7/31/2019 Jones Chapter 1

1/10

I Charles Jonesintroduction To Economic Growth 2nd EditionChapter 1

Introduction: the Facts of Economic Growth

The errors wh ich arise from the absence of facts are farmore numerous and more durable than those whichresult from unsound reasoning respecting true data.

-CHARLES BABBAGE, uoted in Rosenberg (1994,p. 27.

It is quite wrong to try founding a theory on observablemagnitudes alone.. . . It is the theory wh ich decideswhat we can observe.

-ALBERTEINSTEIN, uoted in Heisenberg (l9 7l) ,p. 63 .

peaking at the annua l meeting of the American EconomicAssociation in 1989, the renowned economic historian David S. Landeschose as the title of his address the fundamental question of economicgrowth and development : "Why Are We So Rich and They So Poor?"'This age-old question has preoccupied economists for centuries. It sofascinated the classical economists that it was stamped on the cover ofAdam Smith's famous treatise An Inquiry into the Nature and Causesof the Wealth of Nations. And it was the mistaken forecast of Thbmas~ a l t h u sn th e early nineteenth century concerning the future prospectsfor economic growth that earned the discipline its most recognizedepithet, the "dismal science."'See Landes (1990).

7/31/2019 Jones Chapter 1

2/10

2 1 I N T R O D U C T I O N : T H E F A C T S O F E C O N O M I C G R O W T H T H E D A T A O F G R O W T H A N D D E V E L O P M E N T 3The modern examination of this question by macroeconomists dates

to the 1950s and the publica tion of two famous papers by Robert Solowof the Massachusetts Institute of Technology. Solow's theories help ed toclarify the role of the accumulation of physical capital and eaphasizedthe importance of technological progress as the ultimate driving forcebehind sustained economic growth. During the 1960s and to a lesser ex-tent the 1970s, work on economic growth fl ou ri ~h ed .~or methodolog-ical reasons, however, important aspects of the theoretical explorationof technological change were p ~ s t p o n e d . ~

In the early 1980s, work a t the University of Chicago by Paul Romerand Robert Lucas re-ignited the interest of macroeconomists in eco-nomic growth, emphasizing the economics of "ideas" and of huma ncapital. Taking advantage of new developments in the theory of im-perfect competition, Romer introduced the economics of technologyto macroeconomists. Following these theoretical advances, empiricalwork by a number of economists, such as Robert Barro of Harvard Uni-versity, quantified and tested the theories of growth. Both theoreticaland empirical work has since continued with enormous professionalinterest.The purpose of this book is to explain and explore the modern the-ories of economic growth. This exploration is an exciting journey, inwhic h we encounter several ideas that have already earned Nobel Prizesand several more with Nobel potential. The book attempts to make thiscutting-edge research accessible to readers with only basic training ineconomics and ca lc ~l us .~

The approach of this book is similar to the approach scientists takein studying astronomy and cosmology. Like economists, astronomersare unable to perform the controlled experiments that are the hallmarkof chemistry and physics. Astronomy proceeds instead through an in-terplay between observation and theory. There is observation: planets,2~ far from exhaustive list of contributors includes Moses Abramovitz, Kenneth Arrow,David Cass, Tjalling Koopmans, Simon Kuznets, Richard Nelson, William Nordhaus,Edmund Phelps, Karl Shell, Eytan Sheshinski, Trevor Swan, Hirofumi Uzawa, and Carlvon Weizsacker.3Romer (1994) provides a nice discussion of this point and of the history of research oneconomic growth.4Th e reader with advanced tr aining is referred also to the excellent presentations in Barroand Sala-i-Martin (1998) and Aghion a nd Howitt (1998).

stars, and galaxies are laid out across the universe in a particular way.Galaxies are moving apart, and the universe appears to be sparselypopulated with occasional "lumps" of matter. And there is theory: thetheory of the Big Bang, for example, provides a coherent explanationfor these observations.

This same interplay between observation and theory is used to or-ganize this book. This first chapter will outline the broad empiricalregularities associated with growth and development. How rich are therich coun tries , how poor are the poor? How fast do rich and poor coun-tries grow? The remainder of the book consists of theories to explainthese observations. In the limited pages we have before us, we will notspend much time on the experiences of individual countries, althoughthese experiences are very important. Instead, the goal is to providea general economic framework to help us understand the process ofgrowth and development.

A critical difference between astronomy and economics, of course, isthat the economic "universe" can poten tiall y be re-created by economicpolicy. Unlike the watchmaker who builds a watch and then leaves itto run forever, economic policy makers constantly shape the course ofgrowth and development. A prerequisite to better policies is a betterunders tanding of economic growth.

THE DATA OF G R O W TH A N D D EVELO PMEN TThe world consists of economies of all shapes and sizes. Some coun-tries are very rich , and some are very poor. Some economies are growingrapidly, and some are not growing at all. Finally, a large number of econ-omies -most, in fact- ie between these extremes. In thinking abouteconomic growth and development , it is helpful to begin by consideringthe extreme cases: the rich, the poor, and the countries that are movingrapid ly in between. The remainder of this chapter lays out -the empiri-cal evidence -the "facts" -associated with these categories. The keyquestions of growth and development then almost naturally ask them-selves.

Table 1.1 displays some basic data on growth and developmentfor seventeen countries. We will focus our discussion of the data onmeasures of per capita income instead of reporting data such as life

7/31/2019 Jones Chapter 1

3/10

4 1 I N T R O O U C T I O # : T H E FA C TS O F E C O N O M I C G R O W T H T H E D AT A O F G R O W T H A N D D E V E L O P M E N T 5

AverageGDP per GDP per Labor force annual Yearscapita, worker, participation growth rate, to1997 1997 rate, 1997 1960-97 double

"Rich"countriesU.S.A.JapanFranceU.K.Spain

"Poor"countriesChinaIndiaZimbabweUganda

"Growth miracles"Hong Kong 18,811Singapore 17,559Taiwan 11,729South Korea 10,131

"Growth disasters"Venezuela 6,760Madagascar 577

World TablesMark 5.6, an update ok's Global Development NetworkNotes: The GDP data are in 1985 dollars. The growth rate is the average annual

change in the log of GDP per worker. A negative number i n the "Years to double" columnindicates "years to halve."

expectancy, infant mortality, or other measures of qual ity of life. Themain reason for this focus is that the theories we develop in subsequentchapters will be couched in terms of per capita income. Furthermore,per capita income is a useful "summary statistic" of the level of eco-nomic development in t he sense that it is highly correlated with othermeasures of quality of life.5

We will interpret Table 1 .1 in the context of some "facts," beginningwith the first:fi

":pry .='_ _ There is enormous variation in per capita income

across economies. The poorest countries have per capita in-comes that are less than 5 percent of per capita incomes in therichest countries.

The first section of Table 1.1 reports real per capita gross domesticproduct (GDP) in 1997, together with some other data , for the UnitedStates and several other "rich" countries. The United States was therichest country in the world in 1997, with a per capi ta GDP of $20 ,049(in 1985 dollars), and it was the richest by a s ubstantial amount. Japan,for example had a per capi ta GDP of about $16,000.

These numbers may at first seem slightly surprising. One sometimesreads in newspapers that the United States has fallen behind countrieslike Japan or Germany in terms of per capita income. Such newspaperaccounts can be misleading, however, because market exchange ratesare typically used in the comparison. U.S. GDP is measured in dollars,whereas Japanese GDP is measured in yen. How do we convert theJapanese yen to dollars i n order to make a comparison? One way is touse prevailing exchange rates. For example, in January 1997, he yen-to-dollar exchange rate was around 120 yen per dollar. However, exchangerates can be extremely volatile. Just a little over one year earlier, the ratewas only 100 yen per dollar. Which of these exchange rates is "right"?5Se& for example, the Wor ld Bank's World Development Report. 1991 (New York: Oxfor dUniversity Press, 1991)."any of these facts have been discussed elsewhere. See especi ally Lucas (1988) an dRomer (1989).

7/31/2019 Jones Chapter 1

4/10

6 1 I N T R O O U C T I O N : T H E F A C T S O F E C O N O M I C G R O W T H T H E O AT A O F G R O W T H A N 0 O E V E L O P M E N T 7Obviously, it matters a great deal which one we use: at 100 yen perdollar, Japan will seem 20 percent richer than at 120 yen per dollar.

Instead of relying on prevailing exchange rates to make internationalcomparisons of GDP, economists attempt to measure the actual valueof a currency i n terms of its ability to purchase similar products. Theresulting conversion factor is sometimes called a purchasing powerparity-adjusted exchange rate. For example, the Economist magazineproduces a yearly report of purchasing power parity (PPP) exchangerates based on the price of a McDonald's Big Mac hamburger. If a Big Maccosts 2 dollars in the United States and 300 yen in Japan, then the PPPexchange rate based on the Big Mac is 150 yen per dollar. By extendingthis method to a number of different goods, economists construct a PPPexchange rate that can be applied to GDP. Such calculations suggestthat 150 yen per dollar is a much better number than the prevailingexchange rates of 100 or 120 yen per d ~ l l a r . ~

The second column of Table 1.1 reports a related measure, real GDPper worker in 1997. The difference between the two colum ns lies in t hedenominator: the first column divides total GDP by a country's entirepopulation, while the second column divides GDP by only the laborforce. The third column reports the 1997 labor force participa tion rate-the ratio of the labor force to the population -to show the relationshipbetween the first two columns. Notice that while Japan had a higher percapita GDP than France in 1997, the comparison for GDP per workeris reversed. The labor force participation rate is much higher in Japanthan in the other industrialized countries.

Which column should we use i n comparing levels of development?The answer depends on what question is being asked. Perhaps percapita GDP is a more general measure of welfare in that it tells us howmuch output per person is available to be consumed, invested, or put tosome other use. On the other hand, GDP per worker tells us more aboutthe productivity of the labor force. In this sense, the first statistic canbe thought of as a welfare measure, while the second is a productivitymeasure. This seems to be a reasonable way to interpret these statistics,but one can also make the case for using GDP per worker as a welfaremeasure. Persons not officially counted as being in the labor force maybe engaged in "home production" or may work in the underground7Economist,April 19, 1995, p. 74

econoinj7. Neither of these activities is included in GDP, and in this casemeasured output divided by measured labor input may prove moreaccurate for making welfare comparisons. In this book, we will oftenuse the phrase "per capita income" as a generic welfare measure, evenwhen speaking of GDP per worker, if the context is clear. Whatevermeasure we use, though, Table 1.1 tells us one of the first key thingsabout economic development: the more "effort" an economy puts intoproducing out put, the more output there is to go around. "Effort" in thiscontext corresponds to the labor force participation rate.

The second section of Table 1.1 documents the relative and evenabsolute poverty of some of the world's poores t economies. India andZimbabwe had per capita GDPs around $1,500 in 1997, less than 10percent of that in the United States. A number of economies in sub-Saharan Africa are even poorer: per capita income in the United Statesis more than 40 times higher than income in Ethiopia.

To place these numbers in perspective, consider some other statis-tics. The typical worker in Ethiopia or Uganda must work a month anda half to earn what the typical worker in the United States earns in aday. Life expectancy in Ethiopia is only two-thirds that in the UnitedStates, and infant mortality is more than 20 times higher. Approximately40 percent of GDP is spent on food in Ethiopia, compared to about 7percent in the United States.

What fraction of the world's population lives with this kind ofpoverty? Figure 1.1 answers thi s question by plotting the distribution ofthe world's populati on in terms of GDP per worker. In 1995, more thanhalf of the world's population lived in countries with less than 10 per-cent of U.S. GDP per worker. The bulk of this popu lation lives in onlytwo countries: China, wi th nearly one-quarter of the world's population ,and India, with one-six th of the world's popula tion. Together, these twocountries account for more than 40 percent of the world's population.In contrast, the 39 countries that make up sub-Saharan Africa constituteabout 10 percent of the world's population. .

Figure 1.2 shows how this distribution has changed since 1960.Overall, the distribution h as equalized as the sha re of the world's popu-lation living in coun tries whose GDP per worker is less than 30 percentEf that in the United States has fallen. Of the poorest countries, bothChina and India have seen substantial growth in GDP per worker, evenrelative to the United Stat es. China's relative income rose from 4 percent

7/31/2019 Jones Chapter 1

5/10

1 INTRODUCTION: THE FACTS OF ECONOMIC GROWTH THE DATA OF 6ROWTH AND DEVELOPMENT 9CUMULATIVE DISTRIBUTIO N O F WO RLD PO PULATIO N BYG D P P E R W O R K E R , 1 9 9 5

PERCENTAGEOF WORLD

POPULATION

90

0 20 40 60 80 100GDP PER WORKER

RELATIVETO US.

SOURCE: Penn World Tables Mark 5.6,Summers and Heston (1991), pdated usingEasterly and Yu (2000).

Note: A point (x, ) n the figure indicates that the fraction of the w orld's populationliving in countries with a relative GDP per worker less thanx s equal to y. 136 countriesare included.

of U.S. GDP per worker in 1960 o 9percent in 1995, nd India's relativeincome rose from 7 percent of U.S. GDP per worker to 10percent overthe same period.

The third section of Table 1.1 reports data for several countries thatare moving from the second group to the first. These four so-callednewly industrializing countries (NICs) are Hong Kong, Singapore, Tai-wan, and South Korea. Interestingly, by 1997 Hong Kong had a per

WO RLD PO PULA TIO N BY G O P PER WO RKE R, 1960 AND1995

PERCENTAGE IOF WORLD

POPULATION

GDP PER WORKERRELATIVE TO US.

SOURCE: Penn World Tables Mark 5.6,Summers and Heston (1991), pdated usingEasterly and Yu (2000).

Note: The sample size has been reduced to 114 countries in order to incorporate the1960 data.

capita GDP of $18,811, igher than all of the industrialized countries inthe table except for the United States. This per capita GDP was almosttwice that of South Korea. However, as with Japan, Hong Kong's highper capita GDP is driven to a large extent by its high labor force partic-ipat ion rate. In terms of GDP per worker, Hong Kong comes in belowthe other industrialized economies. Singapore, on the other hand, hasa GDP per Gorker of $36,541, ne of the highest in the world.

An important characteristic of these NICs is their extremely rapidrates of growth, and this leads to our next fact:

7/31/2019 Jones Chapter 1

6/10

10 1 I N T R O D U C T I O N : T H E F A C TS O F E C O N O M I C G R O W T H T H E D A T A O F G R O W T H A N D D E V E L O P M E N T 11- -

Rates of economic growth vary substantially acrosscountries.

The last two columns of Table 1.1 characterize economic growth.The fourth column reports the average annual change in the (natu-ral) log of GDP per worker from 1960 to 1997.' Growth i n GDP perworker in the United States averaged only 1.4 percent per year from1960 to 1997. France, the United Kingdom, and Spain grew a bit morerapidly, while Japan grew at a remarkable rate of 4. 4 percent. The NICsexceeded even Japan's astounding rate of increase, truly exemplifyingwhat is meant by the term "growth miracle." The poorest countries ofthe world exhibited varied growth performance. China and India, forexample, grew substantially faster than the United States from 1960to 1997, but their growth rates were well below those of the NICs.Other developing countr ies such as Zimbabwe and Uganda experiencedlittle or no growth over the period. Finally, growth rates in a number ofcountries were negative from 1960 to 1997, earning these countries thelabel "growth disasters." Real incomes actually declined in countriessuch as Venezuela, Madagascar, and Chad, as shown in the last panelof Table 1.1.

A useful way to interpret these growth rates was provided by RobertE. Lucas, Jr., in a paper ti tled "On the Mechanics of Economic Devel-opment" (1988). A convenient rule of thumb used by Lucas is that acountry growing at g percent per year will double its per capita incomeevery 70/g years.g According to this rule, U.S. GDP per worker willRSeeAppendix A for a discussion of how this concept of growth relates to percentagechanges.YLet~ ( t )e per capita income at time t and It?t yo be some initial value of per capitaincome. Then y ( t )= yo&'. The time it takes per capita income to double is given by thetime t* at which y ( t )= 2yo.Therefore,

zyu = y,)2''log 2* t * = -.g

The rule of thumb is established by noting that log2 = .7. See Appendix A for furtherdiscussion.

double approximately every 50 years, while Korean GD P per worker willdouble approximately every 12 years. I11 other words, if these growthrates persisted for two generations, the average American would be twoor three times as rich as his or her grandparents. The average citizenof Taiwan, Hong Kong, or South Korea would be twenty times as richas his or her grandparents. Over moderate periods of time, small dif-ferences in growth rates can lead to enormous differences in per capitaincomes.

Growth rates are not generally constant over time.For the world as a whole, growth rates were close to zero overmost of history but have increased sharply in the twentieth cen-tury. For individual countries, growth rates also change overtime.

The rapid growth rates observed in East Asia-and even the moremodest growth rates of about 2 percent per year observed throughoutthe industrialized world-are blindingly fast when placed in a broadhistorical context. Figure 1.3 illustrates thi s point by plotting a measureof world GDP per capita over the past five centuries . Notice that becausethe graph is plotted on a log scale, the s lope of each line segment reflectsthe rate of growth: the rising slope over time indicates a rise in theworld's economic growth rate.

Between 1950 and 1990, world per capit a GDP grew at a rate of 2.2percent per year. Between 1850 and 1950, however, the growth rate wasonly 0.88 percent, and before 1850 the growth rate was less than 0.2percent per year. Angus Maddison (1995) goes so far as to suggest thatduring the millennium between 500 and 1500, growth was essentiallyzero. Sustained economic growth at rates of 2 percent per year is just asmuch a modern invention as is electricity or the transistor.

As a result of this growth, the world is substantially richer todaythan it has ever been before. A rough guess is that per capita GDP forthe world as a whole in 1500 was $500 per person. Today, world percapita GDP is nearly ten times higher.

7/31/2019 Jones Chapter 1

7/10

12 1 I N T R O D U C T I O N : T H E FA C T S O F E C O N O M I C G R O W T H O T H E R " S T Y L I Z E D F A C T S " 13$ b* h * * L c Ti + " WORLD P E R CAP I TA GDP AND GROWTH RATE S , 1500-1990

PER CAPITAGDP (1985DOLLARS,

LOG SCALE)4,000

2,000iYEAR

SOURCE: Computed from Lucas (1998)and Maddison (1995).Note: The numbers above each line segment are average annual growth rates.

As a rough check on these numbers, consider the following exercise.Suppose we guess that the world, or even a particular country, hasgrown at a rate of 2 percent per year forever. This means that per capitaincome must have been doub ling every 35 years. Over the last 250 years,income wo uld have grown by a factor of about z7 ,or 128. In this case, aneconomy wi th a per capita GDP of $20,000 today would have had a percap ita GDP of just over $150 in 1750, measured at today's prices-lessthan half the per capita GDP of the poorest coun tries in the world today.It is virtually impossible to live on 50 cents per day, and so we knowthat a growth rate of 2 percent per year could not have been sustainedeven for 250 years.

For individual countries, growth rates also change over time, as canbe seen in a few interesting examples. India's average growth rate corn

1960 to 1997 was 2.3 percent per year. From 1960 to 1980, however,its growth rate was only 1 .3 percent per year; between 1980 and 1997growth accelerated to 3.5 percent per year. Singapore did not experi-ence pqticularly rapid growth until after the 1950s. The island countryof Mauritius exhibited a strong dec linein GDP per worker of 1.3 percentper year in the two decades following 1950. From 1970 to 1997, how-ever, Mauritius grew at 3.6 percent per year. Finally, economic reformsin China have had a substantial impact on growth and on the economicwell-being of one-quarter of the world's popu lation. Between 1960 and1978, GDP per worker grew at a n annual rate of 1.9 percent in China.Since 1979, however, growth has averaged 5.0 percent per year.

The substantial variation in growth rates both across and withincountries leads to an important corollary of Facts 2 and 3. It is so im-portant that we will call it a fact itself:

$ 4 A country's relative position in the world distribu-tion of per capita incomes is not immutable. Countries can movefrom being "poor" to being "rich,"and vice-versa.''

OTHER "STYLIZED FACTS"Facts 1 hrough 4 apply broadly to the countries of the world. The nextfact describes some features of the U.S. economy. These features turnout to be extremely important, as we will see in Chapter 2. They aregeneral characteristics of most economies "in the long run." .'A classic example of the latter is Argentina. At the end of the nineteenth century,Argentina was one of the richest countries in the world. With a tremendous naturalresource base md a rapidly developing infrastructure, it attrac ted foreign investment a ndimmigration on a large scale. By 1990, however, Argentina's per capita income was onlyabout one-third of per capita income in t he United States. Carlos Diaz-Alejandro (1970)provides a classic discussion of the economic history of Argentina.

7/31/2019 Jones Chapter 1

8/10

14 1 I N T R O O U C T I O N : T H E F A C T S O F E C O N O M I C G R O W T H O T H E R " S T Y L I Z E D F A C T S " 15-- --In the United Sta tes over the last century,

1. the real ra te of return to capital, r, shows no trendupwa rdor downward:

2. the sha res of income devoted to capital, rK/Y , and labor,wL/Y , show no trend; and

3. the average growth rate of output per person has been pos-itive and relatively constant over time- .e., the UnitedStates exhibits steady, sustained per capita income growth.

This styl ized fact, really a collection of facts, is drawn largely from alecture given by Nicholas Kaldor at a 1958 conference on capital accu-mulation (Kaldor 1961). Kaldor, following the adv ice of Charles Bab-bage, began the lecture by claiming that the economic theorist shouldbegin with a summary of the "stylized" facts a theory was supposed toexplain.

Kaldor's first fact- hat the rate of return to capital is roughly con-stant- s best seen by noting that the real interest rate on governmentdebt in th e U.S. economy shows no trend. Granted, we do not observereal interest rates, but one can take the nominal interest rate and sub-tract off either the expected or the actual rate of inflation to make thisobservation.

The second fact concerns payments to the factors of production,which we can group into capital and labor. For the United States, onecan calculate labor's share of GDP by looking at wage and salary pay-ments and compensation for the self-employed as a share of GDP.I1These calculations reveal that the labor share has been relatively con-stan t over time, at a va lue of around 0.7. If we are focusing on a modelwith two factors, and if we assume that there are no economic profitsin the model, then t he capital share is simply 1minus the labor share,or 0.3. These first two facts imply that the capital-output ratio, K / Y , isroughly constant in the United States."These data are reported in the National Income and Product Accounts. See. for example.the Council of Economic Advisors (1997).

.: :!y REAL PER CAPIT A GDP I N T HE UNIT ED ST ATES, 187 0-19 94PER CAPITA GDP(1990 DOLLARS,

LOG SCALE)

I I I I I I I1860 1880 1900 1920 1940 1960 1980 2000

YEAR

~OUR(IE:Maddison (1995) and author's calculations.

The third fact is a slight reinterpretat ion of one of Kaldor's stylizedfacts, illustrated i n Figure 1.4. The figure plots per capita GDP (on a logscale) for the United States from 1870 until 1994. The trend line i n thefigure rises at a rate of 1.8 percent per year, and the relative constancyof the growth rate can be seen by noting that apart from the ups anddowns of business cycles, this constant growth rate path "fits" the datavery well.

.Growth in output a nd growth in the volume of in-

ternational trad e a re closely related.

Figure 1.5 documents the close relationship between the growth in acountry's output (GDP) and growth i n its volume of trade. Here , the

7/31/2019 Jones Chapter 1

9/10

16 1 I N T R O D U C T I O N : T H E F A C T S O F E C O N O M I C G R O W T H O T H E R " S T Y L I Z E O F A C T S " 17

AVERAGEANNUAL

GROWTHRATE

OF TRADEVOLUME

0.14

0.12

0.1

0.08

0.06

0.04

0.02

0

- 0.02

G R O W T H I N T R A D E A ND G D P . 1 9 6 0 - 9 0

TCD

KOR0AN

TURTU A HKG

LSO

MOZMDG"~' b CIV

cPQMB

. -MYS SGP

GUY

1 I I I I I I I I 10 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

AVERAGE ANNUAL GROVYMRATE OF GDP

SOURCE: Penn World Tables Mark 5.6, Summers and Heston (1991).

volume of trade is defined as th e sum of exports and imports, but asimilar figure could be produced wi th either component of trade. Noticethat for many countries, trade volume has grown faster than GDP; theshare of exports and imports in GDP has generally increased around theworld since 196 0.~ '

I20n this point, it is interesting to note that the world economy was very open to in-ternational trade prior to World War I. Jeffrey Sachs and Andrew Warner (1995) arguethat mu ch of the trade liberalization since World War 11, at least until the 1980s, simplyre-establishes the global nature of markets that prevailed in 1900.

The relationship between trade and economic performance is com-plicated. Some economies, such as those of Hong Kong, Singapore,and Luxembourg, have flourished as regional "trade centers." The tradeintensity ratio- he sum of exports and imports divided by GDP-for these economies exceeds 150 percent. How is thi s possible? Theseeconomies import unfinished products, add value by completing theproduction process , and then export the re sult. GDP, of course, is gen-erated only in the second step. A substantial component of the stronggrowth performance turned in by these economies is associated with a nincrease in trade intensity.

On the other hand, trade intensity in Japan actually fell from around2 1 percent in 1960 to around 1 8 percent i n 1992 despite rapid percapita growth. And nearly all of the countries in sub-Saharan Africahave trade intensities higher than Japan's. A number of these countriesalso saw trade intensity increase from 1960 to 1990 while economicgrowth faltered.

61 -7l Both skilled and unskilled work ers tend to migratefrom poor to rich countries or regions.

Robert Lucas emphasized this stylized fact in his aforementionedarticle. Evidence for the fact can be seen i n the presence of in-migrationrestrictions in rich countries. It is an important observation becausethese movements of labor, which presumably are often very costly, tellus something about real wages. The returns to both skilled and un-skilled labor must be higher in high-income regions than in low-incomeregions. Otherwise, labor would not be willing to pay the high costsof migration. In terms of skilled labor, this raises an interesting puz-zle. Presumably, skilled labor is scarce in developing economies, andsimplg theories predict that factor returns are highest where factors arescarce. Why, then, doesn't skilled labor migrate from the United Statesto Zaire?

7/31/2019 Jones Chapter 1

10/10

18 1 I N T R O O U C T I O N : T H E F A C T S O F E C O N O M I C G R O W T H T H E R E M A I N D E R O F T H I S B O O K 19

T H E R E M A I N D E R OF T H l S B O O K answers hold the key to unlocking widespread rapid economic growth.Indeed, the recent experience of East Asia suggests that such growth hasThree central questions of economic growth and development are ex- the power to transform standards of liv ing over the course of a singleamined in the remainder of this book. generation. Surveying this evidence in the 1985 Marshall Lecture at

The first question is the one asked at the beginning of this chapter: Cambridge University, RobertE. Lucas, Jr., expressed the sentiment thatWhy are we so rich and they SO poor? It is a question about levels of fueled research on economic growth for the next decade:development and the world distribution of per capita incomes. Thistopic is explored in Chapters 2 and 3 and then is revisited in Chapter 7.

The second ques tion is , What is the engine of economic growth?How is it that economies experience sustained growth in output perworker over the course of a centu ry or more? Why is it that the UnitedStates has grown at 1.8 percent per year since 1870? The answer to thesequestions is technological progress. Understanding why technologicalprogress occurs and how a country such as the United States can exhibitsustained growth is the subject of Chapters 4 and 5.

The final question concerns growth miracles. How is it that econo-mies such as Japan's after World War I1 and those of Hong Kong, Singa-pore, and South Korea more recently are able to transform rapidly from"poor" to "rich?" Such Cinderella-like transformation gets at the heartof economic growth and development. Chapters 6 and 7 present onetheory that integrates the models of the earlier chapters.

The next two chapters depart from the cumulat ive flow of the bookto explore new directions. Chapter 8 discusses influential alternativetheories of economic growth. Chapter 9 examines the potentially im-portant interactions between natural resources and the sustainability ofgrowth. Chapter 10 offers some conclusions.

Three appendices complete this book. Appendix A reviews themathematics needed throughout the book.13 Appendix B lists a nurn-ber of very readable artic les and books related to economic growth thatmake excellent supplementary reading. And Appendix C presents acollection of the data analyzed throughout the book. The country codesused in figures such as Figure 1.5 are also translated there.

The facts we have examined i n this chapter indicate that it is notsimply out of intellectual curiosity that we ask these questions. The

I do not see how one can look at figures like these without seeing them asrepresenting possibilities. Is there some action a government of India couldtake that would lead the Indian economy to grow like Indonesia's or Egypt's?If so, what exactly? If not, what is it about the "natu re of India" that makesit so? The consequences for human welfare involved in questions like theseare simply staggering: Once one starts to think about them, it is hard to thinkabout anything else (Lucas 1988,p. 5 ) .

I3Readers with a limited ex posure o calculus, differential equations, and the mathematicsof growth are encouraged to read Append ix A before continuing with the next chapter.