Catching Up with the Economy - die-gdi.de · / 3y15 mr03 Mp 1 Wednesday Dec 15 07:23 AM LP–AER...

1

/ 3y15 mr03 Mp 1 Wednesday Dec 15 07:23 AM LP–AER mr03

Catching Up with the Economy†

By ROBERT W. FOGEL*

In his Presidential Address five years ago,Zvi Griliches (1994) called attention to the se-vere difficulties that beset current attempts tomeasure the growth of labor productivity inthe American economy. Because of these dif-ficulties, it is likely that the true rate of eco-nomic growth is substantially underestimated.The root of the problem is the difficulty inmeasuring output in the service sector whichnow represents two-thirds of the economy. Insuch sectors as health care and informationservices, the contribution to gross domesticproduct (GDP) is measured by inputs ratherthan outputs, a procedure that makes it impos-sible to gauge accurately improvements in thequality of output. Thus, in the case of com-puters, which are transforming American so-ciety, economists have been unable, so far, tofind a measurable contribution of computers tothe rise in labor productivity—an astonishingparadox.

I want to follow up on this problem of mis-measurements. My thesis is that the profession

† Presidential Address delivered at the one-hundredeleventh meeting of the American Economic Association,January 4, 1999, New York, NY.

* Center for Population Economics, Graduate Schoolof Business, and Department of Economics, University ofChicago, 1101 East 58th Street, Chicago, IL 60637, andNational Bureau of Economic Research. This address isbased on research made possible by grants from theNational Institutes of Health (AG10120), the NationalScience Foundation (SES-9114981), the Walgreen Foun-dation, and the University of Chicago. I am grateful to theUniversity of Chicago Press for permission to make useof material contained in my forthcoming book, The FourthGreat Awakening and the Future of Egalitarianism, whichwill be published in January 2000. I have also drawn ona forthcoming paper by Chulhee Lee (2000) and on con-cepts developed jointly with Dora L. Costa, Irwin H.Rosenberg, Nevin Scrimshaw, and James M. Tanner. Ihave benefited from comments and criticisms by Karen L,Brobst, Katherine A. Chavigny, Dora L. Costa, RoderickFloud, Daniel M. Fogel, Enid M. Fogel, Allison M. Gar-rett, Mark Guglielmo, James A. Heckman, Max Hender-son, D. Gale Johnson, Susan E. Jones, Chulhee Lee,Brigitte Madrian, Robert Mittendorf, Douglass C. North,Nevin Scrimshaw, and Richard Suzman.

is lagging behind the economy more than ithas to. We are, to some extent, entangled inconcepts of the economy and in analyticaltechniques that were developed during the firstthird or so of the century, when economicsemerged as a modern discipline. The range ofthe discipline did not expand greatly duringthe middle decades of the century, due partlyto a concentration on the reformulation of theprevious analytical concepts and techniques inmore sophisticated and more general mathe-matical models. Although the dividends fromthese efforts were high and have contributedto the flexibility and capacity of economics,they did not encourage a reconsideration ofsome of the received assumptions about thescope and focus of economic analysis. Therehas been a significant broadening of the scopeof economics during recent decades, with theemergence of such fields as the new householdeconomics, the new institutional economics,the economics of aging, and medical econom-ics, but much remains to be done.

The balance of this address is divided intofour sections. I begin with the inadequateattention to the accelerating rate of techno-logical change, the implications of this accel-eration for the restructuring of the economy,and its transforming effect on human beings.I then consider the neglect of the nonmarketsector of the economy, the implication of thatneglect for the measurement of consumption,and for the analysis of economic growth. Thethird section deals with the need to shift thefocus of economic analysis from cross-sectional to life-cycle and intergenerationaldata sets, especially in connection with fore-casting. The final section points to the impactof cultural lag in the treatment of material in-equality, and the neglect of the more severeproblem of spiritual inequality. I use the wordspiritual not in its religious sense but as a ref-erence to commodities that lack material form.Spiritual or immaterial commodities make upmost of consumption in the United States andother rich countries today.

2 THE AMERICAN ECONOMIC REVIEW MARCH 1999


FIGURE 1. THE GROWTH OF THE WORLD POPULATION AND SOME MAJOR EVENTS IN THE HISTORY OF TECHNOLOGY

Notes: There is usually a lag between the invention of a process or a machine and its general application to production.‘‘Beginning’’ means the earliest stage of this diffusion process.Sources: Carl W. Bishop, 1936; T. K. Derry and T. I. William, 1960; Graham Clark, 1961; B. H. Slicher von Bath,1963; Stuart Piggott, 1965; Glenn T. Trewartha, 1969; William McNeill, 1971; Jacob Bronowski, 1973; Carlo M.Cipolla, 1974; B. M. Fagan, 1977. See also E. A. Wrigley, 1987; Robert C. Allen, 1992, 1994.

I. Some Implications of Technophysio Evolution

The rapid acceleration in technologicalchange is apparent in the decline in mortalityrates during the twentieth century. Study of thecauses of this decline point to the existence ofa synergism between technological and phys-iological improvements that has produced aform of human evolution that is biological butnot genetic, rapid, culturally transmitted, andnot necessarily stable. This process, which isstill ongoing in both rich and developingcountries, has been called ‘‘technophysioevolution.’’

Unlike the genetic theory of evolutionthrough natural selection, which applies to thewhole history of life on earth, technophysio

evolution applies only to the last 300 years ofhuman history, and particularly to the last cen-tury. Despite its limited scope, technophysioevolution appears to be relevant to forecastinglikely trends over the next century or so in lon-gevity, the age of onset of chronic diseases,body size, and the efficiency and durability ofvital organ systems. It also has a bearing onsuch pressing issues of public policy as thegrowth in population, in pension costs, and inhealth-care costs (Fogel and Dora L. Costa,1997).

Technophysio evolution implies that humanbeings now have so great a degree of controlover their environment that they are set apartnot only from all other species, but also fromall previous generations of Homo sapiens.

3VOL. 89 NO. 1 FOGEL: CATCHING UP WITH THE ECONOMY


TABLE 1—A COMPARISON OF ENERGY AVAILABLE FOR

WORK DAILY PER CONSUMING UNIT IN FRANCE,ENGLAND AND WALES, AND THE UNITED STATES,

1700–1994 (IN KILOCALORIES)

Year FranceEngland and

WalesUnitedStates

(1) (2) (3)

1700a 720 2,3131705 4391750 8121785 6001800 8581840 1,8101850 1,0141870 1,6711880 2,7091944 2,2821975 2,1361980 1,7931994 2,620

Source: Fogel and Floud (1999).a Prerevolutionary Virginia.

This new degree of control has enabled Homosapiens to increase its average body size byover 50 percent, to increase its average lon-gevity by more than 100 percent, and to im-prove greatly the robustness and capacity ofvital organ systems.1

Figure 1 helps to point out how dramatic thechange in the control of environment after1700 has been, and it highlights the astoundingacceleration of technological change over thepast two centuries. The advances in the tech-nology of food production after the SecondAgricultural Revolution (which began about1700 A.D.) were far more dramatic than thoseassociated with the First Agricultural Revolu-tion since they permitted population to in-crease at so high a rate that the line ofpopulation appears to explode, rising almostvertically. The new technological break-throughs in manufacturing, transportation,trade, communication, energy production,leisure-time services, and medical serviceswere in many respects even more striking thanthose in agriculture. Figure 1 emphasizes thatprior to 1600, centuries elapsed between majortechnological advances and the process of dif-fusion was even more extended, continuingover several millennia. Studies of the originand diffusion of the plow, for example, showhow little improvement there was in its designbetween its original development in the Me-sopotamian valley around 4000 B.C. and itsdiffusion across the Mediterranean Sea andnorthward in Europe down to the beginning ofthe second millennium ( Bishop, 1936; E.Cecil Curwen, 1953).

To my mind nothing better illustrates thisamazing acceleration in technological changethan the realization in the twentieth century ofhumankind’s ancient desire to fly. The firstsuccessful motor-driven flight took place in1903, but the fragile aircraft of Wilbur andOrville Wright traveled only a few hundredfeet. Just 66 years later, an astronaut wasstanding on the moon, talking to another as-tronaut on earth, and hundreds of millions of

1 Between circa 1775 and 1995 weight in England in-creased by about 40 percent, while weight in France in-creased by about 60 percent (cf., Fogel and RoderickFloud, 1999).

people around the world overheard andwatched that conversation.

Worldwide, the most important aspect oftechnophysio evolution is the continuing con-quest of chronic malnutrition, which was vir-tually universal three centuries ago.2 Table 1shows that in rich countries today some 1,800to 2,000 or more kilocalories (kcal) of energyare available for work daily per equivalentadult male, aged 20–39.3 At the beginning ofthe eighteenth century, however, France pro-duced less than one-fifth of the current U.S.amount of energy available for work. And En-gland was not much better off. Only the UnitedStates provided potential energy for workequal to or greater than late-twentieth-centurylevels during the eighteenth and early nine-teenth centuries, although some of that energy

2 Even the English peerage, with all its wealth, had adiet during the sixteenth and seventeenth centuries thatwas deleterious to health. Although abundant in caloriesand proteins, aristocratic diets were deficient in some nu-trients and included large quantities of toxic substances,especially alcoholic beverages and salt (Fogel, 1986).

3 Procedure for construction of Table 1 are describedin Fogel and Floud, 1999.



was wasted due to the prevalence of diarrheaand other conditions that undermined thebody’s capacity to utilize nutrients (Fogel andFloud, 1999).

One implication of these estimates of caloricavailability is that mature adults of the eigh-teenth and much of the nineteenth centurymust have been very small by current stan-dards and less physically active. Today the av-erage American male in his early thirties isabout 177 cm (70 inches) tall and weighsabout 78 kg (172 pounds) . Such a male re-quires daily about 1,800 kcal for basal metab-olism and a total of 2,300 kcal for baselinemaintenance. If either the British or the Frenchhad been that large during the eighteenth cen-tury, virtually all of the energy produced bytheir food supplies would have been requiredfor personal maintenance, with little availableto sustain work. To have the energy necessaryto produce the national products of these twocountries circa 1700, the typical adult malemust have been quite short and very light(Fogel, 1997).

Recent studies have established the predic-tive power of height and weight at early ageswith respect to onset of chronic diseases andpremature mortality at middle and late ages.Variations in height and weight appear to beassociated with variations in the chemicalcomposition of the tissues that make up vitalorgans, in the quality of the electrical trans-mission across membranes, and in the func-tioning of the endocrine system and other vitalsystems. Nutritional status thus appears to bea critical link connecting improvements intechnology to improvements in human physi-ology (Fogel and Costa, 1997).

So far I have focused on the contribution oftechnological change to physiological im-provements. The process has been synergistic,however, with improvement in nutrition andphysiology contributing significantly to eco-nomic growth and technological progress in amanner described elsewhere (Fogel, 2000).Here I merely want to point out the main con-clusion. Technophysio evolution appears toaccount for about half of British economicgrowth over the past two centuries. Much ofthis gain was due to the improvement in hu-man thermodynamic efficiency. The rate ofconverting human energy input into work out-

put appears to have increased by about 50 per-cent since 1790 (cf., Partha Dasgupta, 1993).

Technophysio evolution calls into questiona number of easy and frequent assumptions ineconomic analyses such as: tastes are fixed;needs are fixed or exogenously determined;existing life tables are adequate to forecast fu-ture pension costs in 2030 or 2075; and therate of aging is genetically controlled and un-changing from one generation to another.

Technophysio evolution also calls intoquestion such frequently used theoretical as-sumptions as fixed utility functions, fixed ratesof time preference over the life cycle, and therelated assumption that, except for risk differ-entials, economic phenomena should gener-ally be subject to the same rate of discount.Yet the rate of discount varies over the lifecycle; children, after all, have much highertime preferences than adults. It may also varyby religion. Some individuals are prepared towait for their reward in heaven while otherswant it here and now. Moreover, these differ-ences may be evolving and multiplying morerapidly than is now presumed, as is indicatedby the debates over the impact of greenhousegases. Rethinking of the issue is now under-way (see William D. Nordhaus, 1997; cf.,Peter Koslowski, 1992; Nazli Choucri, 1993;Marc Fleurbaey and Philippe Michel, 1994;Michael Toman, 1994) but the profession canbenefit from an increased allocation of re-sources to this problem.

Technophysio evolution requires not justmarginal adjustments, but major leaps in eco-nomic theory. We are slow in pondering suchgrand questions as the implications of the Hu-man Genome Project, which is now nearingcompletion, and the emergence of molecularmedicine for the future of economic life. Wehave entered an era in which purposeful inter-vention in evolutionary processes is passingbeyond plant and animal breeding. The newgrowth economics needs to incorporate at leastsome aspects of directed, rapid human evolu-tion. Endogenous technological change needsto extend to the fundamentals of human be-havior. Theorists also need to grapple with theethical implications of technological changesthat, whatever their positive aspects, threatento undermine the mystery of human life bytransforming people into ‘‘material’’ that is



TABLE 2—SECULAR TRENDS IN TIME USE: THE AVERAGE

HOURLY DIVISION OF THE DAY OF THE AVERAGE MALE

HOUSEHOLD HEAD (BASED ON A 365-DAY YEAR)

c.1880 c.1995 c.2040

Sleep 8 8 8Meals and essential

hygiene 2 2 2Chores 2 2 2Travel to and from

work 1 1 0.5Work 8.5 4.7 3.8Illness 0.7 0.5 0.5Subtotal 22.2 18.2 16.8Residual for leisure

activities 1.8 5.8 7.2

transplanted, cloned, arbitrarily altered in ex-ternal appearance, artificially changed in per-sonality and intelligence, and otherwisemanufactured in ways that challenge the defi-nition of a human being (cf., Zbigniew Brze-zinski, 1996).

II. The Growth of the Nonmarket Sector4

One aspect of technophysio evolution hasbeen a change in the structure of consumptionand in the division of discretionary time be-tween work and leisure. Perhaps the best indexof the growth of the nonmarket sector is thechange in the use of time. Changes in hours ofwork and in the average division of the dayhave paralleled the changes in the structure ofconsumption. Table 2 shows the remarkablereduction in the workyear that has occurred formales in the U.S. labor force over the past cen-tury. Sleep, meals, and essential hygiene,which are biologically determined, requiredabout 10 hours of the day in 1880, as they dotoday. The remaining 14 hours represent ‘‘dis-cretionary’’ time.

The most notable feature of Table 2 is thelarge increase in leisure available to the typicalmale worker. His leisure time has tripled overthe past century, as his workyear declinedfrom about 3,100 hours to about 1,730 hourstoday. Table 2 also forecasts the division ofthe average day in 2040, indicating that by thatdate more than half of the discretionary daywill be devoted to leisure activities. The fore-cast is for a reduction of the workyear fromthe current average of about 1,730 hours to just1,400 hours, with the average workweek downto 30 hours.

The pattern of change among women wassimilar to that among men ( cf., ClaudiaGoldin, 1990; Stanley Lebergott, 1993). Theworkday of women in 1880 was somewhatlonger, and in some respects may have beenmore arduous, than that of men. In householdsof working farmers, artisans, and manual la-borers, wives rose before their husbands andcontinued working until bedtime at 10 or 11P.M. That routine suggests a workday that

4 This section draws on Fogel, 2000.

may have run about 15 minutes longer thanthat of males, implying an annual workyear ofperhaps 3,200 hours.

As a result of the mechanization of thehousehold, smaller families per household,and the marketing of prepared foods, the typ-ical nonemployed married woman todayspends about 3.4 hours per day engaged inhousework; and if she is employed the figurefor housework drops to 2.2 hours. However,women in the labor force average about 4.4hours per day as employees. Hence combining‘‘work’’ with ‘‘chores,’’ men and womenwork roughly equal amounts per day, and bothenjoy much more leisure than they used to.The principal difference is that the gains ofwomen have come exclusively from the re-duction in hours of housework, while the gainsof men have come from the reduction in thehours of employed work ( cf., John P.Robinson and Geoffrey Godbey, 1997).

I have so far retained the common distinc-tion between work and leisure, although theseterms are already inaccurate and may soon beobsolete. The distinction was invented whenmost people were engaged in manual labor for60 or 70 hours per week and was intended tocontrast with the highly regarded activities ofthe English gentry or their American equiva-lent, Thorsten Veblen’s ( 1899 ) ‘‘leisureclass.’’ However, it should not be assumedthat members of the leisure class were indo-lent. In their youth they were students and ath-letes. In young adult years they were warriors.In middle and later ages they were judges,



TABLE 3—ESTIMATED TREND IN THE LIFETIME

DISTRIBUTION OF DISCRETIONARY TIME

c. 1880 c. 1995 c. 2040

Lifetime discretionaryhours 225,900 298,500 321,900

Lifetime earnworkhours 182,100 122,400 75,900

Lifetime volworkhours 43,800 176,100 246,000

bishops, merchant princes, and patrons of thearts. Whatever they did was for the pleasure itgave them since they were so rich that earningmoney was not their concern.

Hence, leisure is not a synonym for indo-lence, but a reference to desirable forms of ef-fort or work. As George Bernard Shaw (1928)put it, ‘‘labor is doing what we must; leisureis doing what we like; and rest is doing nothingwhilst our bodies and our minds are recoveringfrom their fatigue.’’ In order to avoid confu-sion, in the balance of this address I reservethe word ‘‘work’’ for use in its physiologicalsense, an activity that requires energy, overand above the basal metabolic rate (BMR).Activity aimed primarily at earning a living Iwill call ‘‘earnwork.’’ Purely voluntary activ-ity, even if it incidentally carries some pay-ment with it, I will call ‘‘volwork.’’

Why have hours of earnwork declined somuch in recent years? The answer to that ques-tion is suggested by the fact that it is not justdaily and weekly hours of earnwork that havedeclined. The share of lifetime discretionaryhours spent in earnwork has declined evenmore rapidly. Table 2 only dealt with the hoursof earnwork of persons in the labor force. Itdid not reflect the fact that the average age ofentering the labor force is about five years latertoday than it was in 1880, or that the averageperiod of retirement for those who live to age50 is about 11 years longer today than it wasin 1880 (cf., Lee, 1996). A century ago onlyone out of five males aged 65 and older wereretired. Today, six out of seven are retired.

All in all the lifetime discretionary hoursspent earning a living have declined by aboutone-third over the past century (see Table 3)despite the large increase in the total of life-time discretionary time. In 1880, four-fifths of

discretionary time was spent earning a living.Today, the lion’s share (59 percent) is spentdoing what we like. Moreover, it appears prob-able that by 2040 over three-quarters of dis-cretionary time will be spent doing what welike, despite a further substantial increase indiscretionary time due to the continuing exten-sion of the life span (cf., Jesse H. Ausubel andA. Grubler, 1995).

Why this deep desire for volwork? Why doso many people want to forgo earnwork whichwould allow them to buy more food, clothing,housing, and other goods? The answer turnspartly on the extraordinary technologicalchange of the past century, which has not onlygreatly reduced the number of hours of laborthe average individual needs to obtain his orher food supply, but has also made housing,clothing, and a vast array of consumer dura-bles so cheap in real terms that the totality ofmaterial consumption requires much fewerhours of labor today than was required over alifetime for food alone in 1880.

Indeed, we have become so rich that we areapproaching saturation5 in the consumptionnot only of necessities, but of goods recentlythought to be luxuries, or which were onlydreams or science fiction during the first thirdof the twentieth century. Today there is an av-erage of two cars per household in the UnitedStates. Nearly every household with a personcompetent to drive a car has one. On someitems such as radios, we seem to have reachedsupersaturation, since there is now more thanone radio per ear (5.6 per household) . Thepoint is not merely that we are reaching satu-ration in commodities that once defined a highstandard of living and quality of life but alsothat the hours of labor required to obtain thosecommodities have drastically declined. Thetypical household in 1875 required 1,800hours of labor in the marketplace to acquirethe annual food supply, but today it takes just260 hours. All in all, the commodities thatused to account for over 80 percent of house-hold consumption can now be obtained ingreater abundance than previously, with less

5 ‘‘Saturation’’ in this context means that most pur-chases are for replacement rather than for new use.



TABLE 4—THE LONG-TERM TREND IN THE STRUCTURE

OF EXPANDED CONSUMPTION AND THE IMPLIED INCOME

ELASTICITIES OF SEVERAL CONSUMPTION CATEGORIES

Consumptionclass

(Percent)Distribution of

expandedconsumption

c. 1875 c. 1995

Long-termincome

elasticities

Food 49 5 0.2Clothing 12 2 0.3Shelter 14 5 0.7Health care 1 9 1.6Education 1 5 1.6Other 6 7 1.1Leisure 18 67 1.4

than a third of either the market or the house-hold labor once required (Fogel, 2000).

Table 4 shows how sharply the U.S. distri-bution of consumption has changed over thepast 120 years. Food, clothing, and shelter,which accounted for about three-quarters ofconsumption in 1875, accounted for just 12percent in 1995. Leisure, on the other hand,has risen from 18 percent of consumption to67 percent.6 As Table 4 shows, the long-termincome elasticities of the demand for food andclothing are below 0.5 and the elasticity of thedemand for shelter is closer to, but still below,1.0. On the other hand, the long-term incomeelasticities for leisure, education, and for med-ical services are over 1.0.7

Table 4 differs from current official tabula-tions of household consumption in two prin-cipal respects. Official tabulations, with minorexceptions, are limited to the out-of-pocket ex-penditures of households. Table 4, however,adds expenditures for education and healthcare consumed by households but paid by gov-ernment, employers, and other third parties.My procedure does little to change the distri-bution of consumption in 1875 but signifi-cantly increases the education and health-careshares of consumption in 1995.8

Table 4 also differs from official tabulationsby adding to the value of out-of-pocket expen-ditures on leisure, the value of the time de-voted to leisure (volwork). That time is pricedat the average hourly rate of compensation oflabor. The procedure increases the share of lei-sure in consumption in both 1875 and 1995.While leisure (volwork) remains a relativelysmall share of expanded consumption in 1875,

6 ‘‘Other’’ consists mainly of expenditures on utilitiesand transportation, and of purchases of the supplies, fur-nishings, equipment, and services involved in the opera-tion of households.

7 The elasticities are not adjusted for prices. Since theirprices declined, the price-adjusted income elasticities arelower than shown for the first three expenditure categoriesand the last one. Since the prices of the other three expen-diture categories rose, their price-adjusted income elasti-cities are higher than shown.

8 I have also excluded savings and taxes. Nordhaus(1997) points out that when account is taken of the in-crease in health and other forms of human capital, thesavings rate has not declined.

it accounts for two-thirds of expanded con-sumption in 1995.9

The failure to include the value of volworkin the national income and product accounts isperhaps the most glaring example of the cul-tural lag to which I referred at the beginningof this address. That omission also leads to asignificant underestimate of the long-termgrowth rate of per capita income. Before ad-justments for the increased quality of volwork,the growth rate is increased by eight-tenths ofa point (from 1.8 to 2.6 percent per annum).But adjustments for improvements in the qual-ity of volwork might substantially increasethat figure, since leisure-time activities in 1875were limited largely to church on Sundays andcarousing in bars during the rest of the week(cf., Fogel, 2000). Despite some importantcontributions to the development of an eco-nomics of leisure (cf., Gary S. Becker, 1991;Costa, 1998a, c; Daniel S. Hamermesh, 1998;John Pencavel, 1998), the profession has farto go before it catches up with the economy.

III. Shifting to Life-Cycle Data Sets forSuccessive Cohorts and to Intergenerational

Data Sets

It is no secret that cross-sectional data setsare cheaper to construct and more abundant

9 This procedure was previously followed by SimonKuznets (1952) and others who corrected the national in-come and product accounts for the omission of volwork.See Robert Eisner (1988).



than longitudinal data sets. But whatever theirusefulness for other problems, cross-sectionaldata are often highly misleading guides totrends on such critical economic issues of thenew millennium as the prevalence rates ofchronic disabilities, expenditures on healthcare, and pension costs.

Research initiated during the past decadeand a half has called into question previousviews on the length and fixity of the life span,on the shape of the Gompertz curve (whichrelates the log of the age-specific probabilityof dying to age), on the theory of the epide-miological transition, 10 and on the relatedproposition that longer life expectancy impliesworse health among the survivors. It appearsthat some of the earlier propositions were theconsequence of attempts to infer life-cycle be-havior from cross-sectional data sets. Such ef-forts were thwarted by changes in thesampling design of successive cross sectionsand by changes in technology that led to earlierdiagnosis of preexisting conditions (TimothyWaidmann et al., 1995).

The new research also accumulated evi-dence on the outward movement of the sur-vivorship curve. Vaino Kannisto (1994) (cf.,Kannisto, 1996) has shown that, in 14 coun-tries for which the data are adequate, mortalityover age 80 has been declining by about 1 per-cent per annum for about half a century witha significant acceleration in recent decades.John R. Wilmoth (1995, 1997), who exam-ined extreme longevity in five countries,concluded that the right-hand end of the sur-vivorship curve has been shifting outward fortwo centuries. Moreover, there is additionalevidence that the Gompertz curve either levelsoff or declines at old old ages (James W.Vaupel, 1997; cf., S. Jay Olshansky and BruceA. Carnes, 1997).

Another development is the continued ac-cumulation of evidence linking events early inlife, and reflected in height, weight, and bodymass index (BMI), to the onset of chronic con-

10 This theory holds that the prevalence of chronic dis-eases is unrelated to the prevalence of acute infectiousdiseases but depends only on the natural process of senes-cence that precedes mortality.

ditions.11 A number of longitudinal studies thatwere launched in the 1950’s and 1960’s haverecently been extended to cover the entire pe-riod of human growth and early adulthood,through follow-up studies. These have con-firmed the persistence of central nervous sys-tem defects induced by malnutrition in earlychildhood (Nevin S. Scrimshaw, 1995). Nu-merous other follow-up studies have shownthat malnutrition and smoking in adolescenceand in middle ages are risk factors for the earlyonset of chronic conditions and prematuremortality, especially due to coronary heart dis-ease, non-insulin-dependent diabetes, and re-spiratory diseases (Avita Must et al., 1992;Vincent J. Carey et al., 1997; N. K. Chin etal., 1997; Joan M. Dorn et al., 1997; K. Kotaniet al., 1997; Dan S. Sharp et al., 1997; RalfBender et al., 1998). These relationships havebeen established in American, Asian, Austra-lian, European, and Latin American popula-tions—rich and poor. Economists have alsodiscovered a close link between later produc-tivity and height, BMI, and protein consump-tion ( after controlling for caloric intake )( John Strauss and Duncan Thomas, 1995;Thomas and Strauss, 1997).

There has also been an expansion of re-search into the connection between intrauter-ine and infant growth and the onset of chronicdiseases (or premature mortality) . The strong-est evidence for such a link that has emergedthus far is with respect to hypertension, coro-nary heart disease (CHD), and non-insulin-dependent diabetes. A review of 32 papersdealing with the relationship between birthweight and hypertension by Catherine M. Lawand Alistair W. Shiell (1996) showed a ten-dency for middle-aged blood pressure to in-crease as birth weight declined. Evidence of aconnection between birth size and later coro-nary heart disease has been found in England,Wales, Sweden, India, and Finland.12

11 BMI is measured by the ratio of weight in kilogramsto height in square meters (BMI Å kg/m2).

12 See Stephen Frankel et al., 1996; Law and Shiell,1996; C. E. Stein et al., 1996; Swen-Olef Andersson et al.,1997; D. J. Barker, 1997; Barker and C. N. Martyn, 1997;J. L. Cresswell et al., 1997; T. Forsen et al., 1997; Jan A.Henry et al., 1997; Ilona Koupilova et al., 1997; Scrim-



The accumulation of historical and currentbiomedical studies on the trends in health, lon-gevity, and human physiology, combined withcontrolled studies of animal populations, areleading some evolutionary biologists to placeincreasing emphasis on plasticity in human ag-ing (Michael R. Rose, 1991; Caleb E. Finch,1997; Hillard Kaplan, 1997). The term ‘‘plas-ticity’’ refers to the widely varying pattern ofaging within species and the ability of thelength of the life span to be affected by envi-ronmental factors, some of which may operateover several generations ( cf., Kenneth W.Wachter and Finch, 1997).

The search for better ways to forecast long-term trends in health-care costs and in pensioncosts has made it essential for economists tounderstand the underlying physiological, nu-tritional, and epidemiological factors that af-fect the changing demand for health care andretirement. Forecasts about health-care costsare obviously related to assumptions aboutchanges in both life expectancy and age-specific morbidity rates. If current morbidityrates at older ages are presumed to remain con-stant or to increase over the next 30 to 50years, and if life expectancy is also presumedto increase, then national health-care costs be-come astronomic. However, if one takes ac-count of the recent declines in age-specificdisability rates, then the share of health-carecosts in GDP might remain constant (KennethG. Manton et al., 1997b; Manton et al., 1998),provided the demand for health services

shaw, 1997; Stein et al., 1997, A. J. C. Ravelli et al., 1998.However, a number of researchers who attempted to rep-licate the Barker results have been unable to do so (DennyVagero and David Leon, 1994; Kaare Christensen et al.,1995; Kannisto et al., 1997). It has also been pointed outthat some of the key predictors of a link between fetal andinfant development and later chronic disease are oftenfound to be uncorrelated with CHD and other chronic dis-eases (Ivan J. Perry et al., 1995; Michael S. Kramer andK. S. Joseph, 1996). Nevertheless, the volume of studiesconfirming the life-cycle connection has increased sub-stantially since 1994. What is agreed on both sides of thedebate is the need for studies that take better account ofenvironmental factors in utero and between birth and theonset of chronic conditions, including middle-aged vari-ations in nutrition and lifestyle from one population toanother, and other confounding factors, such as mother’slifestyle (including smoking) (Nigel Paneth and MervynSusser, 1995; Perry et al., 1995; Scrimshaw, 1997).

(given disability rates) does not increase. If,however, the income elasticity of the demandfor health is greater than one, then as incomeincreases, the share of income that is spent onhealth care could increase even if disabilityrates decline (see Fogel, 1997). It is quite pos-sible for the demand for medical interventionsto increase, even though age-specific morbid-ity rates decline, for one or more of the fol-lowing reasons: shifts in the age structure ofthe population toward ages with high morbid-ity rates; improvements in medical interven-tions for a wide range of chronic conditionswhich alleviate symptoms but do not cure; thereplacement of current disease-specificstandardsof care by new ones based on improved tech-nologies that are much more expensive thanthose replaced; the elimination of current age re-strictions on such expensive procedures as organreplacements (cf., David M. Cutler and EllenMeara, 1998; Alan M. Garber et al., 1998).

The complexity of the underlying issues inthe economics and biodemography of aginghas led to a significant shift away from cross-sectional data sets and to a concentration onlongitudinal data sets. The earliest of the majorlongitudinal studies sponsored by the NationalInstitutes of Health (NIH), the FraminghamHeart Study, began in 1950 and was focusedon the causes and treatment of coronary heartdisease rather than on the economics and bio-demography of aging. The Retirement HistorySurvey (RHS) followed a cohort of men andwomen aged 58 to 63 from 1969 to 1979 andthen was suspended. Although not initiallyplanned as a longitudinal study, it became themainstay of retirement research during the1980’s. Linked to the social security file oncovered earnings, it provided important in-sights into the labor-force and retirement be-havior of older workers.

In 1990, the National Institute of Aginglaunched a new Health and Retirement Survey(HRS) focused on the cohort born between1931 and 1941, and aimed at collecting a muchwider set of variables than covered by theolder RHS. The new HRS collected numerousmeasures of health (including cognitive abil-ity ) , taxable earnings, job characteristics,hours of work, job turnover, family character-istics, pensions, health insurance, and owner-ship of an array of assets including real estate,



consumer durables, and financial securities.The most ambitious social science project everundertaken, the first wave of HRS cost$14,000,000 and the second wave was budg-eted for $17,000,000. A third wave is currentlyunder way collecting information on the birthcohorts of 1942–1947 (F. Thomas Juster andRichard Suzman, 1995).

NIA launched a second longitudinal surveyaimed at cohorts born in 1924 or earlier years.This sample, which is called Asset and HealthDynamics. Among the Oldest Old (AHEAD),consists of persons aged 70 years and older in1994, when the first wave was undertaken. Al-though the questions in AHEAD overlap withHRS, functional ability is investigated inten-sively and labor-market activity is coveredlightly.

Both HRS and AHEAD have made it pos-sible to examine a wide range of issues withmore reliable and more detailed evidence thanpreviously available. James P. Smith andRaynard Kington (1997), for example, haveused AHEAD to examine disparities in func-tional status and to relate them to socio-economic status ( SES variables ) . Theydiscovered strong feedback effects not onlyfrom health to SES variables but from SESvariables to health. To disentangle these ef-fects they broke household income and wealthinto various categories made possible by thedata and examined the effects by age. Theywere also able to make use of a range of vari-ables bearing on the health and socioeconomicstatus of relatives in three generations (par-ents, siblings, and children). Their analysis in-dicated that current-period health and incomeare attributes of past, concurrent, and futuregenerations. Measuring the full extent of theseinfluences, the direction of causation, andcomplex interactions requires the study of life-cycle histories from birth to very old age. Itwould be inappropriate, they concluded, to useSES variables to explain variations in late-agehealth without taking account of the feedbackmechanisms or identifying within-period in-novations in the stock of health (cf., JohnBound et al., 1998).

Such findings point to the usefulness of cre-ating a prospective life-cycle sample for an ex-tinct cohort (the veterans of the Union Army)by utilizing military, pension, and census records

in archives. This procedure not only creates alongitudinal data base in a small fraction of thetime required to trace a living cohort, but can bedone at less than a tenth of the normal cost. Sucha project was launched by NIA in 1996 (cf.,Clayne L. Pope and Larry T. Wimmer, 1998).

Based on 11 different data sets, the fullylinked life histories contain over 10,000 vari-ables on each recruit, including socioeconomic,ecological, and health variables. Called ‘‘EarlyIndicators of Later Work Levels, Disease, andDeath,’’ this project provides a comprehensivedata set on the life course of over 39,000 UnionArmy veterans. Born mainly between 1835 and1845, these men represented the first cohort toreach age 65 in the twentieth century, and canbe compared with the veterans of World War II.

The preliminary comparisons revealed that atthe same ages the prevalence of chronic diseaseswas much higher among elderly Union Armyveterans than among veterans of World War II.Musculoskeletal and respiratory diseases were1.6 times as prevalent, heart diseases were 2.9times as prevalent, and digestive diseases were4.7 times as prevalent among elderly veterans in1910 than in the mid-1980’s ( cf., Sven E.Wilson and Louis L. Nguyen, 1998). Moreover,young adults born during the second quarter ofthe nineteenth century who survived the deadlycontagious diseases of childhood and early ad-olescence were not freer of degenerative dis-eases than persons of the same ages today, aspropounded by the theory of epidemiologicaltransition, but were more afflicted. Hernia ratesat ages 35–39, for example, were more thanthree times as prevalent in the 1860’s as in the1980’s.

Although the analysis of the data in the Un-ion Army sample is still in progress, some ofthe initial findings have a bearing on forecastsof long-term trends in health status. Costa(1998b) has reported that early-age socioeco-nomic and biomedical stress had a substantialimpact on the likelihood that Union Army vet-erans would have disabling chronic healthconditions by age 60. Thus, veterans raised ina county with high mortality rates were, half acentury later, at elevated risks of sufferingfrom disabling respiratory disease, circulatorydisease, and musculoskeletal problems (cf.,Lee, 1997). Episodes of acute diseases expe-rienced as young adults, such as respiratory



infections, work-related injuries, and extendedbouts of diarrhea, also increased the odds ofsuffering from chronic disabilities by age 60.Costa concludes that about 15 percent of thedecline in the prevalence of joint problems and75 percent of the prevalence of back problemsbetween 1910 and 1980 was due to shifts inthe occupational structure from manual towhite-collar jobs. Moreover, a comparison ofrates of decline in disabilities before and after1980 indicates that disabilities are declining atan accelerating rate (cf., Manton et al., 1997a;Cutler and Elizabeth Richardson, 1998).

The mounting evidence of substantial inter-actions over the life cycle that influence theprocess of aging, the acceleration of techno-logical change which has profoundly affectedthe context in which aging occurs, and the in-creasing evidence that environmental influ-ences on the aging process begin in utero, hasled to the initiation of a new life-cycle projectcalled ‘‘Fetal, Infant, and Later Aging Mark-ers, Cohort b. 1910–35.’’ The acronym forthis project is FILAM.

The central objective of FILAM is the cre-ation of a life-cycle sample of persons bornbetween 1910 and 1935 that would make itpossible to compare changes in the aging pro-cess over the period of 70 years that separatethis new sample from the aging experiences ofthe Union Army cohort. The FILAM cohort isimportant not only because it studies individ-uals currently between ages 63 and 89, but alsobecause of the dramatic environmental andearly life-style changes they experienced.These changes include the rise and partial de-cline of smoking, the decline and the new riseof alcohol consumption, the replacement ofhorses by internal combustion engines as themain source of urban vehicular power, thecleaning up of the water and milk supplies, andthe emergence of a wide range of effectivemedical interventions.

The sample will be drawn from the birthrecords of 10,000 men and women of differingethnicities and races who were born in Boston,New York City, Baltimore, Chicago, IowaCity, and San Francisco. Approximately halfof the neonates will not have survived to thepresent day and these men and women will belinked to their death certificates. The survivorswill be traced and interviewed to determine the

presence of chronic conditions, socioeco-nomic status, and family and own health his-tory. Survivors will be linked to socialsecurity, census, military, and tax records, sub-ject to their written consent.

FILAM will make it possible to investigatethe predictive power of fetal, neonatal, andearly childhood measures of retarded growthsuch as weight for gestational age, ratio of pla-cental weight to birth weight, infant weightgain, thinness at birth, and shortness at birthrelative to head size, on the risk of developingspecific chronic conditions at mid-adult andlate ages. Among the chronic conditions thatwill be examined are coronary heart disease,hypertension, stroke, obstructive lung disease,non-insulin-dependent diabetes, and autoim-mune thyroiditis. FILAM will also make itpossible to investigate how the interaction offetal and infant developments with various riskfactors at later childhood, young adult, andmiddle ages may intensify or moderate risksof chronic conditions and early death after age65. Another objective is the analysis of differ-ences in the health histories, occupational his-tories, and retirement pattern between FILAMand the Union Army sample, with special em-phasis on the similarities and differences ofpredictors of later work levels, morbidity, andwaiting time until death.

The HRS, AHEAD, Union Army sample,and FILAM all contain information on theparents and children of the individuals understudy. Hence, they make it possible to ad-dress some factors that may be transmittedintergenerationally. An effort to utilize ar-chival data to study intergenerational pro-cesses is also under way. Pope (1992) islinking a large sample of genealogies to thelife-cycle sample of Union Army recruits.There are at least 60,000 published familyhistories that contain information on over100,000,000 people who ever lived in NorthAmerica. Pope is drawing a subsample ofthese histories containing 10,000 men ofmilitary age at the time of the Civil War. Itis estimated that 40 percent of these menserved in the Union Army. This new sample(called ILAS, for intergenerationally linkedaging sample ) will be linked to the military,pension, medical, and census data sets pre-viously discussed.



ILAS will make it possible to control for theeffect of wartime stress by comparing subse-quent morbidity and mortality among those whoserved in the Union Army with relatives whodid not. It will also be possible to measure familyeffects on aging and mortality experience. Thiswill be done by including parents’ occupation,wealth, residential history, number of children,place or region of birth, and migration history.Brothers in ILAS share a common prewar en-vironment as well as a common genetic heritage.They may also share a common war experience(see D. S. Lauderdale and P. J. Rathouz, 1998).This common genetic heritage and environmentis not fully captured by the aforementioned in-tergenerational variables because heritability iscomposed of the many different dimensions in-cluded in genetics and environment. However,common family effects may be measured by us-ing independent variables to ‘‘sweep out’’ theeffects of observed variables on death age. Thecovariance of the errors in that regression withbrothers’ estimated death ages is then a measureof the common family effect on death age. Asan alternative to the residual-covariance ap-proach, a kindred-frailty model (Vaupel, 1990)could be used where brothers share a level offrailty. Then, with assumptions about the struc-ture of the frailty distribution, the parameters ofa hazard function and a frailty distribution couldbe estimated ( cf., James J. Heckman andChristopher R. Taber, 1994).

IV. The Production and Distributionof Spiritual Assets

There is, finally, the issue of spiritual or im-material assets. A good place to begin is withSocrates’ question: What is the good life? Thatwas a critical question not only for the sons ofrich Athenians but for sons of the landed richthroughout history. Freed of the need to workin order to satisfy their material needs, theysought self-realization in public service, mili-tary adventures, philanthropy, the arts, theol-ogy, ethics, and moral philosophy. Theirpreoccupation with immaterial commoditiesled Adam Smith to argue that the landed ar-istocracy ignored their property and lacked in-terest in advancing methods of cultivation.‘‘The situation of such a person,’’ he wrote,‘‘naturally disposes him to attend rather to or-

nament which please his fancy, than to profitfor which he has so little occasion’’ (Smith,1937 pp. 364, 891–892).

In a world in which all but a small percent-age are lacking in adequate nutrition and othernecessities of life, self-realization may indeedseem like a mere ornament, but not in a coun-try where even the poor are rich by past orThird World standards. That is the case inAmerica today since the poverty line is at alevel of real income that was attained by onlythose in the top 10 percent of the income dis-tribution a century ago13 (Fogel, 2000). Tech-nophysio evolution has made it possible toextend the quest for self-realization from aminute fraction of the population to almost thewhole of it. Although those who are retiredwill have more time to pursue self-realization,even those still in the labor force will havesufficient leisure to seek it either within theirprofessional occupations or outside of them(Peter Laslett, 1991; Hans Lenk, 1994).

Some proponents of egalitarianism insiston characterizing the material level of thepoor today as being harsh. They confoundcurrent and past conditions of living. Failureto recognize the enormous material gainsover the last century, even for the poor, im-pedes rather than advances the struggleagainst chronic poverty in rich nations, theprincipal characteristic of which is spiritualestrangement from the mainstream society.Although material assistance is an importantelement in the struggle to overcome spiritualestrangement, such assistance will not beproperly targeted if one assumes that im-provement in material conditions naturallyleads to spiritual improvement.

That proposition, so widely embraced by themore secular of the economic reformers of thetwentieth century, did more to promote theconsumerism of the 1920’s and 1930’s than toproduce spiritual regeneration. The middle andworking classes became preoccupied with theacquisition of automobiles and those house-hold appliances made possible by electricity:

13 This calculation neglects a variety of goods and ser-vices, such as heart bypass operations, that are availabletoday through Medicaid, but which were unavailable atany income a century ago.



irons, lamps, telephones, toasters, refrigera-tors, radios, and washing machines. It was thisconsumerism that led such progressive criticsof the era as Vernon Louis Parrington, pioneerin the development of intellectual history, todecry the ‘‘cash-register’’ mentality of modernurban life (Parrington, 1930 p. 81).

The economist’s traditional measures of in-come inequality are inadequate measures ofboth egalitarian gains and egalitarian failures(cf., Amartya K. Sen, 1996). They focus on avariable—money income—that currently ac-counts for less than half of real consumptionand which in a generation may slip to just aquarter of real consumption. The most seriousthreats to egalitarian progress—certainly themost intractable forms of poverty—are relatedto the unequal distribution of spiritual ( im-material ) resources ( cf., William JuliusWilson, 1996).

Realization of the potential of an individualis not something that can be legislated by thestate, nor can it be provided to the weak by thestrong. It is something that has to developwithin each individual. Moreover, which as-pect of one’s potential an individual choosesto develop most fully, such as choosing a pro-fession, is purely an aesthetic consideration.John Dewey and one of his chief disciples,Richard Rorty of the University of Virginia,contend that in a democracy self-realization is‘‘a particularized creative project of individualgrowth’’ (Richard Shusterman, 1994 pp. 396–97). The emphasis on individual choice doesnot mean that other individuals and institutionsplay no role in shaping those choices. Quitethe contrary, the quality of the choices and therange of opportunity depends critically on howwell endowed an individual is with spiritualresources. But the spiritual resources neededfor personal development are unequally dis-tributed among young and old, among menand women, among various ethnic groups, andamong rich and poor. Those rich who are con-tinuously preoccupied with sensual gratifica-tions are as likely to fail in self-realization asthe poor who share that preoccupation. Al-though the rich may have the wealth to buythe treatment needed from highly trained pro-fessionals, inexpensive character and religiouscounseling may serve as well, if not better, forthe addicts of all classes.

The full list of maldistributed spiritual re-sources is too long to discuss adequately herebut I have in mind such vital assets as a visionof opportunity and a work ethic. A commoncharacteristic of such assets is that they aretransferred from one individual to anothermainly very early in the life of the recipient.Self-esteem and a sense of family solidaritybegin to be transferred to children along withmother’s milk and with pabulum. Other spir-itual resources begin to be transferred duringthe toddler and toilet-training stages, includinga sense of discipline, a capacity to resist orcontrol impulses, and a sense of community.Telling nursery rhymes such as ‘‘This littlepiggy went to market,’’ recounting the auto-biographies of the mother and father, and fam-ily histories going back two or threegenerations convey such spiritual resources asa work ethic, a sense of the mainstream ofwork and life, an ethic of benevolence, a visionof opportunity, and a thirst for knowledge.

Although these early transfers of spiritualresources are enriched and expanded by pri-mary, secondary, and college education, andby occupational and other later-life experi-ences, the salience of these later transfers de-pends in no small measure on what happens athome before formal education begins. It is,therefore, necessary to remedy the maldistri-bution of spiritual resources early in life, be-cause the most spiritually deprived infants willoften be born to single, teenaged mothers whoare themselves spiritually deprived.

Some young mothers are too deprived, ortoo young, to call on their own life experiencesto transmit a sense of discipline and of oppor-tunity, a work ethic, a family ethic, a sense ofself-esteem, and a knowledge of the main-stream of work and life. The deprivation canbe addressed by promoting a system of men-toring, taking advantage of the increasinglylarge number of retired men and women whohave abundant spiritual resources. Such men-toring programs would be useful, not only forthe toddlers and their mothers and fathers, butalso for the elderly who are looking for waysto enrich their retirement years.

Despite the improvements in their materialconditions of life, including comfortablestocks of consumer durables, the elderly todaysuffer from a maldistribution of immaterial



resources that traces back to the conditions oftheir youth. Persons aged 80 today were bornin 1918 or 1919. Only 43 percent of that cohortgraduated from high school and less than 15percent entered college. Even among theyoungest cohort of the elderly, those born in1933 and 1934, only half graduated from highschool and about 20 percent entered college(U.S. Bureau of the Census, 1975 p. 379).These cohorts also suffered from high infantdeath rates, poor nutrition in early infancy, andearly onset of chronic diseases, as comparedwith cohorts born since World War II.

Hence, despite their relatively high levels ofincome and stocks of consumer durables, themaldistribution of spiritual resources is sub-stantial. Depression, alienation, and substanceabuse are common (S. C. Samuels, 1997).Those who are most afflicted are lonely, havefew communal contacts, live in retirementhomes rather than in their own households,and sense a loss of control over their personallives ( W. L. Fletcher and R. O. Hansson,1991; K. Pahkala et al., 1992; K. Yamashitaet al., 1993). Recent studies also indicate thatthose who lacked immaterial resources earlyin life have difficulty in attaining self-realization after retirement (J. C. Henretta,1997; J. E. Mutchler et al., 1997).

I have emphasized the level of education be-cause recent studies indicate that the capacityof the elderly to engage effectively in physicalactivity was strongly correlated with educationearly in life. Education also affects cognitiveability and the rate of illness (J. W. Rowe andR. L. Kahn, 1997). Consequently, individualswho were deprived of adequate education inyouth are, for that reason among others, rela-tively deprived of both physiological and spir-itual resources in late life.

Despite the long reach of youthful depri-vation, there are enough other factors affectingthe quality of elderly life to permit redistri-butions that compensate for previous deficits.On the physiological side, for example, thereare effective medical interventions that can in-crease the quality of life and longevity. Al-though the elderly are eligible for Medicare topay for treatment, the quality of treatment isvariable, and many individuals may beshunted to low-quality care. Moreover, someinterventions are denied, or are more reluc-

tantly ordered, for the elderly than for the mid-dle aged.

Because spiritual resources are so un-equally distributed among the elderly, dif-ferent programs are needed for differentstrata. The minority of the current elderlywho are well educated, the 14 percent withat least bachelor’s degrees, most of whomhad professional careers, have developedsome innovative programs (U.S. Bureau ofthe Census, 1997 p. 160) . In Great Britainone of these is called the ‘‘University of theThird Age.’’ This educational program is notaimed at providing credentials for thoseabout to embark upon new careers, but atsatisfying the thirst for knowledge. It isbased on the proposition that education, andthe acquired knowledge and skills, are asource of self-satisfaction, even if they donot enhance an individual’s employability.As Laslett (1991 pp. 171–71) put it:

Reading in a literature, mastering a lan-guage, unraveling a point in logic or phi-losophy, understanding the objectivesset for themselves by poets, painters,novelists or architects, these things ex-tend your appreciation and your masteryof your world, your objective and yoursubjective world as well. They are ful-filling, and adding to other people’sknowledge is the most fulfilling of all.

Programs such as the University of theThird Age will become increasingly importantas the baby boomers and others of the morehighly educated cohorts of the post-WorldWar II era begin to retire. However, today andfor the next decade, the bulk of the elderlylacks the skills to create and participate in suchhigh-level programs as the University of theThird Age. A recent survey of adult literacyrevealed that more than half of the elderly pop-ulation suffers from functional illiteracy.These individuals may be able to sign theirname or read very simple material, but theycannot follow instructions for taking medi-cines or cope with a variety of documents en-countered in daily living (R. Boling, 1998).

Those who suffer from low levels of literacyare educable. Engaging them in intellectual ac-tivities has a significant influence on theirphysiological performance. Recent studies re-



veal more physiological plasticity than waspreviously suspected. The capacity for self-improvement continues into old age and ap-propriately designed programs can returndiminished individuals to earlier levels offunctioning ( Rowe and Kahn, 1997; cf.,Wachter and Finch, 1997).

Peer tutoring has a two-way effect, sinceit is beneficial both to the learner and to thetutor. Both gain from involvement in socialnetworks that enhance mood, combat de-pression, and reduce the risk of suicide. Forwidowed men, the benefits are physiologicalas well as psychological. Men in situationsthat provide higher social support have sig-nificantly lower losses of cortisol, epineph-rine, and norepinephrine ( hormones thatreduce pain, stimulate the functioning of theheart, and improve electrical transmissionacross cells ) . Statistical analysis indicates apositive relationship for both men andwomen between social support and physicalperformance. For the tutors, being active insuch productive and emotionally rewardingactivities serves to retain a sense of relativeyouthfulness. Thus volwork, because it is ef-fective, because it is emotionally rewarding,and because it is what the tutors want to do,adds significantly to national product.

Use of fiscal policy to correct the maldis-tribution of income is based, explicitly orimplicitly, on the ethical proposition thatthose households at the top of the incomedistribution have more income than theyought to have. What about the case of spir-itual redistributions? Are spiritual resourcesmaldistributed because virtue is too heavilyconcentrated? Government cannot legislatethe transfer of virtue as it does with moneyincome. Even if they desired to do so, thoserich in virtue or in the family ethic, or inbenevolence, could not transfer spiritual re-sources by writing out checks denominatedin virtue, benevolence, or family solidarity.Those poor in these spiritual resources ac-quire more of them only through the processof self-realization, through a concerted ef-fort to develop as fully as possible the vir-tuous aspects of their nature.

Those rich in spiritual resources can helpthose who are spiritually deprived by counsel-ing them, by providing spiritual companion-

ship and moral support, by informing andteaching those who are deprived about existingopportunities and procedures, and by helpingto raise their self-esteem. But this process ofcorrecting the maldistribution of spiritual re-sources not only leaves those who are deprivedbetter off, it also increases the spiritual re-sources of those who have virtue in abun-dance. In contrast to income redistribution,spiritual redistribution is not a fixed-sum gamein which some people can become better offonly if other people are made worse off. It isa game in which total resources increase andthe share of the deprived in this larger totalmay also increase without in any way dimin-ishing those who have a superabundance ofspiritual resources.

Some economists may be astonished by myclaim that in some respects the discipline hasfallen seriously behind the economy. After all,if it were so, who would know about it beforethem? The answer lies in the subtext of thisaddress. To understand where the economy isand how it is evolving one needs to study notonly the present but the past. In the 1940’s,Kuznets (1941), musing about some of the an-alytical mistakes made in the aftermath of theGreat Depression said: ‘‘A broader historicalbackground might have prevented some econ-omists from ignoring the dependence of theirgeneralizations upon transient historical con-ditions.’’ That advice is as good today as itwas a half century ago.

REFERENCES

Allen, Robert C. Enclosure and the yeoman:The agricultural development of the SouthMidlands . Oxford: Oxford University Press,1992.

. ‘‘Agriculture During the IndustrialRevolution,’’ in Roderick Floud and DonaldMcCloskey, eds., The economic history ofBritain since 1700 , Vol. 1, 2nd Ed. Cam-bridge: Cambridge University Press, 1994,pp. 96–122.

Andersson, Swen-Olof; Wolk, Alicja; Bergstrom,Reinhold; Adami, Hans-Olov; Engholm,Goran; Englund, Anders and Nyren, Olof.‘‘Body Size and Prostate Cancer: A 20-YearFollow-Up Study Among 135,006 SwedishConstruction Workers.’’ Journal of the



National Cancer Institute , March 5, 1997,89(5) , pp. 385–89.

Ausubel, Jesse H. and Grubler, A. ‘‘Working Lessand Living Longer: Long-Term Trends inWorking Time and Time Budgets.’’ Techno-logical Forecasting and Social Change, No-vember 1995, 50(3), pp. 113–31.

Barker, D. J. ‘‘The Fetal Origins of CoronaryHeart Disease.’’ Acta Paediatrica , July1997, Supp. 422, pp. 78–82.

Barker, D. J. and Martyn, C. N. ‘‘The Fetal Or-igins of Hypertension.’’ Advances in neph-rology from the Necker Hospital , 1997, 26 ,pp. 65–72.

Becker, Gary S. A treatise on the family . Cam-bridge, MA: Harvard University Press,1991.

Bender, Ralf; Trautner, Christoph; Spraul, Max-imilian and Berger, Michael. ‘‘Assessment ofExcess Mortality in Obesity.’’ AmericanJournal of Epidemiology , January 1, 1998,147(1) , pp. 42–48.

Bishop, Carl W. ‘‘Origin and Early Diffusionof the Traction Plow.’’ Antiquities , October1936, 10(39), pp. 261–81.

Boling, R. ‘‘America’s Dirty Little Secret.’’Modern Maturity , March 1998, 41(2) , pp.86–89.

Bound, John; Schoenbaum, Michael; Stinebrick-ner, Todd R. and Waidmann, Timothy. ‘‘TheDynamic Effects of Health on the LaborForce Transitions of Older Workers.’’ Na-tional Bureau of Economic Research (Cam-bridge, MA ) Working Paper No. 6777,November 1998.

Bronowski, Jacob. The ascent of man . London:British Broadcasting Corporation, 1973.

Brzezinski, Zbigniew. ‘‘New Dimensions: For-eign Policy and Human Rights.’’ Current,September 1996, 9(385), pp. 29–33.

Carey, Vincent J.; Walters, Ellen E.; Colditz, Gra-ham A.; Soloman, Caren G.; Willett, Walter C.;Rosner, Bernard A.; Speizer, Frank E. andManson, JoAnn E. ‘‘Body Fat Distributionand Risk of Non-Insulin-Dependent Diabe-tes Mellitus in Women. The Nurses’ HealthStudy.’’ American Journal of Epidemiol-ogy , April 1, 1997, 145(7) , pp. 614–19.

Chin, N. K.; Ng, T. P.; Hui, K. P. and Tan, WanC. ‘‘Population Based Standards for Pul-monary Function in Non-Smoking Adults.’’Respirology , June 1997, 2(2) , pp. 143–49.

Choucri, Nazli, ed. Environmental challengesand international responses . Cambridge,MA: MIT Press, 1993.

Christensen, Kaare; Vaupel, James W.; Holm,Niels V. and Yashin, Anatoli I. ‘‘MortalityAmong Twins After Age 6: Fetal OriginsHypothesis Versus Twin Method.’’ BritishMedical Journal , February 18, 1995,310(6997), pp. 432–36.

Cipolla, Carlo M. The economic history ofworld population . Harmondsworth, U.K.:Penguin Books, 1974.

Clark, Graham. World prehistory: An outline.Cambridge: Cambridge University Press,1961.

Costa, Dora L. The evolution of retirement: AnAmerican economic history, 1880 – 1990 .Chicago: University of Chicago Press,1998a.

. ‘‘The Unequal Work Day: A Long-Term View.’’ American Economic Review ,May 1998b ( Papers and Proceedings ) ,88(2) , pp. 330–34.

. ‘‘Understanding the Twentieth Cen-tury Decline in Chronic ConditionsAmong Older Men.’’ Mimeo, Massachu-setts Institute of Technology and NationalBureau of Economic Research, September1998c.

Cresswell, J. L.; Egger, P.; Fall, C. H.; Osmond,C.; Fraser, R. B. and Barker, D. J. ‘‘Is the Ageof Menopause Determined In-Utero?’’Early Human Development , September 19,1997, 49(2) , pp. 143–48.

Curwen, E. Cecil. ‘‘Prehistoric Farming ofEurope and the Near East,’’ Pt. I., Ploughand pasture, the early history of farming .New York: Henry Schuman, 1953, pp. 3–147.

Cutler, David M. and Meara, Ellen. ‘‘The Med-ical Costs of the Young and Old: A Forty-Year Perspective,’’ in David A. Wise, ed.,Frontiers in the economics of aging . Chi-cago: University of Chicago Press, 1998,pp. 215–46.

Cutler, David M. and Richardson, Elizabeth.‘‘The Value of Health: 1970–1990.’’ Amer-ican Economic Review , May 1998 (Papersand Proceedings) , 88(2) , pp. 97–100.

Dasgupta, Partha. An enquiry into well-beingand destitution . Oxford: Clarendon Press,1993.



Derry, T. K. and William, T. I. A short historyof technology . London: Oxford UniversityPress, 1960.

Dorn, Joan M.; Schisterman, Enrique, F.; Win-kelstein, Warren, Jr. and Trevisian, Maurizio.‘‘Body Mass Index and Mortality in a Gen-eral Population Sample of Men and Women.The Buffalo Health Study.’’ AmericanJournal of Epidemiology , December 1,1997, 146(11), pp. 919–31.

Eisner, Robert. ‘‘Extended Accounts for Na-tional Income and Product.’’ Journal ofEconomic Literature , December 1988,26(4) , pp. 1611–84.

Fagan, B. M. People of the earth , 2nd Ed. Bos-ton, MA: Little, Brown & Co., 1977.

Finch, Caleb E. ‘‘Comparative Perspectives onPlasticity in Human Aging and Life Spans,’’in Kenneth W. Wachter and Caleb E. Finch,eds., Between Zeus and the salmon: Thebiodemographics of longevity . Washington,DC: National Academy Press, 1997, pp.245–68.

Fletcher, W. L. and Hansson, R. O. ‘‘Assessingthe Social Components of Retirement Anx-iety.’’ Psychology and Aging , March 1991,6(1) , pp. 76–85.

Fleurbaey, Marc and Michel, Philippe. ‘‘OptimalGrowth and Transfers between Genera-tions.’’ Recherches Economiques de Lou-vain , 1994, 60(3) , pp. 281–300.

Fogel, Robert W. ‘‘Nutrition and the Decline inMortality Since 1700: Some PreliminaryFindings,’’ in Stanley L. Engerman andRobert E. Gallman, eds., Long-term factorsin American economic growth . Chicago:University of Chicago Press (for the Na-tional Bureau of Economic Research ) ,1986, pp. 439–555.

. ‘‘Economic and Social Structure foran Ageing Population.’’ PhilosophicalTransactions of the Royal Society of Lon-don, Series B: Biological Sciences , De-cember 29, 1997, 352 (1363) , pp. 1905–17.

. The Fourth Great Awakening andthe future of egalitarianism. Chicago: Uni-versity of Chicago Press, 2000 (forthcoming).

Fogel, Robert W. and Costa, Dora L. ‘‘A Theoryof Technophysio Evolution, with Some Im-plications for Forecasting Population,Health Care Costs, and Pension Costs.’’ De-

mography , February 1997, 34(1) , pp. 49–66.

Fogel, Robert W. and Floud, Roderick. ‘‘A The-ory of Multiple Equilibria Between Popu-lations and Food Supplies: Nutrition,Mortality and Economic Growth in France,Britain and the United States, 1700–1980.’’Mimeo, University of Chicago, 1999(forthcoming).

Forsen, T.; Eriksson, J. G.; Tuomilehto, J.; Te-ramo, K.; Osmond, C. and Barker, D. J. P.‘‘Mother’s Weight in Pregnancy and Coro-nary Heart Disease in a Cohort of FinnishMen: Follow-Up Study.’’ British MedicalJournal , October 4, 1997, 315(7112), pp.837–40.

Frankel, Stephen; Elwood, Peter; Sweetnam, Pe-ter; Yarnell, John and Davey Smith, George.‘‘Birthweight, Body Mass Index in MiddleAge, and Incident Coronary Heart Dis-ease.’’ Lancet , November 30, 1996,348(9040), pp. 1478–80.

Garber, Alan M.; MaCurdy, Thomas and Mc-Clellan, Mark. ‘‘Diagnosis and Medicare Ex-penditures at the End of Life,’’ in David A.Wise, ed., Frontiers in the economics of ag-ing . Chicago: University of Chicago Press,1998, pp. 247–74.

Goldin, Claudia. Understanding the gendergap: An economic history of Americanwomen . New York: Oxford UniversityPress, 1990.

Griliches, Zvi. ‘‘Productivity, R&D, and theData Constraint.’’ American Economic Re-view , March 1994, 84(1) , pp. 1–23.

Hamermesh, Daniel S. ‘‘When We Work.’’ Amer-ican Economic Review, May 1998 (Papersand Proceedings), 88(2), pp. 321–25.

Heckman, James J. and Taber, Christopher R.‘‘Econometric Mixture Models and MoreGeneral Models for Unobservables in Du-ration Analysis.’’ National Bureau of Eco-nomic Research ( Cambridge, MA )Technical Paper No. 157, June 1994.

Henretta, J. C. ‘‘Changing Perspective onRetirement.’’ Journals of Gerontology,Series B: Psychological Sciences and So-cial Sciences , January 1997, 52 (1 ) , pp.S1–03.

Henry, Jan A.; Bolla, Manjeet; Osmond, Clive; Fall,Caroline; Barker, David J. P. and Humphries,Steve E. ‘‘The Effects of Genotype and Infant



Weight on Adult Plasma Levels of Fibrino-gen, Factor VII, and LDL Cholesterol AreAdditive.’’ Journal of Medical Genetics, July1997, 34(7), pp. 553–58.

Juster, F. Thomas and Suzman, Richard. ‘‘AnOverview of the Health and RetirementStudy.’’ Journal of Human Resources ,1995, Health and Retirement Study Supp.,30 , pp. S7–56.

Kannisto, Vaino. Development of oldest-oldmortality, 1950–1990: Evidence from 28developed countries , Monographs on Pop-ulation Aging, Vol. 1. Odense, Denmark:Odense University Press, 1994.

. The advancing frontier of survival .Odense, Denmark: Odense UniversityPress, 1996.

Kannisto, Vaino; Christensen, Kaare and Vaupel,James W. ‘‘No Increased Mortality in LaterLife for Cohorts Born During Famine.’’American Journal of Epidemiology , June 1,1997, 145(11), pp. 987–94.

Kaplan, Hillard. ‘‘The Evolution of the HumanLife Course,’’ in Kenneth W. Wachter andCaleb E. Finch, eds., Between Zeus and thesalmon: The biodemographics of longevity .Washington, DC: National Academy Press,1997, pp. 175–211.

Koslowski, Peter, ed. Ethics in economics, busi-ness, and economic policy . New York:Springer, 1992.

Kotani, K.; Nishida, M.; Yamashita, S.; Funa-hashi, T.; Fujioka, S.; Tokunaga, K.; Ishikasa,K.; Tarui, S. and Matsuzawa, Y. ‘‘Two Dec-ades of Annual Medical Examinations inJapanese Obese Children: Do Obese Chil-dren Grow into Obese Adults?’’ Interna-tional Journal of Obesity and RelatedMetabolic Disorders , October 1997,21(10), pp. 912–21.

Koupilova, Ilona; Leon, David A. and Vagero,Denny. ‘‘Can Confounding by Sociode-mographic and Behavioural Factors Ex-plain the Association Between Size atBirth and Blood Pressure at Age 50 inSweden?’’ Journal of Epidemiology andCommunity Health , February 1997,51 (1 ) , pp. 14–18.

Kramer, Michael S. and Joseph, K. S. ‘‘Enigmaof Fetal /Infant Origins Hypothesis.’’ Lan-cet , November 9, 1996, 348(9037), pp.1254–55.

Kuznets, Simon. ‘‘Statistics and Economic His-tory.’’ Journal of Economic History , May1941, 1(1) , pp. 26–41.

. ‘‘Long-Term Changes in the Na-tional Income of the United States ofAmerica Since 1870,’’ in Simon Kuznets,ed., Income and wealth of the UnitedStates: Trends and structure , Income andWealth, Series 2, International Associa-tion for Research in Income and Wealth.Cambridge: Bowes and Bowes, 1952, pp.29–241.

Laslett, Peter. A fresh map of life . Cambridge,MA: Harvard University Press, 1991.

Lauderdale, D. S. and Rathouz, P. J. ‘‘Evidenceof Environmental Suppression of FamilialResemblance: Height Among Civil WarBrothers.’’ Mimeo, University of Chicago,1998.

Law, Catherine M. and Shiell, Alistair W. ‘‘IsBlood Pressure Inversely Related to BirthWeight? The Strength of Evidence from aSystematic Review of the Literature.’’Journal of Hypertension , August 1996,14(8) , pp. 935–41.

Lebergott, Stanley. Pursuing happiness: Amer-ican consumers in the twentieth century .Princeton, NJ: Princeton University Press,1993.

Lee, Chulhee. ‘‘Essays on Retirement andWealth Accumulation in the United States,1850–1990.’’ Ph.D. dissertation, Univer-sity of Chicago, 1996.

. ‘‘Socioeconomic Background, Dis-ease, and Mortality Among Union ArmyRecruits: Implications for Economic andDemographic History.’’ Explorations inEconomic History , January 1997, 34(1) ,pp. 27–55.

. ‘‘The Relationship of the Growth inIncome Inequality to the Organization ofWork and the Structure of Consumption,’’in Robert W. Fogel, The Fourth GreatAwakening and the future of egalitarianism.Chicago: University of Chicago Press, 2000(forthcoming).

Lenk, Hans. ‘‘Value Changes and the Achiev-ing Society: A Social-Philosophical Per-spective,’’ in OECD societies in transition:The future of work and leisure . Paris: Or-ganization for Economic Cooperation andDevelopment, 1994, pp. 81–94.



Manton, Kenneth G.; Corder, Larry and Stallard,Eric. ‘‘Chronic Disability Trends in ElderlyUnited States Populations: 1982 – 1994.’’Proceedings of the National Academy ofSciences, U.S.A. , March 18, 1997a, 94(6) ,pp. 2593–98.

. ‘‘Monitoring Changes in the Healthof the U.S. Elderly Population: Correlateswith Biomedical Research and Clinical In-novations.’’ FASEB Journal , October1997b, 11(12), pp. 923–30.

. ‘‘Economic Effects of Reducing Dis-ability.’’ Paper presented at the AmericanEconomic Association Annual Meeting,January 3–5, 1998, Chicago, IL.

McNeill, William. A world history . New York:Oxford University Press, 1971.

Must, Avita; Jacques, Paul F.; Dallal, Gerard E.;Bajema, Carl J. and Dietz, William H. ‘‘Long-Term Morbidity and Mortality of Over-weight Adolescents. A Follow-Up of theHarvard Growth Study of 1922 to 1935.’’New England Journal of Medicine , Novem-ber 5, 1992, 327(19), pp. 1350–55.

Mutchler, J. E.; Burr, J. A.; Pienta, A. M. andMassagli, M. P. ‘‘Pathways to Labor ForceExit: Work Transitions and Work Instabil-ity.’’ Journals of Gerontology , Series B:Psychological Sciences and Social Sci-ences , January 1997, 52(1) , pp. S4–12.

Nordhaus, William D. ‘‘Discounting in Eco-nomics and Climate Change: An EditorialComment.’’ Climatic Change , October1997, 37(2) , pp. 315–28.

Olshansky, S. Jay and Carnes, Bruce A. ‘‘EverSince Gompertz.’’ Demography , February1997, 34(1) , pp. 1–15.

Pahkala, K.; Kivela, S. L. and Laippala, P. ‘‘Socialand Environmental Factors and DysthymicDisorder in Old Age.’’ Journal of Clinical Ep-idemiology, July 1992, 45(7), pp. 775–83.

Paneth, Nigel and Susser, Mervyn. ‘‘Early Originof Coronary Heart Disease (The ‘Barker Hy-pothesis’).’’ British Medical Journal, Feb-ruary 18, 1995, 310(6977), pp. 411–12.

Parrington, Vernon Louis. Main currents in Amer-ican thought: An interpretation of Americanliterature from the beginnings to 1920. NewYork: Harcourt, Brace & Co., 1930.

Pencavel, John. ‘‘Assortative Mating bySchooling and the Work Behavior of Wivesand Husbands.’’ American Economic Re-

view , May 1998 (Papers and Proceedings) ,88(2) , pp. 326–29.

Perry, Ivan J.; Beevers, D. G.; Whincup, P. H. andBareford, D. ‘‘Predictors of Ratio of PlacentalWeight to Fetal Weight in Multiethnic Com-munity.’’ British Medical Journal, February18, 1995, 310(6977), pp. 436–39.

Piggott, Stuart. Ancient Europe from the begin-nings of agriculture to classical antiquity .Chicago: Aldine, 1965.

Pope, Clayne L. ‘‘Adult Mortality in Americabefore 1900: A View from Family Histo-ries,’’ in Claudia Goldin and Hugh Rock-off, eds., Strategic factors in nineteenthcentury American economic history: Avolume to honor Robert W. Fogel . Chi-cago: University of Chicago Press, 1992,pp. 267–96.

Pope, Clayne L. and Wimmer, Larry T. ‘‘Agingin the Early 20th Century.’’ American Eco-nomic Review , May 1998 (Papers and Pro-ceedings) , 88(2) , pp. 217–21.

Ravelli, A. C. J.; van der Meulen, J. H. P.; Michels,R. P. J.; Osmond, C.; Barker, D. J. P.; Hales, C.N. and Bleker, O. P. ‘‘Glucose Tolerance inAdults After Prenatal Exposure to Famine.’’Lancet, January 17, 1998, 351(9097), pp.173–77.

Robinson, John P. and Godbey, Geoffrey. Timefor life: The surprising ways Americans usetheir time . University Park, PA: Pennsyl-vania State University, 1997.

Rose, Michael R. The evolutionary biology ofaging . New York: Oxford University Press,1991.

Rowe, J. W. and Kahn, R. L. ‘‘Successful Ag-ing.’’ Gerontologist , August 1997, 37(4) ,pp. 433–40.

Samuels, S. C. ‘‘Midlife Crisis: Helping Pa-tients Cope with Stress, Anxiety, and De-pression.’’ Geriatrics , July 1997, 52(7) ,pp. 59–63.

Scrimshaw, Nevin S., ed. Community-basedlongitudinal nutrition and health studies:Classical examples from Guatemala,Haiti, and Mexico . Boston: InternationalFoundation for Developing Countries( INFDC) , 1995.

Scrimshaw, Nevin S. ‘‘More Evidence That Foe-tal Nutrition Contributes to Chronic Diseasein Later Life.’’ British Medical Journal ,October 4, 1997, 315(7112), pp. 825–26.



Sen, Amartya K. Inequality reexamined . NewYork: Russel Sage Foundation, 1996.

Sharp, Dan S.; Burchfiel, Cecil M.; Curb, J. Da-vid; Rodriguez, Beatriz L. and Enright, Paul L.‘‘The Synergy of Low Lung Function andLow Body Mass Index in Predicting All-Cause Mortality Among Older Japanese-American Men.’’ Journal of the AmericanGeriatric Society , December 1997, 45(12),pp. 1464–71.

Shaw, George Bernard. The intelligent woman’sguide to socialism. The collected works ofGeorge Bernard Shaw , Vol. 20, Ayot St.Lawrence Ed. New York: Wm. H. Wise,1931 [1928].

Shusterman, Richard. ‘‘Pragmatism and Liber-alism Between Dewey and Rorty.’’ PoliticalTheory , August 1994, 22(3) , pp. 391–413.

Slicher von Bath, B. H. The agrarian history ofWestern Europe A.D. 500–1850 . London:Edward Arnold, 1963.

Smith, Adam. An inquiry into the nature andcauses of the wealth of nations . New York:Modern Library, 1937.

Smith, James P. and Kington, Raynard. ‘‘De-mographic and Economic Correlates ofHealth in Old Age.’’ Demography , Febru-ary 1997, 34(1) , pp. 159–70.

Stein, C. E.; Fall, C. H.; Kumaran, K.; Osmond,C.; Cox, V. and Barker, D. J. ‘‘Fetal Growthand Coronary Heart Disease in South In-dia.’’ Lancet , November 9, 1996,348(9037), pp. 1269–73.

Stein, C. E.; Kumaran, K.; Fall, C. H.; Shaheen,S. O.; Osmond, C. and Barker, D. J. ‘‘Relationof Fetal Growth to Adult Lung Function inSouth India.’’ Thorax , October 1997,52(10), pp. 895–99.

Strauss, John and Thomas, Duncan. ‘‘HumanResources: Empirical Modeling of House-hold and Family Decisions,’’ in Jere Behr-man and T. N. Srinivasan, eds., Handbookof development economics , Vol. 3A. Am-sterdam: North-Holland, 1995, pp. 1883–2023.

Thomas, Duncan and Strauss, John. ‘‘Health andWages: Evidence on Men and Women inUrban Brazil. ( Analysis of Data onHealth) .’’ Journal of Econometrics , March1997, 77(1) , pp. 159–86.

Toman, Michael. ‘‘Economics and ‘Sustaina-bility’: Balancing Trade-Offs and Impera-

tives.’’ Land Economics , November 1994,70(4) , pp. 399–413.

Trewartha, Glenn T. A geography of popula-tions: World patterns . New York: Wiley,1969.

U.S. Bureau of the Census, Department of Com-merce. Historical statistics of the UnitedStates, colonial times to 1970 . Washington,DC: U.S. Government Printing Office,1975.

. Statistical abstract of the UnitedStates 1997 , 117th Ed. Washington, DC:U.S. Government Printing Office, 1997.

Vagero, Denny and Leon, David. ‘‘IschaemicHeart Disease and Low Birth Weight: ATest of the Fetal Origins Hypothesis fromthe Swedish Twin Registry.’’ Lancet , Jan-uary 29, 1994, 343(8892), pp. 260–64.

Vaupel, James W. ‘‘Kindred Lifetimes: FrailtyModels in Populations Genetics,’’ in JulianAdams, David A. Lam, Albert I. Hermalin,and Peter E. Smouse, eds., Convergent is-sues in genetics and demography . NewYork: Oxford University Press, 1990, pp.155–70.

. ‘‘Trajectories of Mortality at Ad-vanced Ages,’’ in Kenneth W. Wachter andCaleb E. Finch, eds., Between Zeus and thesalmon: The biodemographics of longevity .Washington, DC: National Academy Press,1997, pp. 17–37.

Veblen, Thorsten. The theory of the leisureclass: An economic study of institutions .New York: Modern Library, 1934 [1899].

Wachter, Kenneth W. and Finch, Caleb E., eds.Between Zeus and the salmon: The biode-mographics of longevity . Washington, DC:National Academy Press, 1997.

Waidmann, Timothy; Bound, John and Schoen-baum, Michael. ‘‘The Illusion of Failure:Trends in the Self-Reported Health of theElderly.’’ National Bureau of Economic Re-search (Cambridge, MA) Working PaperNo. 5017, February 1995.

Wilmoth, John R. ‘‘Are Mortality Rates Fallingat Extremely High Ages? An InvestigationBased on a Model Proposed by Coale andKisker.’’ Population Studies , July 1995,49(2) , pp. 281–95.

. ‘‘In Search of Limits,’’ in KennethW. Wachter and Caleb E. Finch, eds., Be-tween Zeus and the salmon: The biodemo-



graphics of longevity . Washington, DC:National Academy Press, 1997, pp. 38–64.

Wilson, Sven E. and Nguyen, Louis L. ‘‘SecularTrends in the Determinants of DisabilityBenefits.’’ American Economic Review ,May 1998 ( Papers and Proceedings ) ,88(2) , pp. 227–31.

Wilson, William Julius. When work disappears:The world of the new urban poor. NewYork: Knopf, 1996.

Wrigley, E. A. People, cities and wealth: Thetransformation of traditional society . Ox-ford: Basil Blackwell, 1987.

Yamashita, K.; Kobayashi, S.; Yamaguchi, S.; Ko-ide, H.; Imaoka, K.; Bokura, H. and Suyama,N. ‘‘Feelings of Well-Being and Depressionin Relation to Social Activity in Normal El-derly People’’ [ in Japanese] . Nippon RonenIgakkai Zasshi , August 1993, 30(8) , pp.693–97.

Catching Up with the Economy - die-gdi.de · / 3y15 mr03 Mp 1 Wednesday Dec 15 07:23 AM LP–AER...

Documents

Transcript of Catching Up with the Economy - die-gdi.de · / 3y15 mr03 Mp 1 Wednesday Dec 15 07:23 AM LP–AER...