Appendix III Demand for Land, 1664-1706

Concomitant with rising population in the extensive development of the Chesapeake in the seventeenth century was the regular acquisition of new land through the process of patenting. Historians have suggested that tobacco productivity per acre changed relatively little from the mid-seventeenth century into the nineteenth century, and thus planters in the Chesapeake expanded production (whether extensively or intensively) by putting more hands and industry on additional land rather than attempting to increase output from existing acreage. ¹

The only constraints on new land acquisition were the requirements of patenting and the continuing obligations of land ownership established by colonial law dating back to the Second Charter in 1609. ² For analyses of the land patents in Virginia we are lucky to have the excellent series of abstracts Cavaliers and Pioneers by Nell Marion Nugent, from which I created a computer data base with all pertinent information for the years 1660-1706. An earlier study showed the patent record to be fairly complete for the years after 1663 and suitable for purposes of the analysis of new land acquisition. ³

Since land was hardly a factor of production in short supply and most planters had enough land on hand to increase production without acquiring new land, we may not expect to find as close a correlation between new land acquisition and tobacco prices as we did with tobacco productivity and demand for labor. And, indeed, the evidence on demand for land is far more ambiguous. (See Figure 6.)

Figure 6

The question arises whether one can study demand for land simply by studying new land acquired through the patenting process and not the regular everyday buying and selling of land which steadily increased in volume over the course of the seventeenth century. If we had good time-series data on the value of land accompanied with particulars on land quality and degree of improvement, then perhaps we could use such sales data to measure changing demand for land. But we do not have such data and so we must fall back on using patent data. Although the patent process may not have directly affected planters who remained in the more settled regions, it did affect them indirectly by siphoning off to less settled regions other planters who would have otherwise competed with them for land. Overall new land acquisition well capture aggregate changes in demand for land in such frontier regions as the seventeenth-century Chesapeake. ⁴

Two potential problems in using land patents as a measure of the demand for land are corruption of the patent process and land speculation. Some contemporaries complained that, whether land was held for hoarding or speculative purposes, avaricious planters abused the patent system by ignoring all the constraints and obligations and simply patenting at will. ⁵ Closely linked to such an interpretation are historical arguments tracing the source of the “insatiable lust for land” to the rapid exhaustion of tobacco lands. ⁶ If corruption were dominant in new land acquisition, undertaken simply because the opportunity for corruption existed, land acquisition might be fairly independent of changing tobacco prices, although it is possible that Virginians engaged in such practices out of personal necessity imposed by falling tobacco prices. But it is also possible that Virginians engaged in corruption and/or speculation only in anticipation of or concomitant with more general increased demand for land. ⁷ Thus corruption and/or speculation could represent a function of general demand for land, a view which an analysis of the relationship between the land patent process and land ownership in 1704 tends to confirm. ⁸

Recent historians, following the work of Carville V. Earle, have downplayed the role of avarice and soil exhaustion in new land acquisition, suggesting that planters in general possessed enough land to allow depleted soils to return to forest and recover their fertility after a period of about twenty years. ⁹ However, modeling such a practice as a variable in an analysis of land acquisition proves difficult to distinguish from soil exhaustion, since both suggest that demand for land was not a function of tobacco prices but of the size of the human and animal population, the quality of the land, and the number of years the soil had been put in tobacco production. Both models suggest that demand for land was a negative function of cumulative land acquisition, a positive function of the tithable population, and, if supply of land proved insufficient to maintain productivity, a positive function of time. ¹⁰ This model simply suggests that planters needed less land per laborer since the land could be recycled.

Historians have long noted a link between the size of the population and new land acquisition in the colonial Chesapeake. Historically the two were linked umbilically through the headright system, almost exclusively the grounds on which land patents were awarded in Virginia throughout the seventeenth century. In this system, the headright guaranteed that a grant of fifty acres be made for every person immigrating to the colony, the grant being “‘made respectively to such persons and their heirs at whose charges the said persons going to inhabit in Virginia shall be transported.'” ¹¹ The work of Edmund S. Morgan and Russell R. Menard, revealing the ready market in headrights and corruption of the system, has destroyed any possibility of simplistically equating land acquisition with immigration. ¹² Nevertheless historians continue to believe the two were strongly linked, like those who emphasize the historical maintenance of an optimum balance between factors of production. In this view, land acquisition was dictated by the availability of labor since labor was the factor in shortest supply. ¹³ Most historians, however, more generally stress “traditional” motivations of farmers for land: the desire to provide land for posterity under “inexorable pressure of demography…always producing more sons than fathers,” the desire of freedmen for land of their own, the dependence of the farmer on capital gains from land improvement rather than annual cash returns from crops. ¹⁴ Thus, whether based on institutional, neoclassical utilitarian, or traditional behavior, historians suggest that new land acquisition was a positive function of total or tithable population and a negative function of cumulative land acquisition, or, more generally, a negative function of population density in terms of cumulative acres per tithable. (See Figure 7.)

Figure 7

One would expect a general decline in the quality of land patented further and further away from navigable streams which might be captured in part by a negative coefficient on cumulative land acquisition. ¹⁵ One should note, however, the significant positive impact on land acquisition resulting from the opening up in 1699 of land on the Pamunkey Neck ceded by the Indians in 1688. ¹⁶

New land acquisition was also a function of capital gains from land improvement and cost of patenting, but there is little evidence to suggest any change over the course of the late seventeenth century. ¹⁷

Finally, as with measures of tobacco productivity in which we had to take into account a lag of one year between planting and marketing (see Appendix I), so with land patenting there were institutional filters which created a delay between demand for and the actual acquisition of land by patenting. ¹⁸ The amount of delay undoubtedly varied from acquisition to acquisition and an exact average is impossible to compute from existing data. The best estimate for the late seventeenth century suggests the average lag was less than one year and the empirical analysis below suggests that a one-year lag produced the best fit. ¹⁹

Test Results

The best-fit equation for new land acquisition for the years 1664-1706 is presented in Table 4. Figure 8 graphically reveals the closeness of the match between predicted and actual measures of new land acquisition. The analysis shows that average and minimum tobacco prices had a statistically insignificant impact on new land acquisition. The model of land acquistion does not explain as much of the variance as the models of tobacco productivity and servant values; the coefficient on the average-price variable (PA2) and the check variable (CHECK) are insignificant albeit with the right sign, and the coefficient on the minimum-price variable (PMIN4) is insignificant and the wrong sign. However, the magnitude of the coefficients at least suggests that demand for land did not move in the opposite direction from demand for labor and tobacco productivity. A fall in the two-year average price of tobacco (PA2) of a tenth of a penny led to an increase in demand for land of 9170 acres, while a fall in the four-year minimum price (PMIN4) of the same amount led, ceteris paribus, only to a decrease of 4550 acres. Furthermore, a fall in the price of tobacco since the previous year, regardless of the magnitude (as captured by CHECK), led to an increase in demand for land of 11130 acres. Thus, as a result of the fall in tobacco prices, if all other factors were held constant, planters in 1694 when prices were relatively low would have demanded 21480 acres more than they did in 1667 when prices were relatively high. ²⁰ On the other hand, with rising prices during the Peace of Ryswick, planters in 1701 would have demanded 14580 acres less than they had in 1694. ²¹

Table 4

Figure 8

That actual land acquisition went in the exact opposite direction, with a fall in new land acquisition of 80700 acres from between 1667 and 1694, and a rise of 41100 acres between 1694 and 1701, shows that other factors were not constant. It also proves the weakness of tobacco price models compared to demographic factors in the demand for land. By 1694 so much land had already been patented (SUM1) that there was little demand for additional land regardless of the fall in tobacco prices, but during the Peace of Ryswick the substantial increase in population (DPOP2) increased demand for land regardless of the rise in tobacco prices.

The evidence on demand for land is thus far more ambiguous than the results on tobacco productivity and demand for labor. It appears that new land acquisition was not on a par in terms of planter responses to changing tobacco prices with increasing production on land already owned and acquiring new servants.