World Bank Reprint Series: Number 279 RE P279March 1983

Nancy Birdsall

Fertility and Economic Changein Eighteenth and NineteenthCentury EuropeA Comment

Reprinted with permission from Population and Development Review, vol. 9, no. 1(March 1983), pp. 111-23.

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

World Bank Reprints

No. 250. J. B. Knight and R. H. Sabot, "From Migrants to Proletarians:Employment Experience, Mobility, and Wages in Tanzania," OxfordBulletin of Economics and Statistics

No. 251. M. Louise Fox, "Income Distribution in Post-1964 Brazil: NewResults," Journal of Economic History

No. 252. Nizar Jetha, "The Welfare Cost of Taxation: Its Meaning and Measure-ment," Bulletin for International Fiscal Documentation

No. 253. Larry E. Westphal, "Fostering Technological Mastery by Means ofSelective Infant-Industry Protection," Trade, Stability, Technology, andEquity in Latin America

No. 254. Gershon Feder, "On Exports and Economic Growth," Journal ofDevelopment Economics

No. 255. Mohan Munasinghe, "Third World Energy Policies; Demand Manage-ment and Conservation," Energy Policy

No. 256. Keith Marsden and Alan Roe, "The Political Economy of Foreign Aid:A World Bank Perspective," Labour and Society

No. 257. James A. Hanson, "Contractionary Devaluation, Substitution inProduction and Consumption, and the Role of the Labor Market,"Journal of International Economics

No. 258. Christiaan Grootaert, "The Conceptual Basis of Measures of House-hold Welfare and Their Implied Survey Data Requirements," TheReview of Income and Wealth

No. 259. Guy Pfeffermann and Richard Webb, "Poverty and Income Distribu-tion in Brazil," The Review of Income and Wealth

No. 260. Pradeep K. Mitra, "A Theory of Interlinked Rural Transactions,".Journal of Public Economics

No. 261. David L. Lindauer and Richard H. Sabot, "The Public/Private WageDifferential in a Poor Urban Economy," Journal of DevelopmentEconomics

No: 262. J. B. Knight and R. H. Sabot, "Labor Market Discrimination in a PoorUrban Economy," Journal of Development Studies

No. 263. Carl Dahlman and Larry Westphal, "Technical Effort in IndustrialDevelopment: An Interpretative Survey of Recent Research," TheEconomics of New Technology in Developing Countries

No. 264. Michael Bamberger, "The Role of Self-Help Housing in Low-CostShelter Programs for the Third World," Built Environment

No. 265. Bela Balassa, "The Adjustment Experience of Developing Economiesafter 1973," IMF Conditionality

No. 266. Bela Balassa, "Outward Orientation and Exchange Rate Policy inDeveloping Countries: The Turkish Experience," The Middle EastJournal

No. 267. Dipak Mazumdar, "Segmented Labor Markets in LDCs," AmericanEconomic Review

No. 268. Stephen P. Heyneman and William A. Loxley, "The Effect of Primary-School Quality on Academic Achievement across Twenty-nine High-and Low-Income Countries," The American Journal of Sociology

Fertility and EconomicChange in Eighteenthand Nineteenth CenturyEurope: A Comment

Nancy Birdsall

In a recent article in Population and Development Review,Richard M. Smith' attributes English fertility trends in the period 1541 to 1871to a "highly distinctive process of household formation." He suggests: "Su-perficially this process would seem susceptible to an economic interpretation.'"The purpose of this note is to propose that in fact there is a sensible economicinterpretation for the sustained rise in fertility in England throughout the eigh-teenth century and for the subsequent long-term decline throughout the nine-teenth century. The interpretation should not be set against the point Smith soappropriately emphasizes, of a particular household formation rule in Europe;it relies on that point. But it goes beyond Smith's discussion in two ways: (I)it explains the continued rise in English fertility through the end of the eigh-teenth century, despite the fall in the marriage rate and in the real wage, as:shown in his Figures 3 and 4 (combined here and reproduced as Figure 1); and(2) it explains the long-run trend of the decline in the nineteenth century (ad-mittedly with fluctuations), despite the sustained increase in real wages through-out that century (see Figure 1).

By an "economic interpretation" is meant simply that the long-run in-crease and then decline in fertility can be systematically related to changes inthe economic environment that induced changes in behavior regarding marriageand fertility. The economic interpretation is set firmly in the framework Smithdemonstrates is important-namely, rules of household formation that appearto be impervious (except perhaps in the very long run) to changes in economicconditions. For the period in Europe under consideration, the first rule was thatnewly married couples set up their own households; the corollary was that mar-riage did not occur until the couple had a sufficient measure of financial inde-pendence to do so. The economic interpretation also relies on the fact thatthroughout the period the structure of the labor market was changing. Smithshows this factor to be important in rural areas in the pretransitional period, but

POPULATION AND DEVELOPMENT REVIEW 9. NO 1 (MARCH 1983) 111

112 Fertility and Economic Change In Eighteenth and Nineteenth Century Europe

FIGURE 1 Gross reproduction rates compared with a real wageindex (both 25-year moving averages): England, 1541-1871

950

900 _

850 I

800 -

750 - 3.2

700 - 3.0

co 650 - i \ /)( ~ 2.8 0

X 600 - s Real wage i /t\s \ \ \ 2.6 a

550 <( _t _ _2.40

450 \ /W GRR \-2.2

450 ~~~~~~~~~~~~~~~~~~~~~2.0

400 _ s-sX _ 1.8

350 . . 1.6

T ,,,,, T1551 1601 1651 1701 1751 1801 1851 1901

YearSOURCE: Smith, cited in note 1, Figures 3 and 4, p. 601.

does not relate it to industrialization, in retrospect the central economic force ineighteenth and nineteenth century England.

The suggested interpretation also resolves the seeming contradictionSmith alludes to between John Caldwell's theory of fertility decline and theMarxist view, adapted by Charles Tilly. Caldwell theorizes that "Westerniza-tion," the accompanying emphasis on economic individualism, and the reversalof "wealth flows" to favor children rather than parents bring on fertility de-cline.2 Tilly suggests that "proletarianization" of the work force (with West-ernization) brings on fertility increases (or at least assures continued highfertility) as the working masses attempt to accumulate the one factor of produc-tion they control-labor power.3 By extending Smith's analysis into the nine-teenth century to take into account the effect of industrialization, and byconsidering more carefully how the labor market changed within the period of"proletarianization," the two theories can be shown to fit easily into a simpleeconomic interpretation of marriage and fertility behavior and to explain verywell both the eighteenth century increase and the nineteenth century decline infertility.

Nancy Birdsall 113

The eighteenth century increase in Englishpopulation growth

The positive association between population growth and the stirrings of in-dustrialization in eighteenth century England is a long-honored one. Kuznetshas even defined modern economic growth in the nonindustrialized countriesas a sustained increase not only in per capita product but also in population.4

The rate of growth in the eighteenth century was probably not much higherthan parts of Europe had experienced in earlier periods when nature wasgenerous: the twelfth and thirteenth centuries, for example, as well as thesixteenth and early seventeenth centuries. It is exceptional only in retrospect,because we know it was followed not by a subsequent drop in population, butby continuous growth.

Smith's discussion assumes that the eighteenth century increase in thepopulation growth rate-due, as is now widely believed, not merely to mortalitydeclines but also to an increase in marriage rates and thus in fertility5 -waspart of the pattern of long swings in fertility that occurred for the severalpreceding "pretransitional" centuries. His contribution lies in showing howthe close relation between changes in real wages and subsequent (lagged)changes in marriage rates was a logical consequence of the European householdformation rule, the institutional arrangement of agrarian service, and whatLesthaeghe calls "communal risk devolution"6 -that is, that in England theparish rather than kin took care of the poor.

The question, however, is whether Smith's emphasis on rural and ag-ricultural labor markets can explain the seeming anomaly that emerges in thelatter half of the eighteenth century, of continued increases in the gross re-production rate at the same time that the real wage declined. This is illustratedin Figure 1, which reproduces the gross reproduction rate as shown in hisFigure 3 and the real wage as shown in his Figure 4, for the period 1541to 1871.

A demographic interpretation of the continuing rise in fertility in thelatter half of the eighteenth century despite the fall in real wages is that it wasa continuation of the same earlier trend, with the lag between the peak inwages (1741) and the peak in fertility (1816) lengthened by measurementfactors associated with changing marriage rates. Wrigley and Schofield, forexample, suggest that throughout the entire period, in the absence of consciousfertility control, fertility response to changing economic circumstances couldtake place only through the timing and incidence of marriage (rather thanthrough immediate response of already-married couples), which would neces-sitate the lag between the wage and fertility series shown in the figure. Thelag in the later period is longer, they propose, because when age at marriageis falling fast, a telescoping effect will increase marriage rates more than thelower marriage age itself would have implied, and similarly increase fertilityrates more. They also suggest that rising rates of pregnancy at marriage andof illegitimacy in the late nineteenth century help explain the long period ofrising fertility in the face of falling wages.7

114 Fertility and Economic Change In Eighteenth and Nineteenth Century Europe

However, the departure from the previous pattern in the form of persistenthigh fertility over a long period of falling wages can also be explained bytaking into account the economic changes occurring with industrialization-changes that altered permanently the previous long-run swings of populationgrowth. Moreover, the new economic reality helps explain the second reversal,that is, rising wages coincident with falling fertility, that had taken hold bythe end of the nineteenth century, signaling the onset of the demographictransition. But first we examine the situation in the late eighteenth century.

Early industrialization and fertilityincrease

Earlier marriage

How can the continued rise in fertility in the latter half of the eighteenthcentury, in the face of declining real wages, be explained? A now-conventionalexplanation is the rise of rural industry. The opportunity to work in cottageor "putting-out" industries destroyed the tie between marriage and access toa living through work exclusively in agriculture and contributed to lower ageat marriage and higher rates of nuptiality (i.e., higher proportions eventuallymarried). It may even have led to higher marital fertility rates, though on thisthere is much less evidence. This explanation is actually a simple extensionof Smith's own approach, taking more explicitly into account new opportunitiesfor wage work outside of agriculture, in that much of the increase in fertilityis attributed to the increased rate of new household formation. In Flanders,where many families supplemented agricultural income by cottage manufactureof linen, the marriage rate responded positively to an increase in the linen/ryeprice ratio.8 Chambers argues that labor force growth in England and the risein fertility (1780 to 1851) were due to the growth of alternative industrialopportunities leading to marriage where impartible inheritance had once re-stricted marriage. For Nottinghamshire in the first half of the nineteenth cen-tury, he finds the highest rates of fertility in the industrial textile areas.9 Levineuses a time series to show the effects of rural industrialization on increasingrates of marriage throughout the eighteenth century in Shepshed, England,where the stocking industry provided ever-expanding opportunities.' 0 More-over, there is some evidence that in nonindustrial areas where a relativelyprosperous smallholder agricultural system maintained its grip on the family(outside of England, where wage labor in agriculture had been prevalent sincethe sixteenth century), fertility (and presumably marriage rates) did not rise-in these areas Smith's pretransitional explanation still applied. For example,fertility in the nineteenth century eastern United States remained relatively low(compared with high fertility in the west, where limited access to land was nota factor). " Compared with the overall population, the agricultural populationof eighteenth century Netherlands increased less rapidly, 12 and had persistentlyhigh rates of celibacy. 13 Where alternative employment was limited, the "desire

Nancy Birdsall 115

to acquire, to hold on to, and to accumulate land" is cited as a cause of thepreindustrial fertility decline in nineteenth century Austria-Hungary." 4

The rise of rural industry explains why a fall in real wages was accom-panied, for the first time in eighteenth century England, by a persistent risein marriage rates and fertility rather than, with some lag, a fall. The effect ofchanges in agricultural wages and prices on the demand for agrarian servicebrought out by Smith, though undoubtedly still a factor in agricultural areas,was countered in the aggregate once there were alternative wage opportunitiesoutside of agriculture. Rural industry may not have increased the averageincome of any one family, but it increased the number of families by affordingto more young people the opportunity to set up a household. This effect waspresumably even more important where impartible inheritance of land was therule. In studies relating family structure to economic change in the nineteenthcentury, several scholars conclude that the single-heir arrangement, with largefamilies but many celibates (thus few families), in aggregate produced lowerpopulation growth than the equal division of land, with smaller individualfamilies but more marriages. '5 The lowest population growth in the Netherlandsis recorded for that province (Overjissel) where until 1795 the law dictatedimpartible inheritance.' 6

Impartible inheritance over the longer run, however, may have had theeffect of easing the transition to a permanent industrial labor force workingin factories, since the nonheir siblings in a family would have responded moreeasily to nonagricultural employment opportunities in cities. Urban factorieswere less likely to compete successfully with rural industry where division ofland was the norm, as in France."' In England, in contrast, the greater prev-alence of a single-heir system, the lower proportion of the rural populationthat owned their own land, and the occurrence of enclosure, reducing evenfurther that proportion, produced low fertility in rural areas but high rates ofemigration to industrial regions, where fertility was high. This may providea partial explanation for the more rapid growth of population in England thanin France during the nineteenth century.

The increasing availability of industrial earnings opportunities had twocomplementary effects: it increased the rate of marriage by opening up income-earning opportunities to more people; and it lowered the age of marriage sinceindustrial workers reach their maximum earnings at an earlier age than agri-cultural workers. In places such as Flanders and textile regions of England,it encouraged more rapid growth within rural areas. Where an agriculturalliving was restricted to a small part of the population, industrial regions becameareas of immigration and of high fertility.'8

The investment value of children

Industrialization in rural areas and later in urban areas introduced another factorthat may have contributed to higher fertility rates: higher marital fertility. Incottage industries, children became an asset to the household. There is virtuallyno evidence of purposeful increases in marital fertility among couples in rural

116 Fertility and Economic Change in Eighteenth and Nineteenth Century Europe

industrial areas; but couples married earlier were bound to have more childrenby reason of a longer period of risk. Mendels reports that in Flanders averagebirths within marriage were 4.8 in linen-producing areas, compared with 3.8in adjacent areas where the soil was more fertile and families were less de-pendent on domestic industry. 19 Among counties in England in 1851 there wasa positive correlation between gross reproduction rates and percentages ofchildren 10-14 years who were employed. 20 Rising rates of illegitimate birthsand of pregnancy at marriage noted by Wrigley and Schofield also suggest thatchildren were becoming less of an economic burden. Moreover, there is someevidence that when employmeht opportunities in rural areas eventually de-clined, as in Shepshed after 1815, there was conscious fertility control toreduce the new burden children posed once they could not work.2 ' In a similarvein, Tilly suggests that proletarians always had higher fertility than nonpro-letarians and that the late eighteenth century rise in natural increase was dueboth to the increase in the proportion of the population that was proletarianand to their higher rates of natural increase. Although the higher rates ofincrease (due to fertility, not lower mortality) could be due entirely to lowermarriage age, Tilly points out that there are well-documented preindustrialcases of groups whose fertility patterns do not conform to a "natural" fertilityregime, but meet the "stringent test of parity-dependence. " 2 2 In short, althoughwe cannot find evidence that couples increased the tempo of reproductionwithin marriage to provide more labor power in the face of falling wages (thiswould be what Smith calls the Marxisant view), there is some evidence inpretransitional societies of divergence from a natural fertility regime, and someevidence immediately after the period of rural industrialization that coupleslimited their fertility within marriage as employment opportunities for childrendwindled.

Although children must have worked in agriculture as well, and partic-ularly on family farmns, the marginal return to their labor was necessarilyrestricted by the size of the holding; this was less true in home industry andcertainly less true when children began to work in factories. In a quite literalsense, children were thus an investment good during the early industrial period.This was particularly the case in England, where child labor laws (discussedbelow) restricted the employment of children at a much later stage in indus-trialization than they did elsewhere.

Furthermore, where parents no longer owned land, children were themost accessible form of investment for the future, when old age would makeworking impossible. In Preston, a Lancashire cotton town, data pieced togetheron household structure in the mid-nineteenth century indicate that 80 percentof grandparents lived with their children, suggesting limited alternatives.2 3

In short, the costs of children in this period were undoubtedly compar-atively low. Since few attended schools and housing standards were poor, themarginal cost of an additional child would be primarily in food. Yet childrencould contribute to household income in rural industrial areas, and parentscould hope for children's support in their old age.

Nancy Birdsall 117

Female employment

When employable women must give up income to take care of children,children have an opportunity cost. 24 This opportunity cost is measured bypotential rather than actual earnings, so that the cost of children to bettereducated women, whose time is consequently worth more, is higher regardlessof whether they actually work. The cost of children rises for a particular motherif the demand for female labor pushes up female wages, again regardless ofwhether she actually works. In fact, studies of fertility determinants indicatethat the husband's income in a family is a good indicator of the positive"income effect" on fertility, and the wife's earnings opportunities a goodindicator of the "substitution effect.' '25

On the other hand, where female employment and child care are com-patible, as was true in rural industrial areas where work took place within thehome, female employment is unlikely to depress fertility. In fact, under suchcircumstances, the additional income may permit working mothers to enlargetheir families, and large families may draw women into the labor force becauseof the need to increase family income. In a study of cottage industries in Japan,fertility of women working in cottage industries was significantly higher thanthat of women who worked away from home, and only slightly lower thanthat of housewives. 26

The situation of many working women in early industrial Europe wasundoubtedly similar to that of these Japanese cottage industry workers. Un-fortunately it is unlikely that there are household data from the period withinformation on both occupational status of women and fertility. Aggregate datado indicate that there was a high correlation, by county in England, betweenthe proportion of women in industrial employment and marriage rates.27 Fur-thermore, among counties with high female employment in 1870, those wheredomestic industry was still predominant had higher marriage and fertility ratesthan those where factory industry dominated. 28

Migration opportunities

Where industrialization per se did not stimulate fertility, the opportunity tomigrate toward industrializing regions may have.29 Rural areas close to theindustrializing regions of Lancashire and Warwickshire experienced both highfertility and high rates of outmigration. 30 The growth of populations throughoutthe European continent in the eighteenth and nineteenth centuries, before in-dustrialization, may also have been due to higher fertility among persons awarethat their children would not be restricted to employment opportunities in theimmediate area. This would be particularly true of regions where land wasalready fragmented or where peasants had only a tenuous right to the fruits oftheir agricultural labor, as in Prussia. It would be less true where land wasmore equitably distributed and where families had a more reasonable hope ofpassing a substantial piece of property on to their offspring, as in France. Tomy knowledge, there is as yet no analysis for different agricultural regions ofthe relationship between outmigration and fertility.

118 Fertility and Economic Change In Eighteenth and Nineteenth Century Europe

Later industrialization and fertility decline

The conventional interpretation of "proletarianization" is the shift from self-employment with control over the means of production to wage work forothers. In England, much of the agricultural labor force was proletarian by thebeginning of the eighteenth century. In late eighteenth century England, andthroughout the nineteenth century in much of Europe, however, industriali-zation led to further proletarianization of the work force. New opportunitiesfor wage work led to lower age at marriage and higher rates of marriage. Betteremployment opportunities, not only for young men but also for married womenand children, may have encouraged higher marital fertility; at the least, theyprovided no stimulus to lower fertility.

Proletarianization proceeded apace throughout the nineteenth century;but fertility by no means remained high. In England the gross reproductionrate, after peaking in 1816, declined throughout the century, except for theperiod between the mid- 1 840s and the 1 870s (see Figure 1). How, then, is thelong-run decline in fertility in England and later elsewhere in Europe to beexplained, and does the decline contradict Tilly's view that wage work causesincreased fertility?

An answer lies in consideration of Caldwell's view that a shift in "wealthflows" within families toward children leads to lower fertility. Caldwell assertsthat Westernization is the spur to such a shift. In an early statement he definedWesternization as the introduction in non-Western societies of Western ideas,via modem means of communication-for example, the radio-but especiallyvia imported education. More recently he has stated that what is significantfor fertility is "whether production is based on the familial organization oflabor or whether there is a free and monetized labor market (i.e., capitalistproduction)." Education remains the centerpiece of his approach, however:"Invariably educated children cost more and give less."3 '

In the modern world, a wage labor force and an educated one tend tocoincide (this is particularly so in Africa, the setting that probably inspiredmany of Caldwell's ideas on fertility). But in the early stages of industrializationin its first setting, eighteenth century England, this was by no means the case.Unskilled wage work (Tilly's "proletarianization") and the implied changein the relations of production may, other things equal, have induced higherfertility, as Marx expected. As suggested above, in the early stages of indus-trialization, after all, children were cheap. The cost of children was no higher,and possibly was even lower, than it had been in an agrarian society-lowerbecause future family welfare was no longer tied to the land that might haveto be divided, and because children could work, in home industries and laterin factories.

The cost of children

Thus, initially the severing of the link between family and income-producingproperty and the proletarianization of the work force were unaccompanied by

Nancy Birdsall 119

any change in the cost of children (except perhaps a lowering of their cost)and reduced the constraints on marriage and fertility. But by 1820, the costsof children probably were increasing. In England, the Factory Act of 1819,for example, prohibited for the first time the employment of children undernine in textile mills and limited those aged 10-16 to a 12-hour day. Althoughthe Act was probably not well enforced and only applied to about 13 percentof the factory labor force, it engendered considerable public discussion of thechild labor issue, and in fact by 1835 less than 0.5 percent of textile employeeswere children under ten.32 By 1840 lack of vigorous enforcement may havemeant the initial impact of the 1819 legislation had been spent. However, newand much more effective legislation was passed in 1867; the Factory ActsExtension Act and the Workshops Regulation Act extended coverage to workersin all factories, and provided for much more extensive enforcement.3 3 Not toomuch should be made of the fact that the 1819 and the 1867 legislation coincidedvery closely with renewed declines in fertility; surely the responses of individualparents to such changes in the costs of children would be neither so calculatednor so rapid. The larger point, however, is important: despite rising income,34

fertility fell throughout much of the nineteenth century, at a time when thecosts of children were increasing.

Education

By the latter part of the nineteenth century, moreover, the importance of passingon a new type of "wealth" further raised the cost of children. The new wealthwas in human capital; families could best ensure the prosperity of each newgeneration by providing it with an education adequate for "white-collar work":govemment, the military, the professions, or business (with the last categorybecoming increasingly complicated technologically and organizationally).

In England, the drop in fertility began among the professional middleclasses,35 which felt most keenly the necessity of educating their children (andnot in the wealthier landholding class). Banks provides a compelling accountof the rising costs of a public school and university education for the Victorianmiddle-class family, whether the objective was a career in law, medicine, orthe military.36 After 1871, military commissions could no longer be purchased,so that education made passage into the military accessible to more families.In the same period, patronage in the civil service was also abolished. Thedevelopment of the Public Boarding School system itself could be taken as anindication of how far the middle and upper classes were willing to go ineducating their children.

There was also a clear drive during this period for education among thelower middle classes. Although the actual costs of training for clerkships werenot large, parents did have to face the prospect of continuing to house, clothe,and feed adolescent sons who in an earlier generation would have been helpingto support themselves.3 7 Finally, in the period between 1840 and 1870 thenumber of persons assessed in the middle-income range expanded more rapidly

120 Fertility and Economic Change in Eighteenth and Nineteenth Century Europe

than the population as a whole.38 And the Elementary Education Act of 1870,followed by legislation in 1876 prohibiting employment of young childrenduring school hours, and legislation in 1891 making school compulsory, allindirectly increased the cost of children to working-class parents.39 As edu-cation became a more universal requirement for income generation in thetwentieth century, the fertility decline spread to other classes.40

Much more data will be required on the actual costs of children beforethe association suggested here can be seriously appraised. It is probably nota coincidence, however, that fertility declined and remained relatively low asthe costs of children rose and as the economic benefits of children to parentsfell, first with the abolition of child labor and later with the spread of education.

Type of employment

The type of industry undoubtedly influenced fertility through its effect onfamily structure. Evidence cited above regarding the apparent increases inmarriage rates and marital fertility in rural industrial areas applied particularlyin those areas where entire families could work at home. The rise of factory-based industry changed the structure of the industrial labor force and tendedto push fertility downward. A striking indication of this turnaround is Lan-cashire, where a fertility ratio "very much above the average in 1821 fellmuch more than the average in the period 1821-51 at a time when cottontextiles was becoming increasingly a factory industry." 4'

The removal of industrial work from the home to the factory increasedthe price parents had to pay for children, since either the mother had to dropout of the labor force or a parent-surrogate, such as a grandparent, had to besupported at home.42 The negative association between labor force participationof women outside the home and fertility is well established for contemporarysocieties (although the direction of causality is not); to establish such a rela-tionship for parts of nineteenth century Europe would require analysis of datathat include information on occupation of all members of households.

Urbanization

Scattered evidence suggests that urbanization, as distinct from industrialization,acted as a fertility depressant, or at least failed to stimulate fertility at a timewhen the development of industry per se may well have exerted such stimulus.The areas of highest fertility in England were those with more dispersed pop-ulations.43 In the great commercial centers of Europe, such as Paris and London,fertility never rose appreciably from preindustrial levels, in contrast to trendsin industrial regions such as the Ruhr.44 In Flanders fertility was lower in thecity of Ghent than in the rural surrounding areas.45 We can tentatively notethat as the later stages of industrialization increased concentrations of people,fertility fell relative to its higher levels in rural industrial areas.

Nancy Birdsall 121

Conclusion

It is true, as Smith shows, that the household formation rule and the peculiarinstitution of agrarian service in pretransitional Errgland help explain the laggedpattern of fertility as it related to changing wages over several centuries. Butby the late eighteenth century an economic explanation, taking into accountthe new force of industrialization, can clarify why, for a long period in England,fertility and real wages literally went separate ways (fertility up, real wagesdown); and why, in nineteenth century England (a period for which the questionof whether workers' living standards rose or fell is still hotly debated), fertilitygradually fell. An economic explanation relies on two critical changes in thelabor market that occurred in sequence as the process of industrialization tookhold: first, the rapidly increasing opportunities for nonagricultural work; andsecond, the gradually increasing importance of skills (and thus education) inthe work force. Initially, fertility.rose with "proletarianization" and the shiftto nonagricultural wage labor. But with child labor laws and the rising im-portance of education, families wishing to preserve their wealth had to raisemore costly children, and the secular fertility decline we associate with theEuropean demographic transition took hold.

Notes

This article represents the views of the author, Haven: Yale University Press, 1966), pp.not those of the World Bank. The author is 19-20.grateful for the comments of an anonymous 5 Mortality also declined, but especiallyreferee and the comments of Richard Smith on for the latter half of the eighteenth century andan earlier draft. early nineteenth century there is evidence that

I "Fertility, economy, and household "a rising birth rate was the major cause of theformation in England over three centuries," growth of the English population"; J. T.Population and Development Review 7, no. 4 Krause, "Changes in English fertility and(December 1981): 595-622. mortality, 1781-1850, "Economic History Re-

2 John C. Caldwell, "Toward a restate- view, 2nd series I1, no. I (August 1958): 70.ment of demographic transition theory," Pop- 6 Ron Lesthaeghe, "On the social con-ulation and Development Review 2, nos. 3-4 trol of human reproduction," Population and(September-December 1976): 321-366; Cald- Development Review 6, no. 4 (Decemberwell, "A theory of fertility: From high plateau 1980): 531.to destabilization," Population and Develop- 7 E. A. Wrigley and R. S. Schofield,ment Review 4, no. 4 (December 1978): with contributions by Ronald Lee and Jim553-577; and Caldwell, "The mechanisms of Oeppen, The Population History of Englanddemographic change in historical perspec- 1541-1871, A Reconstruction (Cambridge,tive," Population Studies 35, no. 1 (1981): Mass.: Harvard University Press, 1981), p.5-27. 419 and pp. 429-430. They also note that

3 Charles Tilly (ed.), Historical Studies wages may have still been rising up to 1781,of Changing Fertility (Princeton: Princeton if rates in the northern regions are taken intoUniversity Press, 1978). account (p. 432).

4 Simon Kuznets, Modern Economic 8 See Franklin F. Mendels, "Proto-Growth: Rate, Structure and Spread (New industrialization: The first phase of the indus-

122 Fertility and Economic Change in Eighteenth and Nineteenth Century Europe

trialization process," Journal of Economic Smith, cited in note 1, p. 613). Even whereHistory 32 (1972); and P. Deprez, "The de- rural industrialization did not occur, there ismographic development of Flanders in the evidence that fertility rose with changes in landeighteenth century," in Population in History, inheritance pattemns or in availability of landed. D. V. Glass and D. E. C. Eversley (Lon- (see notes 11 and 16).don, 1965). 19 Mendels, cited in note 8.

9 J. P. Chambers, "Enclosure and labor 20 Frank Lorimer, Culture and Humansupply in the industfial revolution," Economic Fertility (Paris: UNESCO, 1954), p. 210, re-History Review 5 (1953): 324 and 339. ports a correlation of .489, plus or minus .116.

10 See David Levine, Family Formation 21 Levine, cited in note 10, pp. 65-67.in an Age of Nascent Capitalism (New York:Academic Press, 1977), pp. 58-87. He also 22 Charles Tilly, "Demographic originsshows how demographic trends in Colyton can of the European proletariat," University ofbe related to the rise and fall of rural industry Michigan, Center for Research on Social Or-(pp. 103-115). ganization Working Paper no. 27, December

1979, pp. 60-64. He cites studies by Andorka,11 Richard A. Easterlin, George Alter, Gaunt, Levine (see note 10), and Wrigley.

and Gretchen A. Condran, "Farm and farmrfamilies in old and new areas: The northern 23 Michael Anderson, "Family, house-states in 1860," paper presented at MSSB hold and the industrial revolution," in TheConference on Historical Demography, Wil- American Family in Social-Historical Per-liamstown, July 1974. spective, ed. Michael Gordon (New York: St.

12 Bernard H. Slicher van Bath. "Histor- Martin's Press, 1973), p. 66. This is not to say

ical demography and the social and economic that parish support, as suggested by Smith,development of the Netherlands," Daedalus was not a substitute for family support in old

(Spfing 1968): 604-621. age, but only that family support must have(Spring 1968): 604-621. been preferable.

13 William Petersen, "The demographic 24 For a formalization of this idea, see R.transition in the Netherlands," American So- J Willis "A new approach to the economicciological Review 25, no. 3 (1960): 334-347, theory of fertility behavior," Journal of Po-cited by W. D. Borrie, The Growth and litical Economy 81, no. 2, part 2 (March/AprilControl of World Population (London, 1970), 1973): S14-S64. This is a central tenet of thep. 80. "new home economics," which explains fer-

14 Paul Demeny, "Early fertility decline tility as the result of utility maximization byin Austria-Hungary: A lesson in demographic parents constrained by a household productiontransition," Daedalus (Spring 1968). function and a full-income constraint, both in-

15 H. J. Habakkuk, "Family structure and cluding time of parents.economic change in 19th century Europe," 25 For an example using United StatesJournal of Economic History 15 (1955): 1-12. data, see Glen G. Cain and Adriana Weinin-

16 Slicher van Bath, cited in note 12, ger, "Economic determinants of fertility: Re-p. 618. sults from cross-sectional aggregate data,"

17 Habakkuk, cited in note 15, p. 10. Demography 10, no. 2 (1973): 205-223.18 Phyllis Deane and W. A. Cole, British 26 A. J. Jaffe and K. Azumi, "The birth

Economic Growth 1688-1959: Trends and rate and cottage industries in underdevelopedStructure (Cambridge, 1962), cite Lancashire countries," Economic Development and Cul-as an example of this phenomenon (p. 116). tural Change 4 (October 1960): 52-63.It is worth noting that the argument that fertility 27 William Ogle, "On marriage-rates androse with new employment opportunities does marriage-ages, with special reference to thenot rely on the "ratchet" effect (that an growth of population," Journal of the Royaloversupply of labor, combined with a slump Statistical Society (1890), cited by H. J.in wages, led to more labor effort and exag- Habakkuk, Population Growth and Economicgerated levels of reproduction), nor even on Development Since 1750 (Leicester Universitya change in the relations of production (see Press, 1971), p. 39. Aggregate data on changes

Nancy Birdsall 123

in the distribution of the labor force among (April 1961): 397-416, cited by Branson (seesectors are inadequate for England before the note 32).,census of 1851, and after 1851 still tend to be 35 E. A. Wrigley, Population and Historyinconsistent. Households were sometimes (New York: McGraw-Hill, 1969), Table 5.13,lumped together, so that disaggregation ac- p. 186.cording to female and child employment isimpossible (Deane and Cole, cited in note 18, 36 J. A. Banks, Prosperity and Parent-

138- 140). hood: A Study of Family Planning Among thePP. Victorian Middle Classes (London, 1954),

28 Ogle, cited in note 27. chapter 11.

29 For a theory explaining why the pos- 37 Banks, cited in note 36, p. 193.sibility of outmigration could encourage fer- 38 Deane and Cole, cited in note 18, Tabletility of agricultural households in developing 30, p. 142: "almost a doubling of the numbersand modernizing societies, see Oded Stark, in the trade and commerce group between 1861"The asset demand for children during agni- and 189 1.cultural modemnization," Population and De-velopment Review 7, no. 4 (December 1981): 39 With widespread education, parents671-675. who do not send their children to school must

expect their children's place in the income30 Deane and Cole, cited in note 18, ditibto tob.lwr

p. 121. distrbuton to be lower.

31 Caldwell's early statement is in "To- 40 Wrigley, cited in note 35, p. 186.ward a restatement of demographic transition 41 Habakkuk, cited in note 27, p. 44. Seetheory," cited in note 2. The later emphasis also Krause, cited in note 5, pp. 67-68.on the change in relations of production is in 42 For a careful consideration of the cost"The mechanisms of demographic change in of children and how the shift to earnings-re-historical perspective," cited in note 2. ducing and time-intensive children with eco-

32 William H. Branson, "Social legisla- nomic development raises their costs relativetion and the birth rate in nineteenth century to other commodities, see Peter H. Lindert,Britain," Western Economic Journal (March "Child costs and economic development," in1968): 134-144; see pp. 139-140. Population and Economic Change in Devel-

33 Branson, cited in note 32, p. 142. oping Countries, ed. Richard A. Easterlin33 Banso, cied n noe 32 p. 42. (Chicago: University of Chicago Press, 1980),

34 There is considerable controversy about pp. 5-69. Lindert discusses how his relativeactual trends in living standards in nineteenth cost concept can explain the historical rise andcentury England. For the view that incomes then decline in fertility (p. 62).were rising, based on analysis of aggregate 43 Deane and Cole, cited in note 18,income statistics, wage-price data, and con- p, 116sumption data, see R. M. Hartwell, "The ris-ing standard of living in England, 1800-1850," 44 Wrigley, cited in note 35.Economic History Review, sec. 2, vol. 13 45 Deprez, cited in note 8, p. 627.

No. 269. James R. Follain, Jr., Gill-Chin Lim, and Bertrand Renaud, "HousingCrowding in Developing Countries and Willingness to Pay forAdditional Space: The Case of Korea," Journal of Development Economics

No. 270. Bela Balassa, "Policy Responses to External Shocks in Sub-SaharanAfrican Countries," Journal of Policy Modeling

No. 271. Jaime de Melo and Sherman Robinson, "Trade Adjustment Policiesand Income Distribution in Three Archetype Developing Economies,"Journal of Development Economics

No. 272. J. B. Knight and R. H. Sabot, "The Role of the Firm in WageDetermination: An African Case Study," Oxford Economic Papers

No. 273. William G. Tyler, "The Anti-Export Bias in Commercial Policies andExport Performance: Some Evidence from Recent Brazilian Experi-ence," Weltwirtsclhaftliches Archiv

No. 274. Ron Duncan and Ernst Lutz, "Penetration of Industrial CountryMarkets by Agricultural Products from Developing Countries," WorldDevelopment

No. 275. Malcolm D. Bale, "Food Prospects in the Developing Countries: AQualified Optimistic View," The American Economic Review (withRonald C. Duncan) and "World Agricultural Trade and Food Security:Emerging Patterns and Policy Directions," Wisconsin International LawJournal (with V. Roy Southworth)

No. 276. Sweder van Wijnbergen, "Interest Rate Management in LDCs,"Journal of Monetary Economics

No. 277. Oli Havrylyshyn and Iradj Alikhani, "Is There Cause for ExportOptimism? An Inquiry into the Existence of a Second Generation ofSuccessful Exporters," Weltwirtschaftliches Archiv

No. 278. Oli Havrylyshyn and Martin Wolf, "Recent Trends in Trade amongDeveloping Countries," European Economic Review

Issues of the World Bank Reprint Series are available free of charge fromthe address on the bottom of the back cover.

THE WORLD BANK

Headquarters:1818 H Street, N.W.,Washington, D.C. 20433, U.S.A.

Telephone: (202) 477-1234Telex: WUI 64145 WORLDBANK

RCA 248423 WORLDBKCable address: INTBAFRAD

WASHINGTONDC

European Office:66, avenue d'1ena75116 Paris, France

Telephone: (1) 723-54.21Telex: 842-620628

Tokyo Office:Kokusai Building1-1, Marunouchi 3-chomeChiyoda-ku, Tokyo 100, Japan

Telephone: (03) 214-5001Telex: 781-26838

The full range of World Bank publications, both free and for sale, isdescribed in the Catalog of Publications; the continuing research programis outlined in Abstracts of Current Studies. The most recent edition ofeach is available without charge from:

PUBLICATIONS SALES UNIT

THE WORLD BANK

1818 H STREET, N W.

WASHINGTON, D.C. 20433

U.S.A. ISSN 0253-2131