Jonathan B. Pritchett

Selection in the Market for Slaves: New Orleans, 1830–1860

Greenwald and Glasspiegel argue that adverse selection depressed the market prices of slaves, causing current researchers to overestimate the rate of return from slavery. In this paper we test for the presence of adverse selection by comparing the prices of local slaves with the prices of slaves sold from estate sales. We find no difference in the prices of these slaves, from which we conclude that there was no significant adverse selection in the market. Instead, we propose an alternative explanation for the observed pattern of slave prices based on the costs of shipping slaves to the New Orleans market.

A Peculiar Sample: The Selection of Slaves for the New Orleans Market

Domestic slave traders selected taller slaves for shipment to the New Orleans market in order to increase their profits. Because traded slaves made up a large share of the slaves shipped coastwise, age/height profiles constructed from the shipping manifests are biased upwards. The extent of the bias appears to be small for adult slaves but not for children; those listed on the manifests were taller than the general population of a comparable age.

Cox Regression with Alternative Concepts of Waiting Time: The New Orleans Yellow Fever Epidemic of 1853

Event data can often be analysed using different concepts of waiting time. Our application offers three choices: calendar-time, age, and duration of residence in New Orleans. We exploit the semi-parametric features of Cox regression and estimate parallel specifications in which mortality risk is treated as an arbitrary function of one of the three alternative time measures, while the remaining two enter the hazard parametrically. Comparisons of the parameter estimates with the corresponding estimates of the baseline hazards from the crux of a simple specification checking procedure. In our formal treatment we rely on Aalens Multiplicative Intensity formulation and tackle complications such as left-truncation, functional form specification, and choice-based sampling.

The Interregional Slave Trade and the Selection of Slaves for the New Orleans Market

The Interregional Slave Trade and the Selection of Slaves for the New Orleans Market The United States slave trade and, more generally, the forced migration of slaves, was the result of interregional differences in the productivity of labor. The commercial success of sugar and short-staple cotton increased the prices of slaves in New Orleans relative to the prices of slaves in the upper South. The interregional price differential created arbitrage opportunities for traders, who purchased slaves in the upper South and resold them in New Orleans. Competition among traders placed a bound on the regional price differential equal to the cost of transporting the slaves to New Orleans.1

The Burden of Negro Schooling: Tax Incidence and Racial Redistribution in Postbellum North Carolina

The disparate allocation of public revenues for education reflected racial discrimination in the public sector of the South.1 Suffrage restrictions in North Carolina after 1901, in conjunction with changes in the tax law, enabled whites to increase spending for the public education of white children without increasing it for black children.2 Officials in North Carolina justified (and condemned) this growing racial disparity in spending by asserting that whites as a group subsidized the education of blacks. Their estimates of the tax burden of education expenditures received national attention at the time and continue to be cited by current scholars.3 Their estimates, however, only consider the legal and not the economic incidence of the property tax. Tax shifting increases the burden of school taxes on renters and workers while reducing the tax burden on property owners. Because whites owned a disproportionate share of the land, the shifting of the property tax reduced the burden on whites while increasing it on blacks. This note focuses on the economic incidence of public school taxes. This represents a significant departure from previous studies in that tax incidence is explicitly considered in the context of a single sector model of an agrarian economy. The burden is shown to be largely dependent on the share of land and labor supplied by blacks and on the elasticities of supply and derived demand for property. While no estimates of the burden are presented, the necessary conditions under which blacks paid for their own education are derived.

Quantitative Estimates Of The United States Interregional Slave Trade, 1820 1860

“It is impossible to say definitely what percentage of the movement of slaves to the lower South was comprised in the domestic slave trade.…†L. C. GrayGray, History, p. 658.Between 1790 and 1860 approximately 835,000 slaves moved from the exporting areas to the importing areas of the United States South.Fogel and Engerman, Time on the Cross: The Economics, p. 47. Gutman and Sutch (“Slave Family,†p. 99) estimate “more than a million slaves entered the interstate migration between 1790 and 1860.†For other estimates of the net migration of slaves (or blacks) within the United States, see Collins, Domestic Slave Trade, pp. 61–67; Bancroft, Slave-Trading, pp. 382–406 and sources therein; Lang, Effects, pp. 57–72; Sutch, “Breeding,†pp. 178–80; McClelland and Zeckhauser, Demographic Dimensions, pp. 159–64; and Tadman, Speculators, pp. 237–47. These slaves either moved with their owners as whole plantations or were bought by interregional traders and shipped to the lower South. Until recent years historians had no way of estimating the share of sales in the interregional movement of slaves. Consequently, most of the early quantitative estimates of the interregional slave trade were vague or deliberately imprecise.According to Weld (Slavery, p. 13), “perhaps four-fifths or more†of the slaves were transported through the interregional slave trade. For the period 1820 to 1850, Collins (Domestic Slave Trade, p. 62) estimated that less than two-fifths of the migrants were carried south through the trade. According to Phillips (“Racial Problems, p. 223),†The conjecture of about 25,000 per year which has often been made for the interstate slave trade is probably as just an approximation as can be had.…†Bancroft (Slave-Trading, p. 398) “supposes that†the slave trade accounted for 70 percent of the slave exports. Stampp (peculiar Institution, p. 239) suggests that the slave trade accounted for “perhaps a majority†of the slave exports. According to Miller (“Note,†p. 182), “Estimates indicate that from two-fifths to one-half of these [slaves] may have been sold by their owners and moved by slave-traders to newer regions. Presumption lies in favor of the lower estimate,…â€

North Carolina’s Public Schools: Growth and Local Taxation

The Rapid increase in public spending for white schools that occurred in North Carolina after the turn of the century led to a large racial disparity in the amount spent per child by 1910. Previous scholars have attributed this racial difference in school spending to the disfranchisement of the black voter (Margo, 1982). It was argued that once blacks were prevented from voting, the white members of the school boards were able to divert the public funds which were initially allocated for the education of black children. The most widely accepted version of this theory is credited to Horace Mann Bond (1934) who studied education expenditures for black children in Alabama. Bond argued that the governmental level at which schools were financed was important in determining the racial division of public school funds since the white members of the county school boards were particularly inclined to divert the funds allocated by the state government. The state funds which were allocated to the local school boards in Alabama were not required to be shared equally between black and white students. After blacks had been disfranchised, the county school boards responded by allocating a disproportionate share of these state funds for the education of white children.

A Peculiar Sample: A Reply to Steckel and Ziebarth

Steckel and Ziebarth (2014) find that biases in height by age imposed by traders versus non-traders were negligible. Importantly, their method of identifying traders differs from that of Pritchett and Freudenberger (1992). Using a sample of inward coastwise manifests for the port of New Orleans, we show that Steckel and Ziebarth made errors classifying shippers, that they underestimate the relative number of slaves shipped by traders, and that their empirical estimates of selection bias are attenuated towards zero.