“Theft of Oneself”: Runaway Servants in Early Maryland: Deterrence, Punishment, and Apprehension

Abstract

Immigrant indentured and transported convict servants had an incentive to breech their labor contracts by running away. Masters and servants in colonial Maryland engaged in strategic behaviors to deal with this contract breech incentive. In the seventeenth century, masters altered the colony’s statutory laws to deter and thwart servant escape, and servants chose the escape routes that offered the best chance of not being returned to Maryland. Strategic behaviors changed by the eighteenth century. Masters quickly advertised runaway servants in Maryland newspapers, and servants selected when to run that delayed the appearance of those ads as much as possible.

8.1 Introduction/Summary

Servants who voluntarily entered fixed labor contracts (indentures) comprised a majority of the immigrants arriving from Britain to seventeenth-century Maryland (Grubb 1985a; Smith 1947: 336). Before the late 1680s, and even as late as 1702, they also comprised a majority of the bound labor force in Maryland (Grubb and Stitt 1994). While African slaves became the majority bound labor force in Maryland after 1702, British voluntary indentured servants and transported convict servants continued to arrive in sizable numbers throughout the rest of the colonial period (Bailyn 1986; Grubb 1985a, 2000b).

Indentured immigrants and transported convict servants broke their contracts by running away. This act was called “theft of oneself.” In the seventeenth century, both Virginia and Maryland passed laws to deter and punish such acts. However, Maryland made the legally prescribed punishment for running away harsher than the punishment enacted in the neighboring colony of Virginia. A simple economic model is used to account for this difference. In this model, society desires a “good” called “completed servant contracts.” It produced this good using a cost-minimizing combination of inputs, namely, the apprehension mechanisms and punishments if caught in response to the gains from a successful escape, all of which varied by location within and between the colonies. The method of informing the public of runaway servants and so detaining them in seventeenth-century Maryland was only through “hue and cry”—information passed from one neighbor to the next. Maryland used harsher legal punishments than Virginia not because Maryland planters were crueler, but as a reaction to the greater probability of successful escape and gains from that escape that existed for Maryland servants compared with those faced by servants in Virginia.

By the mid-eighteenth century, Maryland’s laws to deter and punish servants who ran away had become codified and fixed, no longer responsive to changing probabilities of successful escape and gains from that escape. Those probabilities, however, no longer varied substantially across colonies or by location within colonies. Maryland’s method of informing the public of runaway servants and so detaining them had also changed from using just “hue and cry” to using the colony’s weekly newspaper, the Maryland Gazette, to inform fellow colonists and county sheriffs of runaways.

The advertisements for runaway servants in the Maryland Gazette are used to assess the strategic behavior of runaway servants and their masters in mid-eighteenth-century Maryland. No matter the master’s location within the colony, a runaway advertisement was typically placed in the Maryland Gazette within 2 weeks of the servant’s act of running away, appearing in the next issue of the Gazette after the act of running away. Runaways were also more likely to select certain days of the week to run away, namely, the days that increased their chances of successfully escaping. The primary method of informing the public of runaway servants by the mid-eighteenth century led to predictable strategic behaviors on the part of servants and masters regarding contract compliance, enforcement, and apprehension of runaways.

8.2 The Puzzle in Runaway Law in Seventeenth-Century Maryland

The seventeenth-century colonies of Virginia and Maryland were similar. They resided next to each other with economies oriented toward Chesapeake Bay and the export of tobacco to England. Both relied on immigrant indentured servant labor as their principal bound labor force into the late seventeenth century when slaves thereafter became the principal bound labor force. Even so, thereafter immigrant indentured and convict servant labor continued to be imported and used throughout the rest of the colonial period, with most English indentured servants migrating to these two Chesapeake colonies. Trade and migration flourished between the two colonies with little restriction, even though each colony passed its own laws to govern its citizens (Grubb 1985a, 2000b; Grubb and Stitt 1994; McCusker and Menard 1985; Walsh 1977, 1987).

One glaring legal difference between the two colonies was over the treatment of indentured servants, in particular the legally prescribed punishment given the servant when the servant ran away from their master and was subsequently caught and prosecuted. This act of breaking the contract by running away was called “theft of oneself” and was deemed criminal in law in both England and the colonies. Abbot E. Smith (1947: 270, 276–7) summarized and explained the difference as follows:

Why should the laws of Maryland in general have been so much more harsh than those of her neighbors? … Why should a runaway serve ten [extra] days for one [absent] in Maryland, and only two for one in Virginia? Why should a person who harbored a runaway forfeit five hundred lbs. of tobacco for every night in Maryland, and only sixty pounds in Virginia? And most remarkable of all, why, when most colonies progressively made their penal codes milder, should Maryland after abandoning her original death penalty for runaways and substituting double service [two extra days of service for every one day absent], progressively make her code more severe? I know of no answer to these questions, except to assume that the planters of Maryland were a harsher breed than those of Virginia and Pennsylvania. This assumption appears foolish, but one certainly gains the impression from reading court records that not only the laws but also the magistrates of that colony were less merciful. … Obviously the penalties of extra service were imposed principally for the enrichment of the master; there can be no possible reason for the Maryland law with a punishment five times as severe as that of Virginia except that the planters of that colony more openly pursued their own advantage.

Smith’s assertion that differential punishments are due to differential values or tastes is a conclusion, not an explanation. This conclusion is the result of an absence of other explanations. Conclusions such as Maryland planters were harsher or greedier than Virginia planters have the disadvantage of suggesting an end to further investigation. Once an issue arrives at a difference in values, what more can be said? By contrast, the economic approach to explaining differences in servant punishment is to assume fixed values, tastes, and preferences, and look instead at differences in observed opportunities between the two colonies that affected the decision by servants to run away (Becker 1996: 24–49; Diamond 1982).

Published primary sources for the seventeenth century are scarce. The only published court records for Virginia are for Accomack County between 1640 and 1645. Published primary source evidence for Maryland is more plentiful. Secondary sources indicate that Virginia’s penalty of double the time absent added to the contract was constant throughout the seventeenth century. Therefore, the issue to be explained is why Maryland laws differed from Virginia and why Maryland laws changed over time (see Table 8.1).¹

Table 8.1 Maryland laws regarding the punishment for runaway indentured servants

Traditional sentences for violating criminal law in England involved some type of physical chastisement, such as whippings, brandings, or, in extreme cases, hangings (Beattie 1986). Once the punishment was meted out, the person found guilty of the crime (if still living) was returned to society. Incarceration was not used as a sentence or as punishment, but only to hold individuals awaiting trial. The point of punishment in criminal law was to deter similar criminal acts in the future.

In England, as in America, breaching a labor contract by the worker running off was considered a criminal act called “theft of oneself.” Workers who ran off and were caught were returned to their employers to finish their contracts and could also be physically chastised—the physical chastisement being the deterrence part. In England, labor contracts were relatively short and absent workers could be more easily replaced than in colonial America. There was also little labor contract breaking in England due to how English labor contracts were structured (Grubb 2000a). In England, employers lost relatively less from workers running off compared with employers in colonial America.

The punishment of hanging implemented in early Maryland (1639) for the servant’s act of running away from their master was a strong deterrent to servants breaching their contracts. But this punishment also left the master without the rest of the servant’s contracted labor time. Given that servant labor contracts were several years long, the loss of labor value by the master could be substantial. Colonial masters desired compensation for the lost labor value for the time that servants had absented themselves from their masters.

Seeking compensation for damages from the person who breached their contract involves tort law. Immigrant servants had no resources to pay damages except their future post-contract labor time. The punishment in law—that is, requiring servants who ran off and were caught to not just finish their existing labor contracts but to serve extra time beyond the end of their contracts as compensation to masters for the lost labor value while the servants were absent—mixed criminal law (deterrence) with tort law (compensation to the wronged party for breaching a contract).²

The extra time the servant had to serve as compensation for having run off had to be longer than the time the servant had absented himself because (1) it had to also serve to deter future acts of running away and (2) had to compensate the master for the lost time while absent. Because the extra days added to the contract were several years in the future, that future labor value had to be time-discounted back to the present. Everything else held constant, a day’s labor value 4 years from now is worth less than a day’s labor value today. Thus, the extra time served had to be more than the time the servant was absent to fully compensate the master for the lost labor value—to make the master “whole” as phrased in tort law. Colonial law after 1641 required that extra days be added to the end of the contract for the act of running away, more than the number of days the servant was absent (see Table 8.1).

Table 8.1 also shows a sudden shift in punishment in 1666 for the act of running away by the servant. From a level similar to that of Virginia, namely, 2 days extra servitude for every day absent, Maryland substantially increased its punishment in 1666 to 10 days extra servitude for every day absent. While this change generated increased compensation to the master, it also represented increased punishment deterrence to committing this crime. It is difficult to see this change as being driven by a correction in the required equitable compensation for damages. Thus, it looks like it may have been a response to the need for greater deterrence. To explore this, court cases involving runaway servants in Maryland are examined from 1653, the first court case of running away in the surviving records, to 1676, a full 10 years after the change in legal punishment, to explain the cause and effect of changes in Maryland’s runaway laws.

8.3 The Structure of Indentured Servant Contracts and the Incentive to Run Away

Immigrant indentured servitude was a form of long-term labor contracting, basically a forward-labor contract, similar to English apprenticeship contracts and life-cycle-servitude in husbandry agreements (Galenson 1981; Grubb 1985b, 2000a; Kussmaul 1981; Laslett 1971). A key differentiating feature of immigrant indentured contracts was that a large portion of the contract’s value was paid up front in the form of passage to the New World before any work was performed. Servants also received maintenance during the voyage and during the contract. To recoup this outlay, the labor contract had to extend over several years, typically 4 years for an adult (Grubb 1992a). In effect, the servant borrowed on his future labor a sum large enough to pay for passage across the Atlantic.

Prepayment of passage to the New World by masters may have been inescapable. Passage costs were sizable, equal to maybe a full year’s income (Grubb 1985a). Servants did not have enough accumulated savings, and it would take many years to accumulate enough savings to pay such a high passage cost. Borrowing on their future labor by signing an alienable servant contract was likely their only option to secure passage to America.

This large up-front passage fare payment could not be effectively countered over the first half of the labor contract’s length by requiring payment of a contract-completion bonus at the end of the contract. Such bonuses, required in colonial law, were called “freedom dues.” The servant had to pay for any freedom dues contracted or legally due by working for it; thus, freedom dues simply extended the contract’s length without altering the amount of labor time needed initially to repay the passage fare (Grubb 2000a).

When the lion’s share of payment is made prior to the execution of the labor portion of the contract, when it is what is called a “frontloaded contract” from the worker’s perspective, the result is an incentive by the workers to shirk or even breach the contract (Grubb 2000a). Why should the servant work hard or even stick around once the servant has received his passage to America? Having already been paid, i.e., having already been transported to America, the servant had an incentive to avoid the rest of the contract. The incentive was to run away and work for someone who would not have to deduct the cost of passage from the servant’s remuneration.

The economic approach to crime and punishment takes the level of punishment and the probability of capture (conviction) as substitutes in controlling the crime rate (Becker 1976: 39–85; Bodenhorn 2015: 90–3; Ehrlich 1973). The crime here is the unilateral breaking of a labor contract by running away. Instead of thinking in terms of the supply and demand of crimes, it is more cogent to think in derived demand terms of a production function of completed indentured servant contracts. The society of colonial planters produced completed indentured servant contracts through a combination of inputs that influences the servants’ behavior, namely, the likelihood of running away. Planters choose a set of inputs to minimize the cost of producing completed indentured servant contracts per an exogenously given payoff that the servant sees from a successful escape. Changes in the relative costs of these inputs, or changes in the payoff the servant expected from a successful escape, may explain changes in punishment levels inflicted on runaway servants who were caught.

Planters endeavor, through control of the legislature and to some extent the courts, to minimize the cost of producing completed indentured servant contracts. The production function of completed contracts contains a vector of inputs, including the probability of capture and conviction for running away, the degree of punishment for running away, and the expected alternative income if running away is successful. Planters will choose the combination of inputs to satisfy the first-order condition of cost minimization, namely input amounts will be adjusted until the ratio of cost-per-input-unit is equalized across all inputs.

Planters’ control or influence over the inputs and their relative costs varied by location. Both the marginal payoff and the cost per unit of the various inputs can be location specific, thus altering the mix of inputs used to produce the optimal amount of completed contracts by location. The particular combination of inputs used by Maryland planters, via their control of the legislature and courts, was effective in that the number of runaway servants out of the total population of servants was not large (Grubb 2000b; Smith 1947: 270, 278).

8.4 Application to Seventeenth-Century Maryland

The impression in the historical literature is that the problem of runaway servants was large, taking up a substantial portion of court time in seventeenth-century Maryland. In fact, for the period 1653 through 1676, there are only 39 prosecutions of runaway servants. This is a minute fraction of the court’s total time. Cases involving servants suing planters for nonpayment of freedom dues and other rights violations are more frequent (Grubb 2000a). Table 8.2 presents the runaway prosecution cases, their court locations, and the punishments handed down separated into the pre- versus post-1666 change in punishment laws.

Table 8.2 Prosecutions of runaway servants in Maryland, 1653–1676

The Maryland courts followed the letter of the law in meting out punishment except in Charles County after 1666. The number of prosecuted cases increases after the punishment was increased in 1666. More information is needed on the number of servants present in the colony before this increase can be used to infer a rise in the incidence or percentage of servants running away. The number of unprosecuted runaway servants mentioned incidentally in the court records may more accurately reflect changes in the incidence of running away before versus after 1666. Table 8.3 reports these unprosecuted cases of running away in the court records. This evidence indicates that the increase in punishment did not by itself cause an increased incidence of running away.

Table 8.3 Unprosecuted runaway servants in the Maryland court records

Changes in punishment were part of a wider integrated shift in the use of inputs to produce completed indentured servant contracts. One of the other inputs to achieving completed indentured servant contracts was increasing the probability of capture, thus lowering the incentive to run away. Schemes to increase the probability of capture in the seventeenth century were more difficult to devise than ways to increase the punishment level. Nevertheless, a flurry of acts were passed by the Maryland assembly between 1666 and 1676 that attempted to increase the probability of capture. Table 8.4 lists the legislative acts passed to aid in the apprehension of runaway servants.

Table 8.4 Legislative acts to aid in runaway servant apprehension

Changes in alternative income for successfully running away prior to 1666 was another event affecting the use of inputs to produce completed indentured servant contracts. In general, runaways would seek out other white settlements. The wilderness was not a serious option due to the threat from Native Americans and due to starvation. For the years prior to 1666, Virginia and Maryland were relatively isolated. Most runaways would stay within the two colonies or try to board ships leaving the Chesapeake. Many runaway servants were extradited from Maryland to Virginia. Prior to 1664, more runaways were extradited to Virginia from Maryland than were prosecuted as runaways from within the colony of Maryland. Prior to 1664, 14 runaway servants were extradited from Maryland to Virginia. From 1664 through 1676, none were. Also none were extradited to other colonies from Maryland prior to 1676 (Archives of Maryland vols. 1, 2, 4, 10, 41, 49, 53, 54, 57, 60, 65, 66). It may be safe to assume Virginia reciprocated and extradited Maryland runaway servants captured in Virginia back to Maryland.

After about 1660, a new destination for runaway servants appeared, namely, the Dutch-Swedish settlements on the lower Delaware River. Although there was some concourse between these settlements on the Delaware and the English colony of Maryland, the atmosphere was hostile between the two governments, even after the English officially took over these Delaware settlements in 1665. Systematic English government on the Delaware does not appear until 1676. Even then, New Castle, formerly New Amstel (on the Delaware River), court records from 1676 to 1681 reveal no cases involving the extradition to Maryland from Delaware of runaway servants from Maryland. In fact, as late as 1678, the colonists of Delaware were petitioning their governor for the liberty to trade with Maryland for the purpose of acquiring slaves, servants, and utensils.

Delaware was a frequently mentioned destination by Maryland runaway servants. The Maryland court records reveal two specific escape routes to Delaware. One route was up the Elk River and the other route was up the Choptank River. The mouth of the Elk River was close to Augustine Herman’s Bohemian Manor where a prison was established in 1669 to hold runaway servants apprehended nearby (see Table 8.4). The head of the Elk River was a short walk to the head of the Christina River which led directly to New Castle, Delaware. The head of the Choptank River was a short walk into the lower Delaware region. Apparently, escaped servants did not become servants in Delaware. Of the list of tithables for New Castle County in 1677, there were only 19 servants among the 307 inhabitants between the ages of 6 and 60.³ Table 8.5 lists the destination mentioned in Maryland court records for runaway servants.

Table 8.5 Destinations of runaway servants in Maryland records

Maryland formed a buffer between Virginia and the safe haven of Delaware. In a world where the standard form of apprehending runaway servants was through sending out a “hue and cry” through the population—neighbors passing information to the next neighbors—the probability of capture increases with the size of the friendly population between the master and the runaway’s destination. Maryland in effect was performing part of Virginia’s policing effort. Not only was there an increase in the alternative income available for successful runaways after 1660, namely, getting to the Delaware settlements, but the likelihood of successfully getting to Delaware was relatively greater for Maryland servants. Virginia servants faced a longer trek and a higher probability of apprehension because they had to traverse through Maryland to get to the safe haven of Delaware.

The increase in the alternative income occasioned a shift to other inputs such as punishment. The difference in the probability of capture between Virginia and Maryland runaway servants could explain the difference in punishment of their respective runaway servants. One implication of this view is that given some discretion in the county courts, the punishment would be tempered depending on the distance from Delaware, namely, on the probability of their runaway servants being apprehended. The interior counties would be expected to have less severe punishment than the counties on the Delaware frontier. Charles County, in the interior, never inflicted more than 27 lashes. Extra service was not added to the runaway’s contract. Charles County became less severe in its punishment of its runaway servants over the period. By contrast, Talbot County, a county on the Delaware frontier, inflicted the maximum punishment allowed in law in all runaway cases.

The increase in punishment and the increase in the desired rate of apprehension can be seen as part of the same concerted effort to offset the change in some other input to the production of completed indentured servant contracts. That other input was an exogenous change in the probability and payoff from a successful escape occasioned by the presence of a new safe haven for runaway servants on the lower Delaware River.

8.5 Escape and Apprehension Strategies in Mid-Eighteenth-Century Maryland

By mid-eighteenth century, Maryland’s laws to deter and punish runaway servants had become fixed, even ossified, no longer responsive to changing probabilities of escape and gains from escape. The Maryland Assembly saw fit to make no adjustments. However, those escape, and gains from escape, probabilities no longer varied substantially across colonies or by location within colonies. Maryland was now surrounded by other English colonies, i.e., Pennsylvania, Delaware, and Virginia, who advertised and returned servants who ran away from neighboring colonies’ masters. For example, Johann Carl Buettner, a German immigrant indentured servant, ran away from his master in eastern New Jersey in 1775. He made it through Pennsylvania and Maryland and was finally picked up and jailed in Norfolk, Virginia. The Norfolk sheriff advertised his capture in the Pennsylvania Gazette, and Buettner’s New Jersey master was able to recover Buettner from Virginia (Klepp et al. 2006: 186–9).

By mid-eighteenth century, Maryland’s method of informing the public of runaway servants and so detaining them had also changed from just “hue and cry” to using the colony’s weekly newspaper, the Maryland Gazette, to inform fellow colonists and county sheriffs of runaways. The Maryland Gazette was a weekly newspaper issued out of Annapolis. Issues exist from 1745 through 1789. The first page or two of each issue was devoted to news, much of it political, and the next two pages or so were devoted to advertisements. All sorts of things were advertised, including lots of runaways: runaway horses, runaway wives, runaway slaves, and runaway immigrant servants and transported convict servants.

A total of 1765 advertisements for runaway servants (netting out repeat ads) appear in the Maryland Gazette from 1745 through 1789. An advertisement for a runaway typically ran for several issues. Information was taken only from the first appearance of an ad. For an example of a first ad, see the following advertisement in the Maryland Gazette, Thursday, March 30, 1769:

March 29, 1769.

RAN away last Night from the subscribers, living on Kent-Island, Two Convict Servant Men, viz.

EDWARD PONTING, born in Bristol, about 25 Years of Age, 5 feet 6 or 7 Inches high, has a pert impudent Look, thin Visage, with brown curled Hair, is by Trade a Shoemaker, has some blue Marks on the Upper Part of his Hands, near the Thumbs, which are unknown: Had on, when he went away, an old bloom coloured Wilton Coat, spotted Flannel Jacket, a Pair of half worn Leather Breeches, old blue ribb’d Stockings, old Shoes, with plated Buckles, half worn Castor Hat, and a Check Shirt.

… [description of second runaway belonging to Jonathan Roberts] …

Whoever takes up and secures said Convicts, so that their Masters may get them again, shall receive, for each, Thirty Shillings, besides what the Law allows, and reasonable Charges, if brought home, paid by SAMUEL BLUNT [Ponting’s owner] …

Male servants comprised 95% of these ads. Convict servants comprised 48%, immigrant indentured servants 48%, apprentices 1.5%, and re-indented servants 2.5% of these ads. Among the 1085 ads that identified the servant’s ethnicity, 55% were English, 31% were Irish, 4% were Scots, 3% were Welsh, and 2% were German. Table 8.6 reports the counties of the masters of the runaways, from most to least runaways advertised. Three counties, Baltimore County, Annapolis, and Anne Arundel County accounted for over half the runaways.

Table 8.6 Month when the servant ran and the county of owner

Table 8.6 also reports the month the servant ran away based on the date advertised. Servants were more likely to run away in June through September, and less likely to run away in December through February. That servants chose the summer months and avoided the winter months to try and escape makes rational strategic sense on several levels. The opportunities to board a vessel to escape were higher in the summer than in the winter, and the ability to travel far and live off the land was greater in summer than in the winter. There was no nonrandom pattern in terms of what day in the month servants ran away.

Table 8.7 reports the day of the week servants chose to run away. Servants overwhelmingly chose to run away on Sunday. Given that most servants were given Sundays off as the Lord’s day-of-rest and/or to attend Church, running on Sunday likely maximized the amount of time before the master would detect that the servant was missing. Choosing to run on Mondays and Tuesdays was also above random choices, whereas choosing to run on Wednesdays through Saturdays was below random choices. Given that the Maryland Gazette was issued on Thursdays, and given the time it took masters to get a letter to the newspaper to advertise the runaway, running away on Sunday through Tuesday likely meant that the master could not get an advertisement about the runaway into that week’s newspaper. Thus, even if the master sent an ad in immediately after detecting that the servant had run away, the ad would not appear until the following week’s Thursday issue. This time span maximized the servant’s time on the run, a full week and a half, before the public was alerted to the runaway through newspaper ads. This pattern is consistent with servants exercising considerable strategic behavior to maximize their chances of escape and avoiding quick apprehension.

Table 8.7 Day of the week the servant ran away

Many masters dated the letter they sent to the Maryland Gazette; see the example above by the master Samuel Blunt. This evidence gives a sense of how long it took letters to get to Annapolis and then get the ad into the newspaper. Table 8.8 reports that evidence. Half of all the letters got to Annapolis in time to make that week’s newspaper, with 3 days’ time being the mode interval. Almost 80% of the letters got to the newspaper within 2 weeks, namely, by the next issuance of the weekly paper. This evidence indicates that masters did not wait long once detecting that the servant had run away, but more-or-less immediately sent a runaway ad to the Maryland Gazette. Given that masters could be compensated for the cost of placing an ad by having the servant’s labor time extended beyond that required in law meant that masters saw little differential loss from advertising runaway servants regardless of the servant’s contract type or labor value per unit time.

Table 8.8 Days between when the owner’s letter was posted and when the ad appeared and days between when the servant ran away and when the ad appeared

The incentive to advertise runaway slaves was different. Masters could not be compensated by the slave for the cost of advertising or apprehending the slave. Thus, slave owners had a different calculation, namely, they might consider the likelihood that a given runaway slave might return on his or her own, being just a brief absence for some reason, and so the master could avoid the advertising cost (loss). Only when it became clear the slave had permanently fled, and if the remainder of the slave’s life if caught was of greater value, then the master would invest in more advertising and apprehension expenses (Bodenhorn 2015: 93–5).

Given that masters could be compensated for the cost of advertising runaway servants via extra labor time added to the end of the servant contract, beyond the extra time added as set out in law, and so had an incentive to quickly advertise all runaway servants, the strategic behavior of servants in choosing what day of the week to run away makes sense. Servants wanted to maximize that Table 8.8 interval, and so try and push that interval into week 2, because they knew masters would advertise their escape as soon as it was detected. This pattern is internally consistent in a game-theoretic way between masters and servants and their incentives to react to each other’s behaviors.

Table 8.8 also reports the number of days between when the servant ran away and when the ad appeared in the newspaper. Almost 40% of the ads appeared within the week of the act of running away, namely, by the next newspaper issue after running away. The mode interval was 4 days. Almost 70% of the ads appeared within 2 weeks, namely, by the second issue after running away. Again, this evidence indicates that masters did not wait long after detecting that the servant had run away, but more-or-less immediately sent a runaway ad to the Maryland Gazette. Comparing the two columns in Table 8.8 indicates that servants gained some time by strategically choosing what day to runway before the public was alerted to their act through a newspaper ad.

The behavior of masters and servants as revealed through the newspaper ads for runaways still has to be traced through the county courts to see how punishment was administered and whether it varied by location and type of servant contract. The runaway rate for servants in mid-eighteenth-century Maryland is difficult to assess given restrictive arrival sample sizes, though it looks like the runaway rate was about 16% for convict servants and 6% for indentured servants (Grubb 2000b: 108). How many runaways were actually apprehended, and how many of those apprehended were prosecuted and punished across the county courts in Maryland, is a project for future research.

8.6 Conclusions

Immigrant indentured servants and transported convict servants had an incentive to breach their labor contracts by running away. The contracts were frontloaded and so servants had a lot to gain by early departure. Masters and servants engaged in strategic behaviors to deal with this contract breach incentive. Servants maximized their chances of escape, and master maximized their ability to thwart successful escape, given their constraints. In the seventeenth century, that behavior involved masters altering the colony’s statutory laws to mix together changing punishments and apprehension techniques to thwart servant escape. They adjusted these laws to changing locational opportunities of escape. Servants chose the best routes to escape that offered the best chance of not being returned to Maryland. Strategic behavior changed by the eighteenth century due to changing locational opportunities to escape and how runaways information could be delivered to the public. Masters quickly advertised runaway servants in newspapers, and servants selected when to run that delayed the appearance of those ads as much as possible.

Appendix: Grubb’s Murray Tribute

John Murray’s work on orphan apprenticeship in early America and my work on European immigrant indentured servitude in early America had much in common (for examples see Grubb (1992b, 2000a, 2006), Murray and Herndon (2002), and Murray (2013). Regarding these types of labor contracts, John and I often discussed contract structure, why particular contract designs were used, how contract compliance was enforced, and the behavioral incentives faced by masters, servants, and the government regarding contract performance. Typically, we discussed these issues in person at the annual meetings of the Economic History Association.

John was familiar with the work that is provided here through reading earlier working paper versions of it, but mostly through our in-person discussions of it over the years. He had encouraged me to publish it in some form. The initial working paper was written in 1980 as my first work on immigrant indentured servitude in colonial America when I was still a graduate student in economics at the University of Chicago. I sat it aside to work on other aspects of immigrant servitude. I returned to it often over the years, occasionally adding to it, but always setting it aside and not finishing it.

I aspired to use the paper as a vehicle for developing a complex mathematical model of crime and punishment, building on the work of my thesis advisor Gary Becker (1976: 39–85), and then use the colonial data to test the model. I hoped to place the piece in a general model-oriented economics journal. Alas, I have not been up to the task of mathematically modeling the issues as I envisioned them, nor do I think the data were strong enough to test the kind of model I had in mind. So while I frequently revisited the paper, I always sat it aside, waiting for better modeling inspiration that unfortunately (or fortunately) never came.

John liked the story and the data findings in the project and encouraged me to just publish the story and the evidence and not worry about trying to look modeling erudite for my economist peers. He felt that historians and economic historians would find the project interesting and a valuable addition to colonial labor history. As a tribute to his wise advice and, in a gentle way, his mentoring of an older scholar, I cleaned up the project, suppressed efforts to provide an explicit and original mathematical model of crime and punishment, and just presented the story and the evidence. Thanks John. I will miss our conversations.