So far, we have seen that the United States was an agrarian economy. Thomas Weiss tells us that about 75% of the American labor force was employed in agriculture in 1800 and about 55% was still in agriculture in 1860. (Hughes and Cain, p.115 - take special note of the second editor of the book in the citation for Table 6.9) From Callender, the population expanded westward and some specialization and division of labor took place. The west was growing in population extremely fast, going from 15% in 1810 to 46% of the American population by 1850. (Hughes and Cain, p.165) Why did people move west? Whether you consider the Northern or Southern pattern of westward movement, you will come to the same conclusion: people wanted to start farms on better or cheaper land. Lebergott examined two alternatives: fertilizing exhausted soil or moving on to buy new land to farm. It turned out that it was more expensive to try and fertilize old farmland than to simply abandon the old farm and buy fresh, fertile land out west. Lebergott quotes Thomas Jefferson: "We do not use manure because we can buy an acre of new land cheaper than we can manure an old acre..." (Lebergott 1985, p.191-192)
In the Northwest, families migrated to take advantage of cheap land in states like Illinois, Indiana, Michigan, and Ohio to start small wheat or corn farms. Meanwhile in the South, slaveowners usually took large workforces of slaves with them to fertile land in Louisiana, Mississippi, and Alabama to grow cotton on large plantations. Though there were some smaller cotton farms and some non-cotton farms, there were significantly more large plantations in the South than large farms in the North. (Hughes and Cain, p.168)
The 64 dollar question we want to ask is why Northern and Southern agriculture, and more importantly why their methods of farming, went on such different paths. The most interesting question is why the North consisted of farms that relied on family labor while the South used lots of slave labor.
The Harvest Labor Constraint is an idea explained in detail in Gavin Wright's The Political Economy of the Cotton South on p.43-55. The essence of the argument goes like this. Suppose a farmer is deciding how many acres of land to plant with crops. While it takes a considerable amount of effort to plough, seed, and weed the fields, the overriding concern is harvesting the crop. At harvest time, there is a huge spike in the amount of labor needed because the crop will go bad in the fields unless it is harvested within a very short window of time. The farmer knows there will be this harvest labor requirement spike, so he plants only as much crops as he thinks he will be able to harvest. Planting more than you think you can harvest is dumb, because rotten crops unharvested are of no value and take work to raise.
So the Harvest Labor Constraint says the upper limit of crop acreage you should grow depends on how much labor you have available at harvest time. Farmers would like to be able to plant more crops since more crops can be sold for more cash. Land is not a problem because land is cheap ($1.25 per acre of uncleared land from Federal sales). The problem is figuring out how to make sure you have enough labor to harvest the crops. How can the farmer find labor?
We will come back to slavery a little later. For now, consider the use of hired and family labor. Family labor is reliable and can be forced to stay and work. When planning on how much labor he will need to harvest his crops, the farmer can be sure he will have at least his own family's labor to draw on. Now let's consider wage labor, which is a totally different story.
Suppose the farmer hires one worker and plants his fields with the belief that he will have not only his family but this hired hand around at harvest time. The wage laborer might agree to one wage at the start of the season and then threaten to walk out right before the harvest, demanding higher wages. Would the farmer want to pay the hired hand the higher wage?
What would the hired hand demand? He would be able to demand the entire value of the extra acreage. The farmer, being the smart guy he is, realizes that any hired hand he might find would pull this stunt on him. So the extra acreage the hired hand would supposedly allow him to plant and harvest is worthless to the farmer (he ends up giving the entire value of that extra acreage to the hired hand!). Was this threat to walk out unless demands were met credible? You bet. There were tons of farmers looking for help to harvest their crops, and this guy could get his demands from any one of them so he had no reason to stay. Also, hired hands tended to have no families of their own and were single males who could wander with nothing to tie them down in one place.
The fact that hired hands cannot be trusted to stay - or even if they stay they might extort ultra high harvest wages out of you - placed a huge limitation on farmers. It did not make any sense to plant more acres than you could predictably harvest using your own family; wage labor could not relieve the Harvest Labor Constraint.
Northern and Southern farms differed in the way they dealt with the Harvest Labor Constraint. Consider Table 9.3 in Hughes and Cain (p.167). Northwestern states on the bottom half of that table were overwhelmingly mid-sized in the 50-500 acre range. Most of this had to do with the federal policy of selling land in 160 acre lots at first and then 80 acre half lots later. See Atack and Passell, p.277 for a breakdown of the costs of starting a 160 acre family farm in Minnesota.
Faced with unreliable wage labor, Northern farmers decided to simply have smaller farms. Another method used by farmers in the North to deal with the Harvest Labor Constraint was to grow some corn alongside their wheat. The advantage of corn is that it can be left out in the fields for a very long time and harvested at a leisurely pace; there is no harvest labor spike like you get with wheat. Why didn't everyone grow corn then? Because wheat was the cash crop - people wanted to eat bread, not corn, if possible. The status of bread has roots in Roman times (Hence the Roman Meal brand of bread and the line in Billy Joel's song about the girl who lives in a "white bread world.").
There are two advantages to using slave labor: the slave cannot leave and you can make the slave work as hard as you want them to (as long as you don't kill them by overwork). The fact that slaves are bound property and cannot leave solved the Harvest Labor Constraint for Southern planters. Need more labor? Just buy another slave.
The second aspect of slave labor is also critical for expanding your acreage of crops (like cotton) that have a harvest labor spike. (Hughes and Cain, p.170) It isn't enough to get more labor inputs available - you need more labor inputs crammed into a small harvest time window. The Harvest Labor Constraint says you should only plant what you can harvest, so having lots of labor squeezed into the harvest time window really lets you blow up the size of your plantation. The fact that slave owners could work their slaves ultra intensively during labor requirement spike periods made slave labor more efficient than free wage laborers even if those hired hands could be trusted. Free wage laborers were not willing to put up with gang labor at the wages being offered (Atack and Passell, p.305), so slave labor was the way to expand to maximum cash crop production.
First a quote from Gavin Wright's Political Economy of the Cotton South: "During a period of rapid demand expansion, Northern farmers increasingly pressed against labor constraints and searched for mechanical means of increasing acreage and output." (p.108) Why did the North mechanize while the South did not? Again, we come back to the particular solution each found to the Harvest Labor Constraint. Hiring more labor was not an option for Northern farmers, but Southern plantation owners could buy more slaves and expand their reliable labor supply at will. Northern farmers, then, needed to be able to substitute something else for labor - they needed to mechanize to make it possible for the same amount of labor to harvest more acreage. This is exactly what we are talking about when we say that increasing the amount of capital per worker raises the productivity of labor.
At the end of chapter 10, Atack and Passell go through David's analysis of the reaper. This is a standard marginal cost = marginal benefit deal: the farmer figures out how much labor he saves by using a reaper and compares the cost of that labor (in wages) to the cost of buying and maintaining a reaper. If the reaper saves more labor than it costs to maintain, the farmer buys the reaper. Since the value of the reaper depends on the savings from labor costs (higher wages make the reaper more valuable) as well as the benefits from the output the reaper is used to make (higher wheat prices make the reaper more valuable, as does better transportation to the East - remember Callender!), it is hard to tell what the reason is for reaper adoption in the early 1850s. It is clear, though, that Northern farmers had different incentives for mechanization than Southern planters based on their solutions to the Harvest Labor Constraint.
Why might Southerners not want to mechanize? Rich plantation owners had most of their wealth in the form of slaves, whose value was based on the demand for labor. (For those who have taken Econ 305, recall that the price of an asset is the net present discounted value of the future payment streams generated by the asset, so the price of a slave is the total value of crops/goods/services the slave would make over his lifetime.) If you introduce a machine that reduces the need for labor like the reaper, that reduces the value of slaves. Therefore, if you are a rich planter with lots of slaves, you really don't want mechanization because that destroys the value of the slaves your wealth is sunk into.
Back to my main page.
© 2002 Andrew N. Kato (04/27/02)