Industrialization and Fertility in the Nineteenth Century: Evidence from South Carolina

MARIANNE H. WANAMAKER

doi:10.1017/S0022050711002476

Industrialization and Fertility in the Nineteenth Century: Evidence from South Carolina

Published online by Cambridge University Press: 12 March 2012

MARIANNE H. WANAMAKER

Show author details

MARIANNE H. WANAMAKER*: Affiliation:
Assistant Professor, Department of Economics, College of Business Administration, The University of Tennessee, 524 Stokely Management Center, Knoxville, TN 37996. E-mail: wanamaker@utk.edu.

Article contents

Abstract
EMPIRICAL STRATEGY
TOWNSHIP DATA
DIFFERENCE-IN-DIFFERENCE RESULTS
CONTROLLING FOR MIGRATION
MIGRATION RESULTS
MIGRATION, OCCUPATIONAL CHOICE, AND CHILD MORTALITY
EVALUATING POTENTIAL MECHANISMS
CONCLUSION
Footnotes
References

Rights & Permissions

Abstract

Economists frequently hypothesize that industrialization contributed to the United States’ nineteenth-century fertility decline. I exploit the circumstances surrounding industrialization in South Carolina between 1881 and 1900 to show that the establishment of textile mills coincided with a 6–10 percent fertility reduction. Migrating households are responsible for most of the observed decline. Higher rates of textile employment and child mortality for migrants can explain part of the result, and I conjecture that an increase in child-raising costs induced by the separation of migrant households from their extended families may explain the remaining gap in migrant-native fertility.

Information

Type: ARTICLES
Information: The Journal of Economic History , Volume 72 , Issue 1 , 12 March 2012, pp. 168 - 196
DOI: https://doi.org/10.1017/S0022050711002476 [Opens in a new window]
Copyright: Copyright © The Economic History Association 2012

By the dawn of the twentieth century, fertility rates in the United States had undergone a century of steady decline. In 1800 white American females could expect to bear 7.0 children on average; by 1900 this number was 3.6.Footnote ¹ The factors behind the nineteenth-century decline have been the subject of a lengthy literature highlighting the importance of intergenerational bequests, the economic value of children, and the cultural context for American family formation.

Yasukichi Yasuba initiated the literature on the importance of land availability to the bequest process, noting that those locations with lower land prices experienced higher fertility rates and vice versa.Footnote ² William Sundstrom and Paul David and Susan B. Carter, Roger L. Ransom, and Richard Sutch present alternatives to the land availability hypothesis for explaining the nineteenth-century decline emphasizing the role of economic modernization.Footnote ³ The empirical evidence seems to indicate a role for bequest and land availability explanations of fertility decline early in the nineteenth century while economic modernization variables (levels of industrialization, urbanization, child labor laws, average wages, literacy rates, and education levels) hold better explanatory power for the postbellum era.Footnote ⁴ But despite these conclusions, the impact of these economic modernization correlates on fertility rates has yet to be measured in anything more than a cross-section, observational framework.

This article is motivated by the concern that previous estimates of the relationship between industrialization and fertility were measured with simultaneity bias. An empirical observation that industrialization and fertility are negatively correlated is not informative about causation. Falling fertility may have spurred industrialization rather than vice versa, or external factors may have contributed to both simultaneously.

There are a number of mechanisms by which industrialization may have altered a household's fertility outcome. First, several models of economic growth and fertility decline highlight the role of human capital in increasing the incentives of households to invest in child quality over quantity, thereby reducing the number of children born.Footnote ⁵ Second, industrialization may have induced a rise in the implicit costs of raising children. In particular, industries with high rates of female employment increased the opportunity cost of female time. Under the assumption that the child production process is female time-intensive, this would have reduced the incentive to bear children.Footnote ⁶ Third, the movement away from agricultural and at-home production to centralized production, in addition to more restrictive child labor laws, may have reduced the economic return to children, again lowering parental demand and fertility rates.Footnote ⁷ Fourth, industrialization was associated with increased urbanization and the crowding that occurred may have increased the explicit costs of raising children through higher housing and food costs without an associated increase in the benefit.Footnote ⁸ Finally, the developing economy in the United States witnessed decreases in child mortality rates, especially after 1880.Footnote ⁹

I exploit the circumstances surrounding industrialization in South Carolina between 1880 and 1900 to identify a plausibly exogenous change in the level of industrialization. I argue that the emerging geography of South Carolina's textile industry was driven by water power concerns and was unlikely to have been correlated with fertility ex ante. Using a sample of rural townships in South Carolina, including locations that did and did not establish textile mills over this period, I estimate a reduction in fertility of between 6 and 10 percent following textile mill establishment.

These data indicate that an increase in urbanization and the costs of raising children, including a rising opportunity cost of female time, can explain part of the observed reduction in fertility in South Carolina industrial locations. At the same time, I find no support for mechanisms relying on human capital considerations and increases in the adult/child wage differential. Differential infant mortality among industrial workers may also have been a factor in fertility reduction.

EMPIRICAL STRATEGY

This article measures the impact of industrialization on fertility rates using data from rural areas of South Carolina between 1880 and 1900. In 1880 South Carolina was home to 14 textile mills and 84,424 spindles. By 1900 the state housed 93 mills and 1,693,649 spindles, an increase of almost 2,000 percent.Footnote ¹⁰

There are several unique aspects of the South Carolina industrialization process that make it an ideal laboratory in which to test hypotheses about industrialization and fertility decline. First, prior to 1910 or so, textile mills provided the overwhelming majority of industrial employment in the state. In 1900, 30 percent of males and 57 percent of females employed in manufacturing were textile workers. No other industry came close to matching the textile share of employment, and the vast majority of the runners-up involved manufacturing in a decentralized setting.Footnote ¹¹ As a result, South Carolina industrialization in general can be proxied by the textile industry in particular.

Second, the historical record indicates that the Piedmont region, where textile manufacturing focused after 1880, was homogenous in terms of its economic activity, socioeconomic characteristics, and climate prior to industrialization.Footnote ¹² Further, the particular locations of textile mills in rural areas were primarily determined by concerns about water power rather than demographic characteristics of the local population and, in particular, their fertility. In many ways, “if you build it they will come” applied to the South Carolina rural population in this period. Rural households were desperate to leave their farming professions as cotton and tobacco prices plummeted in the 1890s. Mill owners could have chosen virtually any rural location and they would have found a local labor force eager to become mill workers. The uniformity of the Piedmont area gave mill proprietors the opportunity to select locations based on other criteria.

In the early part of the industrialization period, particularly prior to 1890, mills located on the banks of rapidly flowing rivers. These location decisions were driven both by concerns about water power and by the need for natural humidity to support the manufacturing process. Any river location could supply humidity, so the availability of quickly flowing water capable of driving a water wheel dictated location in this early period. The demographics of the local population would have been, if anything, secondary.

The first steam-powered mill was built in South Carolina in 1881 and, after 1890 steam power began to dominate as a power source. “Cotton mills among the cotton fields” became possible, and mill proprietors were freed somewhat from their reliance on fast-flowing water. At first, steam and water were used in combination to power textile mills, but by the second half of the 1890s, most mills ran on steam alone.Footnote ¹³ Proximity to water was still desirable for waste removal and natural humidity.Footnote ¹⁴

After 1900 the pre-1900 drivers of location decisions were waning. Labor market considerations were becoming more important as South Carolina mills began producing a higher quality cloth that required skilled labor, and electric power was freeing firms from prior geographic constraints.Footnote ¹⁵ As a result, 1900 was chosen as the endpoint for this analysis.

Even given the site selection process described above, it is likely that townships with a water supply suitable for water-powered manufacturing were different from other locations ex ante.Footnote ¹⁶ As a result, I use a difference-in-difference estimator to account for any preexisting differences in fertility in estimating the impact of mill establishment on fertility. Using this methodology, the estimated impact of industrialization on fertility is 6 to 10 percent.

In order to determine the importance of migration in this result, I incorporate household-level data and conclude that declining fertility can be entirely attributed to migrating households. Given this result, I evaluate the potential mechanisms at play and find that higher rates of textile employment and child mortality characterized the migrant experience. In addition, I hypothesize that the loss of an extended family raised the implicit costs of child rearing for these households. I further argue that alternative explanations for the observed relationship between industrialization and fertility are not supported for this particular historical episode.

TOWNSHIP DATA

Demographic and fertility data comes from the manuscript returns of the United States Census. The U.S. Census was taken at the individual level. Individuals were grouped into households which were, in turn, grouped into townships. Townships were grouped into counties and then into states. Using data from the online genealogy tool Ancestry.com, I assemble age structure and marital status data for townships in both 1880 and 1900 to measure a township-level impact of industrialization on fertility.Footnote ¹⁷ I focus on marital fertility rates as almost all fertility occurred within marriage and declines in marital fertility, rather than declines in marriage rates, contributed most to the U.S. decline in this period.Footnote ¹⁸ In each year (t = 1880, 1900), I calculate F5 fertility rates:

$\[ \begin{array}{c} {{{\# ^{}}{\rm{o}}{{\rm{f}}^{}}{\rm{children}} gt {5^{}}{\rm{year}}{{\rm{s}}^{}}{\rm{o}}{{\rm{f}}^{}}{\rm{ag}}{{\rm{e}}^{}}{\rm{i}}{{\rm{n}}^{}}{\rm{townshi}}{{\rm{p}}^{}}{j^{}}{\rm{i}}{{\rm{n}}^{}}{\rm{yea}}{{\rm{r}}^{}}t} \over {{\rm{Married,}}{}^{}{\rm{divorced,}}{}^{}{\rm{and}}{}^{}{\rm{widowe}}{{\rm{d}}^{}}{\rm{female}}{{\rm{s}}^{}}{\rm{age}}{{\rm{d}}^{}}{\rm{1}}{{\rm{8}}^{}}{\rm{t}}{{\rm{o}}^{}}{\rm{4}}{{\rm{2}}^{}}{\rm{i}}{{\rm{n}}^{}}{\rm{townshi}}{{\rm{p}}^{}}{j^{}}{\rm{i}}{{\rm{n}}^{}}{\rm{yea}}{{\rm{r}}^{}}t}} \end{array} \$

The F5 fertility rate calculated in this manner will be sensitive to child mortality, a topic I address later in the article.Footnote ¹⁹ I also perform sensitivity analysis in the next section using F3 fertility rates and eliminating widowed and divorced women from the denominator. A comparison of F5 fertility rates for the townships included in the main results of this article, all of South Carolina, all of the South, and the United States appear in Figure 1. Figure 1 also shows a comparison of F5 fertility rates for industrial and nonindustrial townships in South Carolina in both 1880 and 1900.

Industrialization data comes from Davison's Textile Blue Book, a directory of textile mills in the United States. The directory was printed biannually beginning in 1888. I use the 1902/03 edition to generate a timeline of mill establishment in South Carolina prior to 1900.Footnote ²⁰Davison's gives an establishment date for each indexed mill, and Figure 2 shows a map of South Carolina with markers for textile mills that were established between 1880 and 1900. From this data, I generate two industrialization indicators: I ₁₈₉₅ = 1 if a mill was established in township j by 1895 and I ₁₉₀₀ = 1 if a mill was established between 1896 and 1900. The F5 fertility measure described above, when measured in 1900, incorporates all fertility over a 5-year window from 1896 to 1900. Given this time frame, it is important to separately identify townships with mills established prior to the beginning of the fertility measure (I ₁₈₉₅) whose fertility response may conceivably be fully captured by the F5 measure and those townships with mills established within the 1896–1900 window (I ₁₉₀₀) and whose fertility response may not be fully accounted for using F5. I ₁₈₉₅ includes townships exposed to industrialization for the entirety of the F5 window while I ₁₉₀₀ includes townships with only partial exposure to a textile mill during that window.

FIGURE 1

F5 FERTILITY RATES FOR INDUSTRIALIZED AND NONINDUSTRIALIZED SOUTH CAROLINA TOWNSHIPS (TOP PANEL) AND FOR ALL SOUTH CAROLINA, THE SOUTH, UNITED STATES (BOTTOM PANEL)

Notes: Top panel includes only townships in the baseline sample, which includes non-coastal townships with town population less than 2,500 people from Table 3, Column 3. Bottom panel, All South Carolina includes all S.C. townships while Paper Sample is the baseline sample from Table 3. “All South” includes all former Confederate states.Source: Top panel: See the text. Bottom panel: U.S. Bureau of the Census, Tenth Census and Twelfth Census.

FIGURE 2

SOUTH CAROLINA MILL ESTABLISHMENT: 1881–1900

Source: Davison Publishing Company, Davison's Textile Blue Book.

TABLE 1

TOWNSHIP DATA VARIABLE SUMMARY

Notes:

Sample includes all non-coastal townships with no textile mill by 1880 and 1900 town population of fewer than 2,500. Corresponds to results in Table 2.

Sources:

See the text.

I incorporate into the analysis several additional explanatory variables: a binary variable for whether the township was a county seat in year, a proxy for urbanization measuring the inhabitants of a census-defined town divided by all township inhabitants, the nonwhite population percentage, the male to female sex ratio for individuals aged 18 to 42, and the marriage rate among females aged 18 to 42.

DIFFERENCE-IN-DIFFERENCE RESULTS

Table 1 summarizes these variables for both 1880 and 1900 for the sample townships, tabulated by the period in which the first mill was established. By 1900 the average fertility rate in industrialized townships was well below the sample average, and a simple OLS estimate of the effect of industrialization (measured by I ₁₈₉₅ and I ₁₉₀₀) on fertility is negative and significant (not shown). But, as is evident from Table 1, townships that industrialized between 1881 and 1900 also had somewhat lower fertility rates ex ante, as measured by F5₁₈₈₀. While a “falsification test” which repeats the OLS exercise for F5₁₈₈₀ identifies no significant impact on ex ante fertility, the estimated coefficients are large enough to motivate alternative estimators.

To take into account the difference in 1880 fertility levels, I evaluate the impact of mill arrival on fertility rates using a difference-in-difference approach with 1880 and 1900 data.Footnote ²¹ A difference-in-difference estimator compares the change in fertility between 1880 and 1900 for industrializing townships to the change in fertility between 1880 and 1900 for townships where no textile mill was established. The underlying assumption is that, in the absence of industrialization, the rate of change for these two groups would have been the same and any difference in these rates of change is due to the industrialization process itself.

To examine the assumption of similar trends in fertility in both groups, I calculate an annual fertility rate retrospectively from 1881 to 1900 for each township in my sample using the 1900 Census returns.Footnote ²² I regress these annual fertility rates from 1881 to 1900 for townships that do not contain a textile mill in that year on a township fixed effect and a constant and calculate the residuals.Footnote ²³ In Figure 3, I plot these smoothed residuals for two types of townships: those that do not industrialize in the 20-year window (“Not industrialized by 1900”) and those that industrialize at some point before 1900 but have not done so by year t (“Industrialized, but not yet”). The strikingly similar trend lines indicate the fertility rates of industrial townships prior to the arrival of textile mills are declining at a rate similar to townships that do not industrialize by 1900. This finding is consistent with the main assumption behind a difference-in-difference model.

FIGURE 3

RESIDUAL FERTILITY IN NONINDUSTRIAL TOWNSHIPS, BY FUTURE INDUSTRIAL STATUS (1880–1899)

Notes: Plotted residuals from a regression of annual fertility rates from 1881 to 1899 in townships that do not contain a textile mill in that year on a township fixed effect and a constant. Mean residuals are plotted for two types of townships: those that do not industrialize in the 20-year window (“Not industrialized by 1900”) and those that industrialize at some point before 1900, but have not done so by year t (“Industrialized, but not yet”). See “Difference-in-Difference Results” for more details.Source: See the text.

The estimated impact of industrialization on fertility is represented by δ1 and δ2 in equation 1

(1)

where X_jt is a vector of township-year characteristics. c is a constant, λ_j is a township-specific fixed effect, 1(t = 1900) is an indicator for the second year of data, θ is the 1900-specific year effect, and ε_jt is a random error term. The estimation is performed over t =1880, 1900. The first-difference estimating equation is

(2)

$\[ \begin{array}{c} \Delta F{5_j} = \delta 1{I_{j,1895}} + \delta 2{I_{j,1900}} + \beta \Delta {X_{jt}} + \theta + \mathop {{\varepsilon _j}}\limits_{}^\~ \end{array} \$

where .

Variable means and standard deviations are located in Table 1. Table 2 contains the estimates from equation 2 under a variety of specifications. As anticipated, a more significant impact of I ₁₈₉₅ on fertility rates than I ₁₉₀₀, both statistically and economically, is observed in Table 2.

In column 1 of Table 2, I include all townships in the data set, excepting only those that already contained a textile mill in 1880 (1.8 percent of all townships) and incorporating only the political designation of the township as a county seat in X_jt. In column 2, I add two indicators: one for whether the township is located in a county on South Carolina's Atlantic coast and one for whether the size of the town population in 1900 exceeds 2,500. Coastal counties were not as heavily dependent on agriculture prior to the arrival of the mills, did not participate in the textile industry, and were arguably different from the Piedmont counties in other ways as well. The town population measure is a recognition that, in larger towns, industrialization came in forms other than textile mills, and the industrialization proxies used herein are inappropriate.Footnote ²⁴ For both columns 1 and 2, industrialization, as proxied by textile mills, was accompanied by more rapid fertility decline between 1880 and 1900. For a mill built between 1881 and 1895, the estimated additional reduction in fertility is between 6 and 8 percent.

Given these concerns about the adequacy of I ₁₈₉₅ and I ₁₉₀₀ to measure industrialization in coastal counties and in townships with larger town populations, I trim the sample in columns 3–8 to eliminate these counties. This is the baseline specification. The results in column 3 again indicate that textile mill locations exhibited an acceleration in fertility decline between 1880 and 1900 relative to nonindustrial townships. Industrialization prior to 1895 is associated with a differential reduction in fertility of 6.9 percent in this trimmed sample, controlling only for county seat status. This 6.9 percent estimate represents the fertility reduction for the average township. In column 6, I weight the estimate by township population in 1900, which leads to a slightly larger reduction of 7.4 percent as the estimated impact of industrialization.

These results clearly indicate that the presence of a textile mill in a township is correlated with lower fertility rates in 1900. What changes in observable characteristics of textile locations might help explain this correlation? I add controls for the change in the racial composition and density of the township's population in column 4. Industrialization may have induced the small-scale urbanization measured by “town” residents as a percentage of total population and may have altered the ratio of the nonwhite to white population as textile mills employed white workers almost exclusively. Incorporating those two covariates, the estimated impact of industrialization prior to 1895 is reduced by approximately 11 percent, indicating that changes in population density and racial composition accounted for a small proportion of the observed fertility reduction in textile locations.

TABLE 2

IMPACT OF INDUSTRIALIZATION ON FERTILITY DIFFERENCE-IN-DIFFERENCE ESTIMATION RESULTS FOR TOWNSHIP SAMPLE (equation 2)

∗ indicates statistical significance at the 10 percent level.

∗∗ indicates statistical significance at the 5 percent level.

∗∗∗ indicates statistical significance at the 1 percent level.

^aIn Column 7, I ₁₈₉₅ becomes I ₁₈₉₇ and I ₁₉₀₀ represents only those mills built between 1898 and 1900.

Notes:

Point estimates of coefficients. Standard errors are in parentheses.

Sources:

See the text.

The F5 fertility measure used in these regressions is a marital fertility rate, but it may still be the case that changes in the sex ratio or marriage rates altered fertility within marriage if, for instance, a changing marriage rate impacted the age at which women marry and their fertility window within marriage. I add measures of the change in both the sex ratio and the marriage rate for residents aged 18 to 42 in column 5. The estimated mill impact is reduced relative to column 3 by approximately 30 percent, indicating that changes in the sex ratio and marriage rates account for roughly one-third of the overall decline in fertility observed in textile locations.

In columns 7 and 8 of Table 2, I undertake two robustness checks. To determine the sensitivity of the results to the fertility measure used, I repeat the estimation on F3 measures of fertility. There is little change in coefficients as reported in column 7. In column 8, I limit the denominator to married females (excluding widowed and divorced females). Again, there is little change in the results.

The results from columns 3–6 of Table 2 show that textile mill establishment significantly reduced fertility rates in South Carolina between 1880 and 1900 and that changes in population density, racial composition, sex ratios, and marriage rates of these townships can explain less than half of the fertility reduction. Mill establishment prior to 1895 is associated with a 6.9 percent reduction in fertility rates in a specification with no demographic covariates and 4.3 percent after controlling for these changing aspects of the township population.

A remaining concern is whether differential migration not captured by changes in the racial composition, sex ratio, and marriage rate, is responsible for the observed decline. This concern motivates construction of a household-level data set in the next section.

CONTROLLING FOR MIGRATION

The advantage of township data is that township boundaries are the same throughout the 20-year industrialization period, allowing us to measure a “pre-industrialization” and “post-industrialization” fertility metric and generate a difference-in-difference estimate of the impact of textile mill establishment on fertility. But townships are not composed of an immobile set of individuals, and the fertility pattern documented above could simply be a result of selective migration without changing the fertility outcomes of households per se. In this section, I generate a new sample of households including a measure of migration.

In order to identify migrants, the location of the household at two points in time is needed. The 1900 Census contains information on an individual's state of birth, but not their county or township. As the vast majority of South Carolina residents in 1900 were born in South Carolina, this variable has limited power to identify migrants.

I remedy this problem using the genealogy tool Ancestry.com to link individuals between the 1900 Census and the 1880 Census and use their township location in both to infer migration. The 100 percent sample of the 1880 U.S. Census from the North American Population Project has information on 284,412 South Carolina resident males aged 0 to 20 in 1880.Footnote ²⁵ Of those, 34,841 are uniquely matched to the 1900 U.S. Census using the search function of Ancestry.com and the individual's full name, race, and age (1 year) to make a match.Footnote ²⁶ Of those, 12,999 are married with their spouse present in the household and reside in South Carolina townships with fewer than 2,500 town residents in 1900 and outside of the coastal counties. This subsample of married South Carolina males, aged 19 to 41 in 1900 with a spouse present in the household, is the basis for analysis. The 1900 Census return as transcribed by Ancestry.com contains a full list of household members, their name, relationship to the head of household, age, race, marital status, marital duration, and birthdate. The location of the male head at two points in time, 1880 and 1900, gives a measure of household migration.

A caveat is in order. Because the objects of interest are young, fertile households in 1900, I can only link the male head because he is the only member of the household who would have been alive and with the same last name in 1880. Females would have been enumerated under their maiden name in 1880. That presents the possibility that even though the male head did not migrate between 1880 and 1900, his wife did (or vice versa), and I will not capture that in the subsequent analysis.Footnote ²⁷

MIGRATION RESULTS

To measure the impact of migration on the results previously reported, I estimate an equation of the following form

(3)

$\[ \begin{array}{c} \bar F5_{i,j}^{HH} = c + \delta 1{I_{j,1895}} + \delta 2{I_{j,1900}} + \theta {Y_i} + \beta {X_j} + \alpha MOVE{R_i} + {\varepsilon _{i,j}} \end{array} \$

where is the (discrete) number of children less than age 5 in household i in township j, analogous to the F5 fertility rate from the township sample. The vector of township characteristics, X_j, was defined previously. Y_i contains the matched male's age, age squared, wife's age, age squared, and an indicator for whether the matched male is black. MOVER_i is an indicator for whether household i changed township locations between 1880 and 1900. The coefficient α measures differential fertility for township movers. Table 3 gives sample means for variables contained in the household sample. As a consistency check, from Table 1 and from Table 3 are both equal to 1.22.

An ordered probit estimator is used for equation 3 because values for are discrete.Footnote ²⁸ Coefficients and standard errors for δ1, δ2, and α are reported in Table 4, but the coefficients no longer have a marginal effect interpretation. The marginal effect implied by δ1 is listed lower in the table. In column 1 of Table 4, the estimate of the impact of industrialization on fertility controlling only for variables available in the township sample indicates a negative correlation between F5^HH and pre-1896 industrialization of about 3.7 percent in the household sample compared to 7.4 percent in the township data weighted by population.

TABLE 3

1900 VARIABLE SUMMARY FOR HOUSEHOLD DATA

Notes:

Corresponds to results in Tables 4, 6, and 7.

Source:

See the text.

After adding race and age covariates in column 2, the implied reduction in fertility corresponding to pre-1896 industrialization is 4.1 percent.Footnote ²⁹MOVER_i, the indicator for migration, is included in column 3 and the negative coefficient implies that migrants had lower fertility. Comparisons of coefficients for the the sample of movers in column 4 and nonmovers in column 5 shows that migrants who resided in townships industrialized by 1895 had 6.9 percent lower fertility than migrants to nonindustrial townships. On the other hand, nonmovers in those same townships exhibited 1.1 percent higher fertility, although the coefficient is not statistically significant. This is a striking result as it indicates that households who remained in textile locations between census enumerations exhibited no change in fertility relative to those households who remained in a non-textile location. The industrialization process was correlated with lower fertility, as earlier results have indicated, but migrating households were responsible for the negative correlation.

There are at least three potential explanations for this result. First, it is possible that migrants to textile townships were disproportionately from low-fertility native townships and brought these fertility patterns with them. But adding covariates for average fertility of the migrant's home township and population of the migrant's home township, either in 1880 (Table 4, column 6) or 1900 (column 7), does not reduce the estimated impact of industrialization.

TABLE 4

IMPACT OF INDUSTRIALIZATION ON FERTILITY ORDERED PROBIT RESULTS FOR HOUSEHOLD SAMPLE (equation 3)

∗ indicates statistical significance at the 10 percent level.

∗∗ indicates statistical significance at the 5 percent level.

∗∗∗ indicates statistical significance at the 1 percent level.

Notes:

Point estimates of coefficients. Standard errors are in parentheses.

Source:

See the text.

A second potential explanation is that movers to textile locations had lower initial fertility rates than their native township peers. Perhaps these low-fertility households were attracted to the labor markets in industrial locales and brought their low fertility with them. If so, we would expect migrants to townships where mills were built late in the 1881–1900 timeframe to exhibit low fertility as well, but that is not the case. In every specification of Table 4 involving mover households (columns 4, 6, and 7), the impact of being in a township industrialized by 1895 is large and negative while the impact of location in an industrial township where a new mill was added after 1896 is less negative and statistically insignificant. Migrants exhibited lower fertility than nonmigrants in textile locations, but only in locations where they had been exposed to a textile mill for five years or longer. Migrants to townships where mills were built between 1896 and 1900 did not have lower fertility rates than their peers. This points not to selective migration of households, but to a different fertility response to residence in a textile locale.

This gives rise to a third explanation. I document below that women in the households that migrated to textile townships disproportionately chose textile occupations and experienced higher rates of child mortality. In addition, the loss of proximity to extended family for migrants might have contributed to higher costs of raising children.Footnote ³⁰

MIGRATION, OCCUPATIONAL CHOICE, AND CHILD MORTALITY

Occupational choice may have contributed to the differing impact of industrialization on movers and stayers. Since Ancestry.com did not report occupations for the sample used in Table 4, I transcribed the occupation of the male and female household heads for the 2,342 households residing in townships in 1900 that were industrialized between 1881 and 1900 from the original census enumerations. The occupation shares in Table 5 show that 11 percent of male heads of migrant households were employed in textile jobs, compared with only 4 percent of stayers; 7 percent of migrant women compared to 3 percent of stayers were in textile jobs. Movers also had a higher rate of employment in nonfarm occupations in general.

Both nonagricultural employment categories were negatively correlated with fertility. When I estimate fertility for households with occupations listed in the industrial townships as a function only of migration in column 1 of Table 6, fertility is strongly negatively correlated with migrant status in industrial locales. After controlling for the occupations of the male and female household heads in the sample in column 2, households with males and females in the textile industry have lower fertility, while the relationship between fertility and migration is reduced by approximately 25 percent from −9.8 percent to −7.2 percent but remains significant.

TABLE 5

MALE AND FEMALE OCCUPATION, TABULATED BY MOVER/STAYER STATUS

Notes:

See the text. The “Other” category is comprised mostly of service occupations.

Sources:

See the text.

The impact of occupations was strong not only for white residents who were likely to experience textile employment, but also for black migrants who very rarely were textile workers. Column 3 of Table 6 limits the sample to black households and the results are largely unchanged. The impact of textile mills, it appears, extended beyond their own employees, perhaps because they altered the labor market for townships in general.

Occupational choice, then, can explain part of the difference in fertility between movers and stayers in textile locations, but a significant portion of the correlation remains unexplained. One further hypothesis is that migration influenced infant mortality which, in turn, affected fertility. Higher rates of infant and child mortality in industrial locations for migrating households in 1900 may be the cause of lower observed fertility in 1900 if replacement is not perfectly achieved.Footnote ³¹

TABLE 6

IMPACT OF MOVER/STAYER STATUS AND OCCUPATIONAL CHOICE ON HOUSEHOLD FERTILITY IN TOWNSHIPS INDUSTRIALIZED BY 1900 (equation 3)

∗ indicates statistical significance at the 10 percent level.

∗∗ indicates statistical significance at the 5 percent level.

∗∗∗ indicates statistical significance at the 1 percent level.

Notes:

Point estimates of coefficients. Standard errors are in parentheses.

Sources:

See the text.

The 1900 Census asked married women how many children they had borne and how many were surviving.Footnote ³² Using the household sample, it is straightforward to calculate a child survival rate as the ratio of the number of children surviving to the number ever born, and then to examine the relationship between that number, industrial residence, and migration.Footnote ³³

I estimate the relationship between a married female's reported child survival rate and the presence of a textile mill in the household's 1900 township location, conditional on other characteristics of the household. The estimating equation is

(4)

$\[ \begin{array}{c} {S_{ij}} = \tau {D_{ij}} + \alpha MOVE{R_i} + \beta {X_i} + \theta {Y_j} + {\varepsilon _{i,j}} \end{array} \$

where S_ij is the observed child survival rate for household i in township j in 1900. Unlike F5, the survival measure is not limited to the five years prior to the 1900 Census enumeration. Instead, the measure will reflect cumulative infant and child mortality over the entirety of a female's childbearing years and I ₁₈₉₅ and I ₁₉₀₀ are no longer relevant. Instead I construct a variable, D_ij, that represents the proportion of household i's marital duration for which township j housed a textile mill. The same set of control variables are used to estimate equation 4 as for equation 3, and the same sample restrictions apply.

Table 3 gives the mean and standard deviation of S_ij and D_ij. The results in column 1 of Table 7 indicate no discernible difference in infant and child mortality rates between residents of industrialized and nonindustrialized townships.Footnote ³⁴ These are small numbers; the coefficient in column 1 represents a reduction in survival rates of 1.0 percent.

Adding migration status to the regression (Table 7, column 2) does not affect the results. But in column 3, I add an interaction between mover status and the textile exposure measure, D_ij, and the estimates indicate that survival rates were significantly lower among movers to industrial locations than among their peers. A mover to a township with a textile mill present for the entirety of their marriage duration experienced a reduction in child survival rates of 3.7 percent (0.031/0.84) relative to nonmovers or movers to nonindustrial locales. The coefficients when occupations dummies are added to the specification in column 4 indicate that the occupational choice of migrants largely explains the adverse child survival rates. Female employment in a textile mill had a large negative correlation with child survival rates, while employment of the male had a more muted correlation. When the occupation dummies are added the coefficient on the interaction between mover and location in an industrial household turns positive and is statistically insignificant. For black residents (column 5), migration was similarly inconsequential for determining infant and child mortality after controlling for occupation.

TABLE 7

REGRESSION RESULTS FOR CHILD SURVIVAL RATES IN THE HOUSEHOLD SAMPLE (equation 4)

∗ indicates statistical significance at the 10 percent level.

∗∗ indicates statistical significance at the 5 percent level.

∗∗∗ indicates statistical significance at the 1 percent level.

Notes:

Point estimates of coefficients. Standard errors are in parentheses.

Sources:

See the text.

The results in Tables 6 and 7 indicate that occupational choice may explain part of the lower observed fertility rates among migrants in industrial locales. A lower child survival rate in migrant households is entirely attributable to occupational choice, lending futher support to this conclusion. If differential fertility among movers to industrial locations was driven by reductions in survival rates, it appears to have been the result of occupational choice. But a significant portion of the mover-stayer fertility differential remains unexplained, even after controlling for occupation in Table 6.

EVALUATING POTENTIAL MECHANISMS

The introduction to this article listed several ways in which economic growth may have affected fertility rates, and I evaluate those potential mechanisms in turn.

First, the available evidence indicates that increases in the costs of raising children contributed to reduced fertility in textile locations. Female occupation is a strong predictor of household fertility in the sample (Table 6), and low migrant fertility can be explained in part by higher rates of female employment in textile and service industries. Fertility is most strongly associated with textile employment where employment would have prevented active childrearing. But there is a significant impact among service workers as well. The arrival of textile mills increased demand for female-provided services such as cooking, laundering, growing small amounts of crop and livestock, and boarding. These occupations were dominated by African American females and this indirect effect can explain the reduction in fertility among this population.

In addition, the observed difference in fertility for migrants relative to natives is consistent with a different cost of children explanation: the costs of child care. For natives in industrialized locales, extended family could have served as a backstop for childcare. But for migrants, this option was not available. Childcare in the mills was sporadically supplied, and most families would have relied on household members for this service. The lack of extended family members to perform this task would have resulted in an increase in the cost of raising children for migrants relative to natives and may have been another cause of the observed fertility differential.Footnote ³⁵

The available data also suggest an increase in population density as an important factor in inducing lower fertility. In the township results of Table 3, controlling for the town population percentage of a township reduces the coefficient on the industrialization proxy in column 5 by 29 percent for townships industrialized after 1895 and by 18 percent for those industrialized before that date.

On the other hand, there is little evidence that child human capital or quality/quantity tradeoff concerns drove fertility reduction. Textile mills relied on low-skilled operatives, and there is no indication that textile manufacturing increased the return to human capital or incentivized parents to invest in child quality over quantity.Footnote ³⁶ A variant of the original quality/quantity hypothesis focuses on the potential impact of an increase in income on parental preference for quality over quantity. But this also has little support in the data as indicators for male occupational status (a proxy for household income) have a muted correlation with fertility in comparison to those for female occupational status (a proxy for both household income and the opportunity cost of female time) in Table 6.

Finally, a decrease in the labor opportunities of children is an improbable mechanism in this context. Available sources indicate that children represented a large proportion of workers in Southern mills, even when mills self-reported their employment numbers. Further, child labor legislation was not passed in South Carolina until 1903, well after the period examined in this article. If changes in the labor opportunities of children drove the industrialization results, it must be that mill employment, extensive as it was, was less well renumerated than their previous employment, generally as farm laborers. This is a questionable assumption, in part because children, with their smaller and more dextrous hands, were actually better suited for some mill tasks than were adults.

I conclude that increases in child mortality, the opportunity cost of female time, and in other costs of raising children, including those resulting from increased population density, are the most likely explanations for lower fertility in South Carolina textile locations.

CONCLUSION

Fertility decline in the nineteenth-century United States has often been attributed to a quickening pace of industrial activity. I exploit the fact that South Carolina's industrial experience between 1881 and 1900 can be proxied by the textile industry in particular and evaluate the impact of the establishment of a textile mill on rural, marital fertility rates in South Carolina. Using a difference-in-difference approach, I estimate a 6 to 10 percent reduction in fertility following textile mill establishment.

In order to evaluate potential mechanisms to explain this result, I build a household data sample. I use the location of male heads of household at two points in time to measure migration and show that migrants exhibited substantially lower fertility than natives in industrial townships. Observing that migrating households were also more likely to be employed in the textile industry, I argue that an increase in the costs of raising children, including a heightened opportunity cost of female time, led to lower household fertility in mill locations. I hypothesize that the separation of migrating households from their extended families may also have raised the costs of childrearing and resulted in lower migrant fertility in industrial locales. Finally, the results indicate that reductions in child survival rates among migrating industrial workers and increases in population density also contributed to the observed fertility reduction.

Footnotes

1 Total fertility rate for American females as reported by Haines, “White Population.”

2 The work of Richard Easterlin and his students on the targeted bequest hypothesis is another step in this direction. See Yasuba, Birth Rates; Easterlin, “Does Human” and “Factors”; and Schapiro, “Land Availability.”

3 David and Sundstrom, “Old-Age Security Motives”; and Carter, Ransom, and Sutch, “Family Matters.” In addition, see Haines, “Economic History,” pp. 212–13, for a discussion of the cultural concerns.

4 See Steckel, “Fertility Transition”; Guest, “Social Structure”; David and Sundstrom, “Old-Age Security”; Vinovskis, “Socioeconomic Determinants”; and Wahl, “Trading Quantity.” These studies employ cross-sectional variation in state-level labor force composition, manufacturing wages, average farm values, literacy rates, school attendance, child employment, sex ratios, presence of financial intermediaries, and/or land availability to explain fertility rates and reach their respective conclusions.

5 See Galor and Weil, “Population”; Kögel and Prskawetz, “Agricultural Productivity”; and Tamura, “Human Capital.”

6 See Lagerlöf, “Gender Equality”; and Galor and Weil, “Gender Gap.” Schultz, “Changing World Prices,” Brown and Guinnane, “Fertility Transition,” and Crafts, “Duration,” document a negative correlation between female labor force opportunities and fertility rates in Sweden, Bavaria, and England/Wales, respectively.

7 Doepke, “Accounting for Fertility Decline,” argues that child labor regulation is a critical component in explaining fertility declines during periods of economic growth. This view is shared by Hazan and Berdugo, “Child Labor.” Moehling, “State Child Labor Laws,” on the other hand, argues that in the United States, child labor legislation was enacted only after industry's resistance had waned and thus had very little impact on actual child employment.

8 It is an empirical regularity that urban fertility rates were lower than rural ones as early as the nineteenth century. See Vinovskis, “Socioeconomic Determinants,” for the empirical results and Guinnane, “Historical Fertility Transition,” for a more detailed discussion.

9 See Haines, “White Population.” In the short run, industrialization brought increased population density and hazardous work conditions that may have contributed to higher child mortality rates. But in the long run, rising incomes outweighed this effect by providing families the means for better nutrition and sanitation. The expected correlation between child mortality and fertility is ambiguous. See Doepke, “Child Mortality,” for a negative correlate and Sah, “Effects,” and Kalemli-Ozcan, “Stochastic Model,” for a positive one.

10 Carlton, Mill and Town, p. 40.

11 For example, carpentry at 7.5 percent of male manufacturing employment; and dressmaking and seamstresses at 32.3 percent of female manufacturing employment. Author's analysis of IPUMS 1900 1 percent sample.

12 Tang, Economic Development, p. 64.

13 U.S. Commissioner of Labor, Seventh Annual Report.

14 Steam power also required massive amounts of coal input, and coal was an expensive commodity to transport. Proximity to railroad lines likely replaced water flow as the primary driver of mill location by the late 1890s. If railroad location was, in turn, dependent on some aspect of local population also correlated with fertility, this is cause for concern. In unpublished OLS analysis, I explicitly control for the presence of a railroad and find no significant effect on fertility and no change in the estimated impact of industrialization on fertility. I have railroad data for 1890 only, and so do not include this variable in the reported difference-in-difference results for 1880 to 1900.

15 The first electrically powered mill, Orr Mills, was built in Anderson, South Carolina in 1899.

16 See Bleakley and Lin, “Portage,” for potential differences in portage locations.

17 There are 484 South Carolina townships in 1900 and 461 in 1880. In order to perform difference-in-difference analysis, I must create consistent township barriers between years. I do this using the county boundary descriptions from the two census enumerations. The process is not a precise one and requires some judgment calls. I also must consolidate some townships, and the final number for estimation is 391.

18 Sanderson, “Quantitative Aspects.”

19 There is nothing special about the 5-year cutoff here, and a sensitivity test using F3 generates similar results. Using longer fertility windows increases the probability that fertility measures will be biased by migration or mortality. Preston, Lim, and Morgan, “African American Marriage,” document a tendency of African American women to legitimate out-of-wedlock births by claiming a marital status of married, widowed, or divorced when the female was actually single. This should serve to make my estimate more accurate as it ensures that even out-of-wedlock fertility is appropriately captured.

20 Davison Publishing Company, Davison's Textile Blue Book.

21 I have also performed the analysis using a propensity score matching estimator that derives an estimated probability of textile mill establishment between 1881 and 1900 based on 1880 characteristics of the township. Matching procedures are then used to estimate the fertility impact. The estimates of the impact of industry on fertility from this matching exercise are larger in absolute value than the difference-in-difference estimates.

22 The strong caveat here is that the accuracy of this fertility measure will be lower during the early years of this 20-year window. The 1881 fertility rate, for example, is calculated as the number of 19-year-olds observed in a township in 1900 (who would have been born in 1881) divided by the number of women aged 37 to 61 in 1900 (who would have been 18 to 42 in 1880). This number will clearly be subject to bias from migration and mortality if those rates differ by industrial status, but the bias likely diminishes over time such that values closer to 1900 are more accurate of township-level fertility than earlier dates. This issue, plus the unavailability of other covariates (sex ratio, marriage rate, urbanization, racial composition, etc.) for years between 1881 and 1899, precluded the possibility of performing detailed panel data analysis.

23 The regression is where the summation is taken over all townships. The residuals then represent deviations from the township average fertility rate for 1880–1899. I also smooth the results by averaging these residuals across two-year intervals. A chart without smoothing contains many more spikes in the residual, but generates the same conclusion.

24 Sensitivity tests to setting the cutoff at 1,000, 2,000, 3,500, and 5,000 do not show any remarkable difference in conclusions relative to the baseline.

25 North Atlantic Population Project and Minnesota Population Center, Complete Count Microdata.; and U.S. Bureau of the Census, Tenth Census.

26 U.S. Bureau of the Census, Twelfth Census. Every individual in the 1900 U.S. Census has been digitized on Ancestry.com and is searchable by a subset of household characteristics, including name, race, and age. Failed matches result both from locating no matching individuals in 1900 (75 percent of failed matches) and from locating more than one matching individual in 1900 (25 percent of failed matches).

27 The size or direction of this bias is unknowable.

28 Poisson estimation does not remarkably alter the results.

29 The analysis excludes households with a female spouse less than 15 years of age and greater than 50 years of age as these entries are more likely data errors than actual spousal matches.

30 It does not appear that migrants’ fertility rates were mismeasured due to leaving children behind, the so-called “missing children” hypothesis. See Moehling, “Broken Homes.” Comparing the ratio of children present in the household to surviving children for young migrant mothers and young stayers in industrial townships, there is minimal difference. The number of additional “missing” children for migrants relative to stayers is 3 out of 1,000, much smaller than the fertility differential observed in Table 4.

31 On the other hand, higher rates of infant mortality may lead to higher fertility by reducing birth intervals.

32 Unfortunately, standardized infant and child mortality statistics for South Carolina townships in 1900 do not exist.

33 A caveat is in order: this statistic will reflect the cumulative number of deaths of the household's children (in infancy or otherwise) over the entire span of the female's childbearing. It is impossible to determine from this data whether these deaths occurred in years before or after the introduction of a township's textile mills.

34 Restricting the sample to younger women or newly married couples does not change the result, and replacing with the industrialization proxies employed previously generates the same conclusion.

35 This hypothesis is further supported by the fact that all migrants, in industrial locales and otherwise, exhibited lower fertility rates than their peers, and the absence of an extended family would have affected migrants no matter their final destination. But migrants to textile locations exhibited even lower fertility rates and may also have been differentially affected by the loss of extended family. Their increased likelihood of being engaged in occupations which were incompatible with childrearing would have increased their reliance on child care relative to migrants to other locations.

36 See Becker, Hornung, and Woessman, “Education,” for evidence that the textile industry in Prussia exhibited low returns to education relative to metal, rubber, and other industries.

References

REFERENCES

Atack, Jeremy, al, et. “Did Railroads Induce or Follow Economic Growth? Urbanization and Population Growth in the American Midwest, 1850–1860.” Social Science History 34, no. 2 (2010): 171–97.Google Scholar

Beakley, Hoyt, and Lin, Jeffrey. “Portage: Path Dependence and Increasing Returns in U.S. History.” Forthcoming in the Quarterly Journal of Economics, 2012.Google Scholar

Becker, Sascha O., Hornung, Erik, and Woessmann, Ludger. “Education and Catch-Up in the Industrial Revolution.” American Economic Journal: Macroeconomics, forthcoming.Google Scholar

Brown, John C., and Timothy, W. Guinnane. “Fertility Transition in a Rural, Catholic Population: Bavaria, 1880–1910.” Population Studies 56, no. 1 (2002): 35–49.Google Scholar

Carlton, David L.Mill and Town in South Carolina: 1880–1920. Baton Rouge: Louisiana State University Press, 1982.Google Scholar

Carter, Susan B., Ransom, Roger L., and Sutch, Richard. “Family Matters: The Life-Cycle Transition and the Unparallelled Antebellum American Fertility Decline.” In History Matters: Essays on Economic Growth, Technology, and Demographic Change, edited by Guinnane, Timothy W., Sundstrom, William A., and Whatley, Warren, 271–327. Stanford, CA: Stanford University Press, 2002.Google Scholar

Crafts, N.F. R.“Duration of Marriage, Fertility, and Women's Employment Opportunities in England and Wales in 1911.” Population Studies 43, no. 2 (1989): 325–35.CrossRef Google Scholar

David, Paul, and Sundstrom, William. “Old-Age Security Motives, Labor Markets, and Farm Family Fertility in Antebellum America.” Explorations in Economic History 25, no. 2 (1988): 164–97.Google Scholar

Davison Publishing Company. Davison's Textile Blue Book. Ridgewood, NJ, various years.Google Scholar

Doepke, Matthias. “Accounting for Fertility Decline During the Transition to Growth.” Journal of Economic Growth 9, no. 3 (2004): 347–83.CrossRef Google Scholar

Doepke, Matthias. “Child Mortality and Fertility Decline: Does the Barro-Becker Model Fit the Facts?” Journal of Population Economics 18, no. 2 (2005): 337–66.Google Scholar

Easterlin, Richard A.“Does Human Fertility Adjust to the Environment?” The American Economic Review 61, no. 2 (1971): 399–407.Google Scholar

Easterlin, Richard A.“Factors in the Decline of Farm Family Fertility in the United States: Some Preliminary Research Results.” Journal of American History 63, no. 3 (1976): 600–14.Google Scholar

Galor, Oded, and David, N. Weil. “The Gender Gap, Fertility, and Growth.” American Economic Review 86, no. 3 (1996): 375–87.Google Scholar

Galor, Oded, and David, N. Weil. “Population, Technology, and Growth: From Malthusian Stagnation to the Demographic Transition and Beyond.” American Economic Review 90, no. 4 (2000): 806–28.CrossRef Google Scholar

Guest, Avery M.“Social Structure and U.S. Inter-State Fertility Differentials in 1900.” Demography 18, no. 4 (1981): 465–86.Google Scholar

Guinnane, Timothy W.“The Historical Fertility Transition and Theories of Long-Run Growth: A Guide for Economists.” Journal of Economic Literature, forthcoming.Google Scholar

Haines, Michael R. “Economic History and Historical Demography: Past, Present, and Future.” In The Future of Economics, edited by Field, Alexander J., 185–253. Hingham, MA: Kluwer Academic Publishers, 1995.Google Scholar

Haines, Michael R. “The White Population of the United States, 1790–1920.” In A Population History of North America, edited by Haines, Michael and Steckel, Richard, 305–69. Cambridge: Cambridge University Press, 2000.Google Scholar

Hazan, Moshe, and Berdugo, Binyamin. “Child Labor, Fertility, and Economic Growth.” The Economic Journal 112, no. 482 (2002): 810–28.Google Scholar

Kalelmi-Ozcan, Sebnem. “A Stochastic Model of Mortality, Fertility, and Human Capital Investment.” Journal of Development Economics 70, no. 1 (2003): 103–18.Google Scholar

Kögel, Tomas, and Prskawetz, Alexia. “Agricultural Productivity Growth and the Escape from the Malthusian Trap.” Journal of Economic Growth 6, no. 4 (2001): 337–57.Google Scholar

Lagerlö, Nils-Petter. “Gender Equality and Long-Run Growth.” Journal of Economic Growth 8, no. 4 (2002): 403–26.CrossRef Google Scholar

Moehling, Carolyn M.“State Child Labor Laws and the Decline of Child Labor.” Explorations in Economic History 36, no. 1 (1999): 72–106.Google Scholar

Moehling, Carolyn M.“Broken Homes: The ‘Missing’ Children of the 1910 Census.” Journal of Interdisciplinary History 33, no. 2 (2002): 205–33.CrossRef Google Scholar

North Atlantic Population Project and Minnesota Population Center. NAPP: Complete Count Microdata. NAPP Version 2.0 [computer files]. Minneapolis, MN: Minnesota Population Center [distributor], 2008.Google Scholar

Preston, Sameul H., Lim, Suet, and Philip Morgan, S.. “African American Marriage in 1910: Beneath the Surface of Census Data.” Demography 29, no. 1 (1992): 115.Google Scholar

Ruggles, Steven, al, et. Integrated Public Use Microdata Series: Version 4.0 (Machine-Readable Database). Minneapolis, MN: Minnesota Population Center. 2008.Google Scholar

Sah, Raaj K.“The Effects of Child Mortality Changes on Fertility Choice and Parental Welfare.” Journal of Political Economy 99, no. 3 (1991): 582–606.Google Scholar

Sanderson, Warren. “Quantitative Aspects of Marriage, Fertility, and Family Limitation in Nineteenth-Century America: Another Application of the Coale Specifications.” Demography 16, no. 3 (1979): 339–58.Google Scholar

Schapiro, Morton O.“Land Availability and Fertility in the United States, 1760–1870.” The Journal of Economic History 42, no. 3 (1982): 577–600.Google Scholar

Schultz, T. Paul. “Changing World Prices, Women's Wages, and the Fertility Transition: Sweden, 1860–1910.” Journal of Political Economy 93, no. 6 (1985): 1126–49.CrossRef Google Scholar

Steckel, Richard H. “The Fertility Transition in the United States: Tests of Alternative Hypotheses.” In Strategic Factors in Nineteenth-Century American Economic History, edited by Goldin, Claudia and Rockoff, Hugh, 351–74. Chicago: University of Chicago Press, 1992.Google Scholar

Stover, John R.The Railroads of the South, 1865–1900. Chapel Hill: North Carolina University Press, 1955.Google Scholar

Tamura, Robert. “Human Capital and the Switch from Agriculture to Industry.” Journal of Economic Dynamics and Control 27, no. 2 (2002): 207–42.Google Scholar

Tang, Anthony M.Economic Development in the Southern Piedmont, 1860–1950: Its Impact on Agriculture. Chapel Hill: University of North Carolina Press, 1958.Google Scholar

Thompson, Holland. From the Cotton Field to the Cotton Mill: A Study of the Industrial Transition in North Carolina. New York: The MacMillan Company, 1906.Google Scholar

Todd, Petra E. “Matching Estimators.” The New Palgrave Dictionary of Economics Online. In The New Palgrave Dictionary of Economics, Second Edition, edited by Durlauf, Steven N. and Lawrence, E. Blume. New York: Palgrave Macmillan, 2008 (http://www.dictionaryofeconomics.com/article?id=pde2008_M000365, accessed 09 December 2011).Google Scholar

United States of America, Bureau of the Census. Tenth Census of the United States, 1880. Washington, DC: GPO, 1880.Google Scholar

United States of America, Bureau of the Census. Twelfth Census of the United States, 1900. Washington, DC: GPO, 1900.Google Scholar

United States of America, Bureau of the Census. Index to the 1900 United States Federal Census. Digital copy of original records in the National Archives, Washington DC. Available at http://www.ancestry.com, subscription database, 2008.Google Scholar

United States Bureau of Labor. Report on Condition of Woman and Child Wage-Earners in the United States. Washington, DC: GPO, 1910.Google Scholar

United States Commissioner of Labor. Seventh Annual Report. Cost of Production: The Textiles and Glass. Washington, DC: GPO, 1892.Google Scholar

Vinovskis, Maris. “Socioeconomic Determinants of Interstate Fertility Differentials in the United States in 1850 and 1860.” Journal of Interdisciplinary History 6, no. 3 (1976): 375–96.Google Scholar

Wahl, Jenny Bourne. “Trading Quantity for Quality: Explaining the Decline in American Fertility in the Nineteenth Century.” In Strategic Factors in Nineteenth-Century American Economic History, edited by Goldin, Claudia and Rockoff, Hugh, 375–97. Chicago: University of Chicago Press, 1992.Google Scholar

Wright, Gavin. “Cheap Labor and Southern Textiles Before 1880.” The Journal of Economic History 39, no. 3 (1979): 655–80.Google Scholar

Yasuba, Yasukichi. Birth Rates of the White Population in the United States, 1800–1860: An Economic Study. Baltimore, MD: Johns Hopkins University Press, 1962.Google Scholar

FIGURE 1F5 FERTILITY RATES FOR INDUSTRIALIZED AND NONINDUSTRIALIZED SOUTH CAROLINA TOWNSHIPS (TOP PANEL) AND FOR ALL SOUTH CAROLINA, THE SOUTH, UNITED STATES (BOTTOM PANEL)