Joshua Rosenbloom and Brandon Dupont The Impact of the Civil War on Southern Wealth Mobility WEAI Conference, 2015 Joshua Rosenbloom and Brandon Dupont

Impact of the Civil War

Did the War Displace the Southern Elite? The number of Southern landholdings doubled and their average size was cut in half after the Civil War (Shugg, 1939) Southern planter elites were therefore uprooted by the war (C. Vann Woodward, 1951) More recent work has challenged this view Wiener (1979): 43% of 236 largest landholders in five Alabama counties remained among the elite in 1870 Campbell (1982): Geographical persistence in one Texas county – 43% of large planters persistence from 1860 to 1870; 22% of poor whites persisted

The Open Questions Wiener and Campbell study transitions within a circumscribed geography (counties) Recent advances make it possible to link records nationally We ask two questions Where did the 1870 top wealth holders come from? What happened to top 1860 wealth holders by 1870?

Methods and data 1860 and 1870 Census included questions on real and personal property holding We start with top 10% of wealth holders in each year Link backward/forward using Ancestry.com Household heads 25-65 years old in each year Living in Northeast, North Central, South Atlantic or South Central (exclude Western areas) Search based on name, age (+ 2 years), birthplace Still working on linkage Results here reflect backward linkage for ~top 5% in 1870

Results We find a fair amount of turnover at the top But considerably more in the South than the North Of top 5% of wealth holders in the South, 48% had been below the 90th percentile in 1860 In the North this figure was just 33%

Backward Linkage We start with 1870 1.2% IPUMs We have so far examined 2,506 (of about 7,700) records for linkage We accept a link if name and age match 46% of records are successfully linked Requiring birthplace would lower linkage to 45% We assume high wealth individuals are easier to link Wealth cut off determined by percentiles in the 1% sample We start with the wealthiest individuals and are working down the list

Linked observations based on Linkage Results Linked observations based on Difference in means Name and + No Link Age Birthplace HH Member P value of (1) (2) (3) (4) (2)-(1) difference Nobs 1,352 1,154 1,124 991 age 47.8 51.9 51.8 52.8 4.11 0.000 Nonwhite 1.1% 0.2% 0.1% -0.9% 0.003 North 52.1% 68.9% 69.1% 68.8% 16.8% Urban Resident 62.9% 64.1% 64.2% 65.1% 1.3% 0.515 male 91.3% 94.3% 94.4% 95.3% 3.0% Foreign Born 16.1% 10.9% 11.1% 11.4% -5.2% Value of Real Property $24,166 $29,037 $28,722 $28,438 $4,871 0.002 Value of Personal Property $12,285 $15,583 $15,084 $14,014 $3,298 0.029 Has wife 80.7% 86.7% 86.9% 89.2% 6.0% Has child 82.7% 89.7% 89.6% 91.0% 7.0% Living Outside State of Birth 50.3% 43.3% 43.4% 43.8% -7.0% Professional & Technical 7.6% 8.8% 8.9% 8.3% 0.303 Farming 41.2% 43.9% 45.7% 2.6% 0.196 Clerical&Managerial 28.6% 25.8% 25.4% 24.9% -2.8% 0.116 Sales 1.7% 1.8% 0.5% 0.331 Craftsmen 3.7% 5.0% 5.1% 5.2% 0.107 Operatives & Kindred workers 1.9% 2.8% 0.8% 0.164 Service workers 0.4% 0.3% -0.2% 0.434 Laborers 0.9% -0.5% 0.079 Non-Occupational 14.3% 11.5% 11.6% 10.7% 0.035 Linked individuals are Older More northern More likely to be male Native born Wealthier More likely to be married or have children More likely to live in state of birth

1860 Wealth Levels of the 1870 Top 5% of Wealth Holders, by Region Shows the origins of 1870 top-wealth holders that we have been able to locate in the 1860 census in terms of their relative wealth holding at the time of that census. The first column for each region reports totals, while the next two columns break the top wealth group into two groups: the wealthiest 1 percent, and those in the top 95-99 percent of wealth holders. In both regions roughly similar fractions (10.3 percent in the South vs. 11.6 percent in the North) moved from the bottom 55% into the top tier of wealth holders But a much larger number of those in the 55th-90th wealth percentiles in 1860 had been able to move up in the South. In 1870, nearly 40 percent of the top wealth holders were drawn from this group. In contrast, this group made up just over 20 percent of top wealth holders in the North.

Probits: probability that an individual in the top 5% of 1870 wealth holders was among the top 10% of wealth holders in 1860 One concern in interpreting the results in previous table: may be influenced by differences in sample composition across regions. Probability of linkage varied systematically with a number of personal characteristics, and it is possible that these differences affect differences in measured persistence across regions. To test this hypothesis we report the results of estimating a probit regression where the dependent variable takes the value 1 if the individual was in the top 10% of 1860 region wealth holders. Whether demographic controls included or not: 1870 residence in North increases the probability that a top wealth holder was in the top 10% in 1860 by about 20 percentage points.

Interstate Migration Between 1860 and 1870 Only a few wealthy individuals migrated between regions in the 1860s: only 5% of top southern wealth holders and 1% of top northern wealth holders had lived in a different state 10 years earlier. Interstate migration was also relatively uncommon, and about equally likely within each region. Among top wealth holders in 1870 only 12.3 percent of those in the South had moved across state lines in the previous decade while the corresponding figure for the wealthiest northerners was 9.3 percent.

Conclusion We are expanding our sample of top 1870 wealth holders linked to the 1860 census and gathering data needed to link top wealth holders in 1860 to the 1870 census. By examining the fates of those at the top of the wealth distribution in 1860 we expect in the future to be able to provide greater insight about where those who fell out of the top wealth group landed.