which could potentially have shared agricultural technologies, were separated by areas in which these technologies would not be useful and thus would not spread. Similarly, although the cli- mates of southern Africa and the Mediterranean basin were similar enough that European crops would have grown in southern Africa, the two regions were separated by a whole continent in which European crops would not flourish. Consequently, these crops did not reach south- ern Africa until they arrived by sea.

The availability of food crops and domestic animals in Europe and Asia allowed for more ef- ficient food production, denser populations, and

the rise of advanced civilizations. Surplus food could support a large class of rulers, priests, and warriors. This context fostered the rise of new technologies, including writing, metallurgy, and the oceangoing ships that took European coloniz- ers around the world. The east-west orientation of the Eurasian landmass allowed important in- ventions such as the wheel (invented around 3000 b.c. in the Black Sea region) to spread throughout the continent. And the vastness of the Eurasian landmass permitted a large population to share these new technologies. As a result, Eurasia was more economically advanced—and more densely populated—than any other part of the earth.

Finally, domestic animals and dense popu- lations gave Europeans one more crucial ben- efit. People’s close association with large animals allowed many animal diseases, such as mea- sles and smallpox, to transfer to humans. And dense populations caused Europe to sustain a number of endemic diseases that, in a sparser population, would have died out. Over time, Europeans developed partial immunity to these diseases, although they still harbored the agents of infection. Thus, when Europeans came into contact with unexposed American Indians, the results were devastating. Diseases killed far more Americans than any deliberate action of the Europeans, and the massive depopulation left the Americas (and later Australia as well) open to domination and colonization.

*Diamond (1997).

458 CHAPTER 15 Geography, Climate, and Natural Resources

Geographic Concentration and Spillovers A look at the world map inside the front cover of this book suggests that wealthy coun- tries tend to be near one another. Europe is the best example. Among non- European countries that are wealthy, there is also a good deal of clustering, such as Canada and the United States, Japan and South Korea, and Australia and New Zealand.

One possible explanation for this clustering is that it reflects countries’ influ- ence on their neighbors. Economists use the term spillovers to describe these cross-border effects. We already saw that countries that are near each other are more likely to trade with each other. Wealthy countries also tend to spread jobs to their poor neighbors to take advantage of low wages; one prominent example is the maquiladora assembly plants in Mexico that import components from the United States and ship their output back over the border. A wealthy neighbor also provides a positive example, a source for ideas to copy, and opportunities for train- ing. In contrast, a politically unstable neighbor is likely to be a source of refugees or military aggression. Because poor countries are more prone to suffer from such instability, having rich neighbors is an aid to a country’s growth.

A second possible explanation for this clustering is that nearby countries share common characteristics that are important for growth. For example, countries that are close to each other share the same climate. Similarly, neighboring countries may have common characteristics that are difficult for economists to measure. In Chapter 14, for example, we weighed the possibility that the fast-growing East Asian countries share a Confucian culture.

Depending on the exact source of the relationship between a country’s in- come and that of its neighbors, the fact that wealthy countries are geographically clustered may represent an additional obstacle for the development of many poor countries. Specifically, if the clustering of economic growth is indeed the result of spillover effects, then this is good news for a few developing countries that are near richer countries (e.g., Mexico, Morocco, and China), but it is bad news for most of the developing world, particularly for sub-Saharan Africa. If, however, the cluster- ing of economic growth simply reflects common factors among neighboring coun- tries, then the clustering itself does not represent additional bad news. According to this second interpretation, countries that put in place the building blocks of growth—good government, accumulation of physical capital and human capital, and so on—will grow rich even if their neighbors remain poor.

Geography’s Effect on Government Another way in which geography has been theorized to affect economic growth is through its effects on the size of states as well as the conduct of government.3 The background to this theory is the observation of differences in the historical

3Jones (1987).

15.1 Geography 459

formation of states in Europe in comparison to most of the rest of the world— and, in particular, in comparison to China—in the period before the Industrial Revolution. Europe was exceptional for its lack of a unified government. The divi- sion of Europe into a large group of independent countries has been a feature of the continent’s history since the dissolution of the Roman Empire. In 1600, there were 500 more or less independent political units in Europe, and even in 1900, af- ter centuries of consolidation, there were 25 independent European states.

In contrast to Europe, China has had a history of remarkable centralization, beginning with the first unified Chinese state in 221 b.c. The last extended period of partition ended in the 13th century. Even as China was conquered by foreigners including the Mongols (13th century a.d.) and the Manchus (1644), the country continued to be governed as a single unit. Beyond China, empires were also the general rule in most of the preindustrial world.

Was Europe’s lack of unification good for economic growth? A priori, we might expect exactly the opposite. A large, unified country will have a large mar- ket and thus the potential for gains from specialization. Productive ideas should also spread more easily in a unified country. Disunity raises the prospect of war between neighboring states, which wastes resources—and indeed, preindustrial Europe experienced a great deal of fighting among neighboring states.

Despite these theoretical advantages of unification, historical experience points to a number of ways in which the lack of centralization in Europe proved advantageous for economic growth. First, external competition served as a check on governments’ power. Although a given ruler might be tempted to enact policies that would stifle eco- nomic innovation to maintain the status quo, there was always the danger that neigh- boring countries would allow innovation and thus gain an advantage. Government’s size was limited by the ability of capital owners to move their wealth—and of workers to move themselves—from one jurisdiction to another if they found taxes or other restrictions too burdensome. These constraints forced Europe’s monarchs to be less prone to wasteful extravagance than their Chinese peers. Second, when governments did try to suppress destabilizing economic innovation (or the destabilizing ideas that went along with it), the innovators often could move to a neighboring country. In China, by contrast, there was usually no outside competition for the government to worry about, nor was there any place for suppressed innovators to go.

The danger of the unified government in China was most forcefully demon- strated in the 15th century, when the imperial court turned violently against oceanic exploration. The huge Chinese fleets, which had sailed as far as the east coast of Africa, fell into disrepair. By 1500, building a ship with more than two masts had become a crime punishable by death. The advantage of European fragmentation was demonstrated in a similar area: When Genoese navigator Christopher Columbus was unable to get financing for his voyage of exploration from the Portuguese, he turned to their neighbors, and competitors, the Spanish. Europeans were no less disposed to suppress useful ideas than were Chinese—witness the trial of Galileo in 1633, in which the great astronomer was forced to recant his belief that the earth

460 CHAPTER 15 Geography, Climate, and Natural Resources

rotates around the sun (his writings were not removed from the Catholic Church’s index of banned books until 1835). But even though Galileo’s work was suppressed in Catholic areas, his ideas continued to advance in Protestant regions.

In sum, the historical record seems to show that the virtues of competi- tion—which we examined in Chapters 10 and 11 with respect to firms— apply to countries. The question then becomes: What was the source of Europe’s exception- alism? In other words, why did Europe remain politically fragmented while large empires were the norm in the rest of the world?

One prominent theory attributes Europe’s fragmented political structure to geography. Figure 15.3 shows that Europe’s most fertile lands—areas including the

FIGURE 15.3

Core Areas in Preindustrial Europe

Source: Pounds and Ball (1964).

Tirane

Minsk

Sarajevo

Sofia

Zagreb

Prague

Copenhagen

Helsinki

Paris

Athens

Budapest

Rome

Riga

LIECH. Vaduz

Vilnius

Skopje

Chisinau

Monaco

Amsterdam

Oslo

Warsaw

Bucharest

Moscow

Bratislava

Madrid

Stockholm

Bern

Kiev

Vienna

Brussels

Tallinn

Berlin

Dublin

LUX. Luxembourg

Lisbon

Ljubljana

London

Core areas of modern states

Peripheral and external core areas

Ephemeral core areas

Atlantic Ocean

North Sea

Mediterranean Sea

Baltic Sea

Black Sea

RUSSIA

FINLAND

AUSTRIA

ITALY

SPAIN

SWEDEN

NORWAY

GERMANY

FRANCE

PORTUGAL

HUNGARY ROMANIA

BULGARIA

DENMARK

POLAND

BYELARUS

UKRAINE CZECH. SLOVAKIA

GREECE

NETHERLANDS

BELGIUM

IRELAND

SERBIA

ALBANIA

MOLDOVA

LITHUANIA

LATVIA

ESTONIA

MONTENEGRO

BOSNIA

CROATIA

SLOVENIA

SWITZ.

MACEDONIA

U.K.

F

15.1 Geography 461

London basin, the Ile de France, and the plain of the Po River, which would become the cores of modern states—are widely dispersed among large areas of reduced fer- tility.4 As historian Eric Jones writes, “On the modern map the intervening spaces have been cleared, drained, cultivated and filled up with people, but until the end of preindustrial times Europe was a succession of population islands in a sea of forest and heath.”5 Europe is also cut apart by numerous natural barriers, including moun- tain chains such as the Alps and the Pyrenees, as well as bodies of water such as the English Channel. Although the different parts of the continent can communicate and trade with one another, they are sufficiently separated that they are difficult to govern as a single unit. By contrast, as shown in Figure 15.4, China has only four such core

FIGURE 15.4

Core Areas in Preindustrial China

Source: Stover (1974).

South China Sea

C H I N A

TAIWAN

Beijing

Chengdu Chongqing

Guangzhou

Kaifeng

Shanghai

Xi`an Hwai-an

Hangzhou

Luoyang

Tianjin

Hankow Nanjing

Ka n

R .

Yang tze

R.

Ye llo

w R.

0 200

Miles

400

Grand

Canal

Yel low

R .

F

4Pounds and Ball (1964). 5Jones (1987), p. 106.

462 CHAPTER 15 Geography, Climate, and Natural Resources

regions, of which two, centered along the Yellow and Yangtze rivers, are dominant.6 Further, the main core areas of China were connected by the Grand Canal in the fourth century b.c. Thus, geography made it likely that China would be governed as a single unit, a circumstance that held back economic growth.

Like many other explanations for the different historical experiences of Europe and China, however, this theory has to be treated with some caution. An interest- ing counterexample comes from the history of the Indian subcontinent, which has a geography much like Europe’s with scattered areas of fertile land separated by desert, hills, and jungle. India was not politically unified between the Gupta em- pire in the fifth century a.d. and the Mughals in the 16th century a.d. Yet political fragmentation in India did not have the same growth-inducing effects that it did in Europe. This example is a useful reminder that geography is not necessarily des- tiny—geography may be part of the explanation of why Europe developed before the rest of the Eurasian continent, but not all of it.

15.2 CLIMATE One of the most important aspects of a country’s geography is its climate—that is, the seasonal patterns of temperature, precipitation, winds, and cloud cover. We be- gin our analysis of climate by looking at data on how income and population differ among the world’s climate zones. We then investigate some of the potential causes of this variation.

Previously in this chapter, we noted the strong relationship between a country’s distance from the equator (i.e., its latitude) and its income per capita. Because latitude is linked to climate, this finding suggests a role for climate in determining income per capita. But climate does not depend on latitude alone; factors such as prevailing weather systems, distance from the ocean, and altitude also matter. Thus, for example, the nations of Western Europe, because they are warmed by the Gulf Stream, are not as cold as other countries at similar latitudes. Altitude explains why the highlands of Kenya are relatively moderate in their climate despite their proximity to the equator. For an accurate assessment of cli- mate’s role in determining income per capita, we need to look at detailed data on how climate varies.

Geographers divide the earth’s climate into 12 zones, which are detailed in Table 15.1. The left-most column of the table gives the climate designation ac- cording to the Koppen-Geiger system of climate classification (the exact meaning of these designations is not important for our purposes).7 The second column describes the climate zone, and the third column lists some representative cities in each zone. The fourth and fifth columns show the percentage of the world’s

6Chi (1963). 7For a full discussion, see Strahler and Strahler (1992).

15.2 Climate 463

TABLE 15.1

World Climate Zones F

Climate Zone

Description

Representative

Cities

Percentage of World

Landmass

Percentage of World

Population

GDP per Capita Relative to

World Average

Af Tropical Rain Forest Jakarta, Indonesia; Manaus, Brazil

4.0 4.4 0.64

Am Tropical Rain Forest with Seasonal Monsoon

Manila, Philippines; Cochin, India; Belém, Brazil

0.8 2.4 0.41

Aw Tropical Savannah Dhaka, Bangladesh; Kinshasa, Congo; Havana, Cuba

10.8 17.5 0.38

Cw Subtropical: Mild Humid with Dry Winter

Hanoi, Vietnam; Kanpur, India; Lilongwe, Malawi

4.3 16.0 0.44

Cf Mild Humid Climate with No Dry Season

New York, USA; Paris, France; Shanghai, China; Sydney, Australia

7.7 19.5 2.24

Cs Mediterranean Climate: Mild, Humid with Dry Summer

San Francisco, USA; Rome, Italy; Santiago, Chile

2.2 4.3 2.10

Df Snowy-Forest Climate with No Dry Season

Chicago, USA; Moscow, Russia

23.0 5.8 1.90

Dw Snowy-Forest Climate with Dry Winter

Seoul, South Korea; Vladivostok, Russia

6.2 5.3 0.64

BS Semi-arid Steppe San Diego, USA; Odessa, Ukraine

12.3 11.8 0.55

BW Desert: Annual Precipitation Less than 15 in. (38 cm)

Cairo, Egypt; Karachi, Pakistan

17.3 6.2 0.58

H Highlands Mexico City, Mexico 7.3 6.8 0.78

E Ice Climates: Average Temperature in Warmest Month Less Than 50° F (10° C)

Nuuk, Greenland 4.0 <0.1 —

Source: Data on landmass, population, and GDP per capita are from Mellinger, Sachs, and Gallup (1999).

landmass in each zone and the percentage of the world’s population living in each zone. Finally, the sixth column of the table presents average GDP per capita in each zone relative to the world average.

The first three climate zones in Table 15.1 (Af, Am, and Aw) are classified as tropical. These are areas in which average temperatures during the coldest month of the year do not fall below 64° F (18° C). As the table shows, these tropical re- gions, which together contain 24.3% of the world’s population, are poor. As a group, their income per capita is only 43% of the world average. The fourth climate zone, designated Cw, is subtropical and shares many of the characteristics of the tropical group, including its relative poverty.

464 CHAPTER 15 Geography, Climate, and Natural Resources

The second group of four climate zones (Cf, Cs, Df, and Dw) is referred to as the temperate region. As can be seen in Table 15.1, these are generally the wealthi- est parts of the world, the exception being the relatively small Dw zone, which falls exclusively in northeastern Asia. Taken together, the temperate region con- tains 34.9% of the world’s population and has income per capita that is 1.94 times the world average. In western Europe, 96% of the population lives in temperate zones. In North America, the percentage is 88%. By contrast, only 12% of Latin Americans and 4% of sub-Saharan Africans live in such zones, and South Asia has no temperate climate zone whatsoever.

In this section, we investigate the links between climate and economic growth. Climate has direct effects on productivity, most importantly in agriculture. Climate also influences the human input into production because the prevalence of disease is linked to climate and because people’s ability to work is affected by temperature. Finally, climate affects the economy by making a location more or less pleasant as a place to live.

Climate and Agricultural Productivity In 2009, agriculture accounted for 35% of world employment. Historically, the figure was much higher, and in many developing countries this is still the case. For example, the share of employment in agriculture is 59% in sub-Saharan Africa and 54% in South Asia.8 Thus, differences in agricultural productivity among coun- tries profoundly affect income per capita.

Measures of agricultural output per worker differ greatly between tropical and temperate regions. Figure 15.5 shows the relationship between a country’s latitude and the production per worker in agriculture. The range of the data is enormous, with workers in wealthy, temperate countries producing as much as 300 times the agricultural output of workers in poor, tropical countries.

The finding that agricultural output per worker is low in the tropics does not necessarily mean that the tropics are a bad place for agriculture. That is, this fact alone does not indicate whether low agricultural output in tropical countries is the result of inherent differences in the agricultural environment, such as can be traced to climate, or whether there is some other cause. Other factors that might explain the gap in ag- ricultural output between tropical and temperate countries include differences in the use of inputs to production such as agricultural machinery and fertilizers, the human capital of farmers, and the amount of land available per farmer. Similarly, differences in agricultural productivity might result from differences in the institutional environ- ment, such as the quality of government, between tropical and nontropical countries. Only after we account for all of these other potential sources of difference can we judge whether climate can explain the low output of tropical agriculture.

8International Labour Organization (2011).

15.2 Climate 465

9Gallup (1998), Table 2.

Studies of the agricultural production function, employing techniques similar to those we used in Chapter 7 to analyze the production function for the economy as a whole, show that even when we account for differences in farm machinery, fertilizer inputs, and the human capital of workers, agricultural productivity in tropical countries is lower than in the temperate zone. One estimate is that using the same capital, labor, and fertilizer inputs, land in wet tropical climates produced 27% less and land in the dry tropics produced 31% less output than land in the wet temperate zone. Land in the dry temperate zone produced 15% less than land in the wet temperate zone.9

At first, the finding that tropical climates are inhospitable to agriculture may be surprising because we tend to associate tropical climates with lush vegetation. Tropical areas also have longer growing seasons than do temperate regions. On closer inspection, however, tropical climates suffer from several disadvantages in producing useful crops. Although the tropics do receive heavy rainfall, the pattern in which rain falls is not good for farming. In much of the tropics, rain falls sea- sonally, so torrential monsoons alternate with long dry seasons. Even where this

FIGURE 15.5

Latitude versus Agricultural GDP per Agricultural Worker

Source: United Nations Food and Agriculture Organization (2010).

Agricultural GDP per agricultural worker, 2009 (ratio scale)

705040 602010 300

10,000

1,000

100,000

100

10

Latitude (absolute degrees distant from the equator)

Australia

China

Ethiopia

France

Hungary

India

South Korea

Libya

Norway

Pakistan

Russia

Saudi Arabia

Singapore

United Kingdom Venezuela

F

466 CHAPTER 15 Geography, Climate, and Natural Resources

seasonal pattern does not occur, tropical rain tends to fall in deluges that can erode the soil. For example, on the island of Java, home to the majority of Indonesia’s population, a quarter of annual rain comes in storms of 2.4 inches (60 millimeters) per hour. Similarly, the seasonal pattern of sunlight in the temperate zones—long days in the summer and short days in the winter, as opposed to the relative con- stancy of sunlight in the tropics—is optimal for growing staple grains such as wheat and corn (maize).

Even more significant as a factor affecting agricultural productivity in the tropics is the absence of frost. Frost, which occurs when the ground-level tempera- ture falls below freezing, has been called “the great executioner of nature” for its effect in killing exposed organisms. Tropical areas are characterized by a wealth of insect life, which competes vigorously with humans in consuming food crops. A second benefit of frost is that by killing some of the microorganisms in the soil, frost slows the decay of organic materials. In tropical regions, the rapid breakdown of organic matter into its component minerals causes farmed land to lose its fertil- ity rapidly unless farmers apply fertilizer. Finally, frost helps to control the types of animal diseases that place a heavy burden on tropical agriculture.10

Climate and Disease In Chapter 6 we looked into how the health of a country’s residents affects income per capita. Healthy people can work harder and longer than unhealthy people; they can also think more clearly. Similarly, students who are healthier do a better job of accumulating human capital in the form of education. We saw a strong cor- relation between health, as measured by life expectancy, and income per capita.

Much of the correlation between health and income per capita results because people in richer countries can afford better inputs into health. But differences in health may also result from factors other than income—the “health environment” of a country that we identified in Chapter 6. In our analysis of the simultaneous determination of income and health, we found that if countries differ in their health environment, these differences will affect both income and actual health. Countries with a poor health environment will in equilibrium be both poorer and less healthy than countries with a good health environment. Further, differ- ences in the health environment among countries will be subject to a multiplier effect: Countries with better health environments will have healthier workers who produce more output, allowing for better nutrition and medical care, which will further improve health.

There is good evidence that the tropics constitute a bad health environment. Tropical regions are rife with diseases that are harmful to humans, including ma- laria, yellow fever, sleeping sickness, and schistosomiasis (also called bilharzia), to

10Masters and McMillan (2001), Sachs (2000).

15.2 Climate 467

name but a few. The concentration of diseases in the tropics results from two fac- tors. First, regions where the temperature never reaches freezing support a much wider selection of parasites and disease-carrying insects than do temperate zones. Second, because protohumans evolved in tropical regions of Africa and spent mil- lions of years there, local parasites had ample time to evolve to take advantage of them. Starting in the 15th century, Europeans inadvertently spread these African parasites to other tropical areas. In contrast to the tropics, the temperate zones have been home to humans for only a few tens of thousands of years, so fewer dis- ease parasites in these regions have evolved to attack humans.11

Of all tropical diseases, malaria has the largest effect on economic growth. In 2010, there were an estimated 216 million cases of malaria, resulting in 655,000 deaths; 80% of cases and 90% of deaths occurred in Africa.12 A typical bout of the disease entails four to six days of nearly total incapacitation, followed by four to eight days of fatigue. Long-term effects on children include brain damage and learning disabilities and on adults include anemia, which lowers the energy level.

Malaria is caused by a one-celled parasite called a plasmodium and is trans- mitted among humans by the bites of mosquitoes of the Anopheles genus. Female mosquitoes that bite an infected person carry away some of the parasite along with the blood they ingest; when the mosquitoes bite another person, the parasite is transmitted into that person’s bloodstream. The malaria parasite has no host other than mosquitoes and humans, and a person who is infected with the parasite will become noninfectious within about two months of having received the parasite. Therefore, malaria can be maintained only in climates where mosquitoes are active almost year-round.

Today malaria is confined almost exclusively to the tropics in what is a signifi- cant retreat from its earlier geographic extent. Before 1945, areas at a high risk for malarial infection included Greece, Spain, Italy, and large parts of the U.S. South. For example, in 1935 there were 135,000 cases of malaria and 4,000 deaths from the disease in the United States. In Europe some geographical areas, such as the Pontine Marshes in central Italy and the Plain of Marathon in Greece, were consid- ered uninhabitable because of the pervasiveness of malaria.

As with other tropical diseases, malaria’s restriction to the tropics could con- ceivably be the result of the poverty of tropical nations rather than to any charac- teristic of malaria itself. That is, tropical countries might be poor for some reason unrelated to malaria (e.g., the effect of climate on agricultural productivity that we considered in the previous section), and such countries might suffer from malaria because they are too poor to undertake proper preventive measures. To assess this issue, researchers have constructed an index of “malaria ecology” for various countries. The malaria ecology index measures the susceptibility of a

11McNeill (1976). 12World Health Organization (2011).

468 CHAPTER 15 Geography, Climate, and Natural Resources

country’s climate to mosquito breeding (which depends on adequate rainfall and warm temperatures), as well as the prevalence of mosquito species that feed only on humans (because human malaria is transmitted by a single mosquito biting two people, transmission is much more likely when mosquitoes limit their meals to humans).

As Figure 15.6 shows, the malaria ecology index is closely related to the actual incidence of malaria, measured as the fraction of the population that was exposed to risk of malaria in 1994. Among countries with low values for the malaria ecol- ogy index, little or none of the population was at risk for the disease. In most of the countries with a high value for the index, 100% of the population was at risk. For intermediate values of the malaria ecology index, we find interesting variation among countries. For example, Spain and India have roughly the same malaria ecology index—that is, their climates are equally hospitable to malaria mosqui- toes—but 66% of India’s population was at risk for the disease whereas none of Spain’s population was. Other countries are notable as outliers. For instance, Pakistan had surprisingly high exposure to malaria given its climate, and Brazil had very low malaria given its climate.

FIGURE 15.6

Malaria Ecology versus Incidence of Malaria

Sources: Kiszewski et al. (2004).

Percentage of the population at risk for malaria, 1994

80

60

100

120

–20

0

20

40

0.000001

Malaria ecology index (ratio scale)

0.0001 0.10.00001 0.001 0.01 1001 10

Finland

Netherlands Belgium

United States Italy Spain

Brazil Mauritius

Thailand

Papua New Guinea

India

Pakistan

Nigeria Ghana

Burkina Faso

Greece

Mexico

Colombia

Syria

F

15.2 Climate 469

Overall, Figure 15.6 shows that malaria ecology is the dominant factor in ex- plaining the actual incidence of malaria. This finding is profoundly bad news for the development prospects of tropical countries. For countries such as Pakistan and India, with moderate levels of malaria ecology, there are hopeful examples in which countries that have a similar climate have eliminated the disease. But for countries with a very high malaria ecology index, there are almost no examples of the successful eradication of the disease. The only country with an extremely fa- vorable malaria ecology that has wiped out the disease, Mauritius, is a special case. The environment of Mauritius is so favorable to malaria that in 1867 an epidemic of the disease killed 10% of the country’s population. However, Mauritius has the advantage of being a small island of only 788 square miles (2,040 square kilome- ters). Being small made it relatively easy for the country to eradicate malaria by draining swamps and spraying DDT. Once malaria was eliminated, Mauritius did not face the common problem of the reintroduction of the disease from neighbor- ing areas because it was cut off by the ocean.13

Climate and Human Effort A final aspect of climate’s effect on economic growth is associated with the effort that people put into their work. French philosopher Montesquieu observed in 1748 that “people are more vigorous in cold climates.”14 Montesquieu’s own expla- nation for this phenomenon—involving nervous juices, bodily fibers, humors (flu- ids thought to affect health and temperament), and other aspects of 18th-century medical theory—has been discredited, as has been his belief that a warm climate makes a person more sensitive to both pleasure and pain. But there is a more straightforward explanation for the relationship between temperature and vigor, based on human physiology. Simply put, people in warm climates cannot work hard because they will overheat. Three-quarters of the energy released by a work- ing muscle takes the form of heat, which must be dissipated in order for the body to go on functioning. In a warm climate—especially a warm, humid climate, where the evaporation of sweat cannot keep the body cool—people must work slowly if they are to survive.

The relationship between climate and effort can be modified by technology. Humans, after all, are tropical creatures. In their natural state, humans are far more vulnerable to exposure during winters in the temperate zone than to overheating in the tropics. This vulnerability to cold is the reason that humankind spent most of its history in the tropics. Technological progress in the form of clothing, shelter, and fire allowed humans to leave Africa and populate the rest of the planet. So suc- cessfully did preindustrial technology keep people warm that human habitation

13World Health Organization (2003), Gallup and Sachs (2001). 14Montesquieu (1914), Book XIV.

470 CHAPTER 15 Geography, Climate, and Natural Resources

spread as far north as the Arctic Circle. But technological development in the area of temperature control has been asymmetric: We have been able to warm ourselves for thousands of years, but only in the last century, following the invention of the air conditioner in 1902, did we learn to cool ourselves. Even now, cooling technol- ogy is much more cumbersome and much less portable than heating technology in the form of clothing.

Because of its high cost, air-conditioning has had most of its impact so far in developed countries rather than poor countries in the tropics. In the United States, for example, air-conditioning has contributed to the rise of the “Sun Belt” in the decades since 1950, when air-conditioning started to come into widespread use. The fraction of the U.S. population living in the South, having declined in the first half of the 20th century, rose from 31% in 1950 to 37% in 2010. Cities such as Houston, Atlanta, Phoenix, and Las Vegas owe much of their rapid growth to the fact that air-conditioning made them livable year-round. Many pundits have ar- gued that the air conditioner also deserves some of the credit for the growth of big government in the United States after World War II. Before the invention of air- conditioning, Washington, D.C. was so miserable in the summer that the federal government almost closed down for several months each year.

In most tropical countries, with a few exceptions like Singapore, air- conditioning has thus far had little impact on worker productivity. Ironically, as tropical countries grow wealthier, the last occupation that will benefit from air- conditioning is the occupation that is most prevalent in the tropics: agriculture. It is much easier to air-condition a factory or an office than it is to keep a farmer cool out in the fields.

15.3 NATURAL RESOURCES One seemingly obvious geographic determinant of income is the presence or absence of natural resources. Output is not produced by capital, labor, human capital, and technology alone: Also needed are natural resources such as farm- land, forests, and minerals, which are combined with capital and labor to produce output. It seems obvious that countries with more natural resources per capita should be richer than those without resources. Yet the relationship between re- source endowments and income has proved to be more complex than one would initially think.

The Relationship between Natural Resources and Growth Until the 19th century, the most important natural resource for determining economic growth was fertile land. Although European colonists were first drawn to the New World in pursuit of gold and silver, the availability of abundant land

15.3 Natural Resources 471

is what drove most of the economic growth of the Americas. The importance of land as a natural resource is not surprising because, as we saw in Chapter 3, land was more significant than capital as a factor of production (and earned a higher fraction of national income) before industrialization. The newly settled, land- abundant countries were among the richest in the world in the 19th century. In 1870 Australia, the United States, and Canada had higher real wages than any country in Europe, and Argentina’s real wages were higher than those of any European country other than Great Britain. Not surprisingly, these land-rich countries were magnets for immigration: In the hundred years following 1820, some 60 million Europeans set sail for the New World.15 Along with this flow of labor came an inflow of European capital. As we saw in Chapter 11, financial markets were globalized in the 19th century, and a good deal of the investment that took place in the Americas was financed in Europe. Thus, growth of many countries in the Western Hemisphere was very much “resource driven.” Observers of the time took it for granted that a country’s bounty of natural resources was a primary determinant of its economic success.

And yet, for all the countries that grew rich from natural resources, in many others, abundant natural resources did not result in economic growth. The sugar-producing islands of the West Indies, which at the time of the American Revolution were the richest parts of the British Empire in America, remained economically stagnant after 1800. Similarly, Mexico and Peru were the most im- portant Spanish New World colonies because of their rich endowments of gold and silver, but they failed to move beyond resource extraction into sustained growth. During the 19th century, there were booms in the production of cotton (Egypt), copper (Chile), sugar (Cuba), coffee (Brazil, Colombia, and Costa Rica), guano (Peru), and nitrate (Chile), all of which failed to lay the foundation for long-term growth.

In the post–World War II period, the relationship between economic growth and natural resource endowments has also been uneven. Many countries—most notably, the oil-exporting countries of the Persian Gulf—have grown rich from natural resources. But there are also prominent examples of countries that have grown rich despite being exceedingly poor in resources. Conversely, many coun- tries have remained poor despite a generous endowment of resources.

To assess more systematically the role of natural resources in determining countries’ income, we use World Bank data that measure each country’s natu- ral capital. Natural capital is the value of a country’s agricultural lands, pasture lands, forests, and subsoil resources, including metals, minerals, coal, oil, and natural gas.16 Unlike physical and human capital, natural capital is not created by

15O’ Rourke and Williamson (1999). 16World Bank (2006).

472 CHAPTER 15 Geography, Climate, and Natural Resources

deliberate investment. Rather, a country’s natural capital represents the resources that exist irrespective of human activity.

As shown in Figure 15.7, natural capital and GDP per capita are positively related: Countries with more natural resources tend to have higher income. (Several wealthy, oil-rich countries such as Saudi Arabia, Kuwait, and Qatar are omitted from the figure because of lack of data.) However, there are many exceptions to the rule that resources make a country richer. Japan, South Korea, and Belgium are among the world’s most resource-poor nations but are also among those with the highest income per capita. Similarly, Syria, Venezuela, Ecuador, and the Republic of Congo all have low income relative to their levels of natural resources. Thus the figure suggests that natural resources are helpful to economic growth but are neither necessary nor sufficient to achieve growth.

Analyses similar to Figure 15.7 but looking at the growth of income rather than the level of income have come to a conclusion that is even less favorable to natural resources. For example, one study examined the role of natural re- sources by calculating the share of total national wealth (defined as natural plus physical plus human capital) that consisted of natural capital. The author found

FIGURE 15.7

Natural Capital versus GDP per Capita

Sources: World Bank (2006), Heston, Summers, and Aten (2011).

100

1,000

10,000

100,000

100 1,000 10,000 100,000 Natural capital per capita, 2000 (ratio scale)

GDP per capita, 2009 (2005 Dollars, ratio scale)

Norway

New Zealand

Venezuela

Republic of Congo

Syria

Nigeria

South Korea Belgium

Japan

Mauritius Gabon

United States

Ecuador

Gambia

Bangladesh

F

15.3 Natural Resources 473

that countries where natural capital made up a large fraction of national wealth grew significantly more slowly over the period 1965–1998 than did countries where natural capital made up a small share of national wealth. Similarly, another study found that countries that relied on natural resource exports grew more slowly between 1970 and 1990 than those that did not rely on such exports.17

These studies present a puzzle. Surely, a bounty of natural resources should, other things being equal, make a country better off. Yet the effect of natural re- sources on income is weak at best. In the next section, we explore potential solu- tions to this puzzle.

Explanations for the Resource Curse As we have seen, an abundance of natural resources is often a spur to economic growth. At the same time, many countries’ experience with natural resources has been surprisingly disappointing. This negative experience has led some observers to conclude that in the long run, the presence of natural resources can actually im- pede economic growth—in effect, that there is a “resource curse.”

Where could such a resource curse come from? In Chapter 14, we identified one possible channel: that countries rich in natural resources do not develop the cultural attributes necessary for economic success. In addition to this culture- based explanation, three other prominent theories attempt to come to grips with this phenomenon, one focusing on the level of saving, one on the process of indus- trialization, and one on politics.

Overconsumption. One explanation for the resource curse begins with the observation that booms in income resulting from natural resources tend to be temporary. The discovery of a new resource or a sudden increase in the price of a resource will lead to a spurt in a country’s export earnings. Eventually, however, ei- ther the resource runs out, or its price in the world market decreases as importing countries find a substitute or as new sources of supply come on line.

Windfalls of income associated with resource booms seem to lead coun- tries to raise their consumption to levels that cannot be sustained once the boom ends. As a result, countries in which booms have ended experience low rates of saving (and thus investment) in an effort to maintain their consump- tion. Because economic growth requires investment, the final result is a lower level of income after a temporary resource boom than if the boom had never taken place.

In some cases, the situation is even worse. During a resource boom in de- veloping countries, their governments sometimes assume that revenue from the

17Gylfason (2001), Sachs and Warner (2001).

474 CHAPTER 15 Geography, Climate, and Natural Resources

resource not only will remain high but will keep on rising. In anticipation of higher future revenues, such governments borrow money on the world market, often to fund ill-conceived public investment projects. When the resource boom ends, these countries are left with unsupportable overhangs of foreign debt.

What is bad for growth is not the presence of natural resources per se, but rather the interaction of natural resource abundance with some other factor. In the case of a temporary resource boom, a country whose people or government was sufficiently forward-looking could take a large fraction of the resource windfall and invest it productively for the future. Thus, the country could end up richer in the long run.18

Dynamics of Industrialization. A second explanation for the resource curse is that natural resources distort the structure of an economy in a way that produces short-run benefits but long-run costs. A country with natural resources to export will import other products, generally manufactured goods, for its own consump- tion. The shift to importing manufactured goods will cause the country’s own manufacturing sector to contract. In the short run, this contraction represents an efficient adjustment. However, manufacturing industries are those with the most rapid technological progress. A country that imports its manufactures will miss out on this progress and may be worse off in the long run than a country that lacks a natural resource to begin with. The process by which the presence of a natural resource is ultimately detrimental to the domestic manufacturing sector is called the Dutch disease because it was first analyzed when the development of large natural gas fields off the coast of Holland in the early 1960s led to a contraction of that country’s manufacturing sector.

One of the most prominent historical examples of the Dutch disease is the experience of Spain after Europe’s discovery of the New World. Spain became fantastically rich from the inflow of gold and silver from the Americas. This in- flow had the same effect on the Spanish economy as if the resources had actually been located in Spain itself. Spain traded its gold and silver to the rest of Europe in exchange for manufactured goods. But when the flow of gold and silver ran out, other countries had gained experience and knowledge in production, and Spain was left an economic backwater.

One implication of this theory is that the effect of resources on a country’s economic growth will depend crucially on the degree to which the exploitation of a natural resource stimulates or impedes production in other sectors of the economy. Resource extraction can stimulate production in other sectors through backward linkages, in which locally produced goods are used as inputs by the re- source extraction industry, and forward linkages, in which the natural resource is processed or used to produce other goods. When backward and forward linkages

18Rodriguez and Sachs (1999).

15.3 Natural Resources 475

are present, the exploitation of a natural resource can drive the development of the economy as a whole. Backward linkages may provide enough demand to allow for the creation of industries for which the country otherwise would have been too small. Similarly, countries may start up the ladder of industrial development by processing their own natural resources. Finally, industrial development resulting from backward or forward linkages can stimulate growth in other economic sec- tors. For example, a railway line built to take iron ore from a mine in a country’s interior to a port on the coast can also link farmers in the country’s interior to world markets.

Whether exploiting a given natural resource will result in backward and forward linkages that promote economic growth depends on the nature of the resource, transport costs, and the state of the economy in which the resource is found. In the case of a land-rich country such as the United States in the 19th century, the export of grain, a resource-intensive product, involved a rich set of forward and backward linkages. The economy required agricultural machinery to harvest crops, railroads to move them to ports, and a banking system to finance the whole operation. Through these linkages, resources contributed to the indus- trialization of the United States itself.

Ironically, one consequence of the reduction in transport costs over the last 200 years is that much of the processing of natural resources now occurs in a place other than where the resources are located. In the late 19th century, the United States, Britain, and Germany all experienced rapid growth in their steel industries owing to the availability of domestic iron and coal deposits. By con- trast, after World War II, Japan and South Korea became major steel producers using imported raw materials. For resource-producing countries, the reduction in transportation costs means that the production of natural resources is often unac- companied by significant forward or backward linkages. Some resource exports in the developing world today take the form of enclaves—small pockets of economic development that have almost no contact with the rest of the country’s economy. The most extreme version of such an enclave is offshore oil production, in which capital and workers are imported and oil is pumped and exported without any contact with the local economy.

Politics. The theories of why resources may not lead to development we’ve con- sidered so far—overconsumption, the Dutch disease, and lack of linkages—all suggest that government actions can play a role in turning the resource curse into a blessing. Government policy, for example, can help to ensure that the windfall in- come from a commodity boom is used for investment or put aside for a rainy day rather than being used for consumption. Similarly, because many resource extrac- tion industries are owned by a government itself, government policy can be used to establish forward and backward linkages between resource extraction and other sectors of the economy. In addition, governments can collect taxes on resource exports and use this revenue to provide a public good such as infrastructure or to

476 CHAPTER 15 Geography, Climate, and Natural Resources

invest in education. Thus, we might expect that, with proper government policy, the presence of natural resources could be a significant boon to growth.

The fact that natural resources frequently do not support growth suggests, then, that governments are not undertaking effective policies. But many observers go one step further, arguing that the presence of natural resources actually makes governments undertake worse policies than they otherwise would. That is, natural resources may have a toxic effect on the political system.

The negative effects of natural resources on government policy come through two channels. First, natural resources often lead to an overexpansion of the gov- ernment sector of the economy. This ballooning of government occurs both because resources provide a ready source of government revenue and because large government is often a means by which the revenue from natural resources is distributed, legally or otherwise, to powerful groups within a country. Second, by increasing the revenue that the government can distribute to favored groups, the presence of natural resources raises the stakes in the struggle for control of the government, thus encouraging people to put more effort into maintaining or seiz- ing power.

The following description of Venezuela during the oil boom of the 1970s cap- tures the typical ways in which natural resource revenue is redistributed to favored groups:

In this climate, massive government contracts were awarded outside of regular procurement procedures; large sums of money passed through state agencies with- out controls; millions of dollars worth of loans were granted without regulation. Previously existing forms of illegality were exacerbated: for example, the juggling of bank accounts, the awarding of contracts without a public bidding process, the pri- vate purchase of properties with public monies, the diversion of budgetary funds for purposes other than their allocation, the awarding of commissions on loans and con- tracts, [and] the issuance of large loans without sufficient security.

From 1970 to 1978, the size of the public sector of Venezuela’s economy (gov- ernment and government-owned firms) doubled, and the country’s foreign debt rose by a factor of nine.19

The struggle for the control of government that results from these opportu- nities to grab a share of natural resource revenue takes a variety of forms, from political corruption to civil war to foreign invasion. The 10-year war that displaced a third of Sierra Leone’s population in the 1990s, fought over and funded by lucra- tive diamond exports, is one of the worst cases of this sort of struggle. In countries in which resources are distributed unevenly along ethnic lines, the situation is even more terrible. For example, in Sudan a 20-year civil war pitting the politically dominant Muslim north of the country against the non-Muslim, oil-rich south

19Karl (1997), p. 146, Table 8, and Table A.9.

15.3 Natural Resources 477

killed 1.9 million civilians. (A peace treaty was signed in January 2005, and the new country of South Sudan was established in 2011.)

The toxic effects of natural resources on the political system are most se- vere in the case of resources associated with large economic rents. Recall from Chapter 10 that an economic rent is a payment to a factor of production that is in excess of what is required to elicit the supply of that factor. Where there are large economic rents, the benefits of rent seeking are larger, and so, correspond- ingly, are the resources wasted on such activities. Resources produce large rents if they are easily extracted relative to their price. For example, in Saudi Arabia the cost of producing oil, inclusive of capital expenditures, is in the neighborhood of $4-6 per barrel.20 In early 2012, that oil was selling for more than $100 per barrel. Similarly, resources produce rents if extraction is done by foreigners; for example, when oil is produced by platforms off the coast of Angola, the foreign companies that pump the oil simply send royalty payments to the Angolan government, without requiring anyone in Angola to do anything. By contrast, resources such as fertile farmland produce smaller (or no) rents because getting value out of farmland requires capital and labor.

Of course, the presence of rent-producing natural resources does not neces- sarily lead to political problems. In Norway, for example, large oil exports have fattened the government’s coffers, but the country has remained free of the rent seeking, corruption, and mismanagement that have plagued most oil exporters. In addition, Norway has managed to put much of its extra income aside as a reserve to be used when oil revenues fall. The southern African country of Botswana pro- vides another example in which government has apparently been immune to the toxic effects of natural resources. Despite a heavy reliance on diamond produc- tion (which accounts for 33% of the country’s GDP), Botswana has maintained a stable, democratic, efficient, and honest government. In the 43 years follow- ing independence in 1966, GDP per capita grew at an annual rate of 6.0%. Why has Botswana been able to profit from natural resources whereas so many other African countries have failed? Among the explanations that have been put forward are Botswana’s high degree of ethnic homogeneity, the survival of precolonial tribal institutions that constrained political elites and allowed for the expression of dissent, and the exceptional talent and lack of greed of Botswana’s first postinde- pendence leader, Seretse Khama.21 The examples of Norway and Botswana imply an important interaction between the presence of natural resources and other factors—for example, culture and human capital—that also influence the quality of government. Perhaps once a country is rich enough or has a long enough tradi- tion of honest government, it is relatively immune to the toxic effects of natural resources on the political system.

20Reuters (2009). 21Acemoglu, Johnson, and Robinson (2003), Beaulier (2003).

478 CHAPTER 15 Geography, Climate, and Natural Resources

In an effort to moderate the most toxic effects of resources on development, the Extractive Industries Transparency Initiative (EITI) was created in 2002. The initiative brings together corporations in the oil, gas, and mining industries with governments of resource producing countries. Corporations agree to report all payments (taxes, royalties, bonuses) made to governments in resource pro- ducing countries and to submit to external auditing of their accounts, whereas governments agree to disclose all revenues received from corporations. An in- dependent group in each country is then appointed to reconcile the two sets of reports. As of 2011, 60 of the world’s largest oil, gas, and mining companies, and 35 governments, mostly in Africa, had signed on to the initiative. The hope is that transparency will prevent the worst abuses associated with resource extraction. Corporations will not be able to disguise bribe payments as other items in their ac- count books. Similarly when citizens know exactly how much revenue was turned over to their governments, they will be in a position to demand that the money be used well. Whether the EITI is successful remains to be seen, however. Pessimists argue that with prices for the extractive products rising once again, the incentives faced by corporations to cut corners to secure supplies will simply be too great to resist. Further, the entry into the market of large Chinese buyers, which may not be constrained by scruples that western governments impose on their corporations, may undo some of the benefits of the EITI.

15.4 CONCLUSION Is geography (along with climate and natural resources) destiny? Because a country’s geography, climate, and natural resources are not determined by human activity, these characteristics are particularly attractive as fundamental explana- tions for differences in economic outcomes among countries. It is intellectually more satisfying to say that “Country X developed before Country Y because of dif- ferences in these fixed characteristics” than it is to cite accidents of history. In this chapter we have seen strong evidence that geographic and climatic characteristics are related to countries’ economic outcomes. In the case of natural resources, the evidence is much weaker.

Still, some obstacles stand in the way of concluding that geography and climate are destiny. To see the difficulties, we can return to the chapter’s opening example of latitude and income per capita. Countries near the equator have more disease-prone and enervating climates, are generally farther from the centers of economic activity, and have lower agricultural productivity than countries at higher latitudes. These characteristics might be underlying channels that explain the strong correlation be- tween income per capita and distance from the equator. Many of these explanations would lead to the depressing conclusion that countries near the equator will never catch up to the developed world. But even with these potential explanations for the link between latitude and income, we still cannot be fully confident that proximity to the equator necessarily explains why these countries are poor. For every one of

15.4 Conclusion 479

the channels we considered as a way in which geography can affect income, we can find exceptions. For example, even though most economic activity is conducted near seacoasts and in temperate climates, the fastest-growing city in the United States is Las Vegas, which is located in a desert, far inland. Similarly, the prevalence of disease in tropical areas has not always been a barrier to economic achievement. For example, although the French attempt to build a canal across Panama in the 1880s was defeated by yellow fever and malaria, at the cost of some 20,000 lives, in a subsequent effort by Americans, these diseases were eradicated before construction took place. Indeed, even the relative unhealthiness of the tropics is not an exogenous factor for certain. One reason that the tropics are so unhealthy for humans is that less money has been spent on studying tropical diseases than on studying diseases of temperate climates. This imbalance in spending is easily explained by the fact that rich countries—whose citizens buy new medicines and whose governments spon- sor research—are located in temperate climates. Similarly, one reason that tropical agriculture is relatively unproductive is that agricultural research and development (R&D) is concentrated in temperate zones, so the technologies developed in these regions are not appropriate for the tropics.

Given these doubts, there is always the possibility that our explanations for the poverty of tropical countries are after-the-fact rationalizations. Perhaps if Sweden were poor rather than rich, economists would blame its poverty on the difficulties produced by cold weather and snow or the depressing effects of long winters.

An even more difficult question is what the existing relationship between lati- tude and income per capita tells us about income growth in the future. The example of the air conditioner altering the pace of economic activity in the U.S. South sug- gests that technological progress can significantly change the relationship between geographic factors and economic growth. It is also possible that the climate-income correlation we observe reflects the circumstances of history rather than any delete- rious effects of climate on health or productivity today. We saw in Chapter 12 that climate was an important determinant of the kind of government Europeans estab- lished in their colonies. In climates in which disease took a heavy toll on European settlers, colonial powers tended to create “extractive” institutions. Conversely, in climates in which the mortality of Europeans was lower, permanent settlers brought with them European forms of government. The institutions created by colonial powers tended to carry over into the postcolonial period, endowing tropical former colonies with bad governments, which in turn stifled economic growth. If this mech- anism is the primary explanation for the relationship between climate and income in the world today, then tropical countries that manage to improve their governing institutions will not experience any negative effect of climate on economic growth.22

One of this chapter’s most important lessons is that a country’s natural re- sources have only a limited effect on economic growth. Countries that lack many of the raw materials necessary to produce output in a modern economy can

22Acemoglu, Johnson, and Robinson (2001).

480 CHAPTER 15 Geography, Climate, and Natural Resources

technology were not advanced enough to exploit Chinese coal deposits until long after the begin- ning of industrialization. By contrast, Europe’s largest coal deposits reside in Britain, the coun- try that led the Industrial Revolution. British coal is located near centers of economic activity that predated the Industrial Revolution and is easily accessible to canals and ocean transport.

The second piece of bad luck for China derives from differences in the geology of the Chinese and British coal mines. Because British coal mines are susceptible to flooding, constant effort has to be applied to pump water out of them. This coincidence of a great need for me- chanical energy for pumping at the same location as a cheap source of fuel ensured a ready demand for coal-fired steam engines. Indeed, early steam engines were so woefully inefficient in their use of energy and so bulky that they had no prac- tical use other than pumping out coal mines. Only after several decades, as experience with steam power produced a series of technologi- cal improvements raising energy efficiency and reducing engine size, did the steam engine move beyond coal mines to become the dominant source of power for the Industrial Revolution.

In contrast to the British mines, Chinese coal mines are dry, and the greatest danger is not flooding but spontaneous combustion. To prevent spontaneous combustion, the Chinese developed sophisticated ventilation technology. Unlike the steam engine, however, this technology did not have uses in other parts of the economy.*

*Pomeranz (2000).

M uch of the evidence in this chapter points to the conclusion that the presence of natural resources is not essential for economic growth in the world today. Simply put, countries can make up for any resources they lack by importing them from abroad. A different issue is whether the presence or absence of natural resources helps to explain the history of eco- nomic growth. In this area, many scholars argue that natural resources help to shed light on one of the most puzzling phenomena in eco- nomic history: the emergence of the Industrial Revolution in Europe rather than China, despite the latter’s longer history of economic integra- tion and, for many centuries, its more advanced level of technology.

The particular natural resource underlying this theory is coal. In Chapter 9 we saw the central role that coal played in the Industrial Revolution in Europe. Coal was the fuel that allowed for a vast increase in iron production starting in the 18th century. Even more significantly, coal-fired steam engines powered factories and drove the railways and ships that produced such a spectacular in- crease in trade in the 19th century.

China has vast coal deposits and today is the world’s largest coal producer. But from the perspective of early industrialization, China had two pieces of bad luck involving coal. The first disadvantage was the location of the coal. China’s deposits are primarily in the north- west of the country, particularly in the province of Shanxi, far from centers of population and economic activity, and are inaccessible to any easy water transport. Engineering and railroad

FRESOURCES AND EARLY INDUSTRIAL DEVELOPMENT: THE CASE OF COAL

nonetheless grow rapidly, and countries that are well endowed with resources of- ten grow slowly. The explanation for this surprising finding is simply that many natural resources are easy to transport across country borders.

The observation that the resources available to a given country do not deter- mine its growth tells us nothing about whether the resources available to the world

Problems 481

as a whole will affect global growth. Unlike individual countries, the world as a whole cannot import natural resources, at least for the foreseeable future. We re- turn to the issue of natural resources at the world level in Chapter 16.

K E Y T E R M S

spillovers 458 maquiladora 458 natural capital 471

Dutch disease 474 backward linkages 474 forward linkages 474

enclaves 475

Q U E S T I O N S F O R R E V I E W

1. Why are geography, climate, and natural re- sources particularly interesting to study as possible determinants of economic growth? How do these characteristics differ from other factors affecting economic growth, such as quality of government, income inequality, and culture?

2. What evidence do we have that proximity to ocean transport is an important contributor to economic growth?

3. How did the availability of useful plant and animal species affect economic development in different regions before the 16th century? Why were Europe and Asia advantaged in this regard?

4. How were differences in geography reflected in the political histories of Europe and China? How did these political differences in turn affect economic growth?

5. Through what channels might a country’s climate affect the country’s level of income per capita?

6. What characteristics of tropical climates are particularly hostile to economic growth?

7. What is natural capital? In what ways is it similar to physical capital and human capital? In what ways is it different?

8. What is the resource curse? What are some potential explanations for it?

P R O B L E M S

1. What does the Malthusian model presented in Chapter 4 predict will be the effects of climate-based differences in agricultural productivity on population density and on the standard of living in different geographic regions? Does this prediction match what we see in the data?

2. What does Jared Diamond’s theory regarding the role of geography in explaining preindustrial economic growth say about the different fates of Europe and China?

3. How does globalization change each of the fol- lowing relationships? In each case, cite specific examples.

a. The relationship between a country’s natural resources and its level of income per capita

b. The relationship between a country’s geo- graphic characteristics and its level of income per capita

c. The relationship between a country’s climate and its level of income per capita