1

Who Benefited From North American

Slavery?

U.C. Berkeley

IntroductionAsk a historian, or a political scientist, or a politician the question,

“Who benefited from North American slavery?” and the answer

you will probably get is, “The slaveholders, of course.” The

slaveholders got to work their slaves hard, pay them little, sell

what they made for healthy prices, and get rich.

We economists have a different view. Consider North American

slaves growing cotton in the nineteenth century. Those

slaveholders who owned slaves when it became clear that Cotton

would be King—that the British industrial revolution was

producing an extraordinary demand for this stuff and that Eli

Whitney’s cotton gin meant that it could be produced

cheaply—profited immensely as the prices of the slaves they

owned rose. But slaveholders who bought their slaves later on and

entered the cotton-growing business probably profited little if any

more than they would have had they invested their money in

transatlantic commerce or New England factories or Midwestern

land speculation: with the supply of slaves fixed, the excess profits

produced—I won’t say earned—by driving your slaves hard were

already incorporated in the prices you paid for slaves.

And there is another group who benefited mightily from North

American slavery: consumers of machine-made cotton textiles,

from peasants in Belgium able for the first time to buy a rug to

London carters to Midwestern pioneers who found basic clothing

the only cheap part of equipping a covered wagon. Slave-grown

Who Benefited From Slaver

U.C.

IntroductioAsk a historian, or a political scientist, or a politician "Who benefited from North American slavery?" and you will probably get is, "The slaveholders, of slaveholders got to work their slaves hard, pay them what they made for healthy prices, and

We economists have a different view. Consider slaves growing cotton in the nineteenth slaveholders who owned slaves when it became clear would be King—that the British industrial producing an extraordinary demand for this stuff and Whitney's cotton gin meant that it could cheaply—profited immensely as the prices of the owned rose. But slaveholders who bought their slaves later entered the cotton-growing business probably profited little more than they would have had they invested their transatlantic commerce or New England factories land speculation: with the supply of slaves fixed, the produced—I won't say earned—by driving your slaves already incorporated in the prices you paid

And there is another group who benefited mightily American slavery: consumers of machine-made from peasants in Belgium able for the first time to buy a London carters to Midwestern pioneers who found the only cheap part of equipping a covered wagon.

Economics 113, Spring 2018

2

cotton could be produced cheaply, yes, but the cotton-growers did

not collude and so sold their cotton at prices that incorporated only

a normal rate of profit. Cotton could be spun and woven by

machines at amazingly low prices, yes, but British factories did not

collude and sold their garments at prices that incorporated only a

normal rate of profit.

And there is yet a third group that benefited: northern and western

Americans whose taxes are lower because of the tariffs collected

on imports of goods financed by cotton exports.

A lot of surplus was extracted from North American slaves. And

the bulk of the surplus went to three places: to the pockets of those

who owned slaves when slave prices boomed as the extent to

which Cotton was King became clear, to those who

could—because of slavery—buy machine-made cotton textiles

much cheaper than would otherwise be the case, and to those in the

north and west of the United States whose taxes were lower than

would otherwise have been the case.

That’s the verbal preview of the economic argument. Now let’s

turn to the argument itself. It will proceed in three stages. First, it

will sketch out a picture of what the world in the first half of the

nineteenth century would have been like in the absence of North

American slavery. Second, it will sketch out a picture of what the

world actually looked at. Third, it will compare the two.

The World in the Absence of SlaverySuppose that there were no North American slavery in the first half

of the nineteenth century. Britain (and American) textile factories

would still have gotten the bulk of their cotton from the American

south: India was far and transport costs were high, and while Egypt

was closer its cotton-growing capacity was limited. The big

difference would have been that cotton would have had to have

cotton could be produced cheaply, yes, but the cotton-not collude and so sold their cotton at prices that a normal rate of profit. Cotton could be spun and machines at amazingly low prices, yes, but British factories collude and sold their garments at prices that incorporated normal rate

And there is yet a third group that benefited: northern Americans whose taxes are lower because of the on imports of goods financed by

A lot of surplus was extracted from North American the bulk of the surplus went to three places: to the pockets who owned slaves when slave prices boomed as the which Cotton was King became clear, to could—because of slavery—buy machine-made much cheaper than would otherwise be the case, and to those north and west of the United States whose taxes were would otherwise have been

That's the verbal preview of the economic argument. turn to the argument itself. I t will proceed in three stages. will sketch out a picture of what the world in the first half nineteenth century would have been like in the absence American slavery. Second, it will sketch out a picture of world actually looked at. Third, it will compare

The World in the Absence Suppose that there were no North American slavery in the of the nineteenth century. Britain (and American) would still have gotten the bulk of their cotton from south: India was far and transport costs were high, and was closer its cotton-growing capacity was limited. difference would have been that cotton would have had

2

3

been grown by free American farmers rather than by slaves, and

that the price of cotton would have to be high enough to offer those

who grew it earnings high enough to attract them out of northern

wheat farming.

Let’s roll all the costs of growing cotton (including any labor

engaged in supervision, storage, transportation, plus normal

profits, et cetera) into one variable: the (high) “wage” of American

labor growing cotton: wA. And let’s make the constant returns to

scale assumption: American land is free and American labor is

abundant so that as long as free cotton-growers in America can

earn wA more and more will enter the cotton-growing business, and

America can grow as much cotton at the price corresponding to

that wage as the world demands.

Now let’s move across the ocean, and make the same argument for

the factories in Britain that spin and weave the cotton into cloth.

British capital and labor are abundant, so we can once again

assume constant returns to scale. British costs of production can be

rolled into one variable, wB, the “wage” of British cotton-

processing labor ((including labor engaged in management,

construction of factories, machine-making, machine fixing,

storage, transportation, plus normal profits, et cetera).

The assumptions we have made tell us that the world supply of

cotton textiles takes a very simple form. It is, with P being the

price and Q being the quantity of cotton textiles produced in the

world:

P = wA + wB

If the price of cotton textiles were higher, American cotton-

growers would earn more than wA and British cotton-processors

would earn more than wB, more would enter the cotton-growing

and cotton-processing businesses, and production would expand. If

the price of cotton textiles were lower, American cotton-growers

would earn less than wA and British cotton-processors would earn

been grown by free American farmers rather than by that the price of cotton would have to be high enough to who grew it earnings high enough to attract them out wheat

Let's roll all the costs of growing cotton (including engaged in supervision, storage, transportation, profits, et cetera) into one variable: the (high) "wage" labor growing cotton: wA. And let's make the constant scale assumption: American land is free and American abundant so that as long as free cotton-growers in earn wA more and more will enter the cotton-growing America can grow as much cotton at the price that wage as the

Now let's move across the ocean, and make the same the factories in Britain that spin and weave the cotton British capital and labor are abundant, so we can assume constant returns to scale. British costs of production rolled into one variable, wB, the "wage" of processing labor ((including labor engaged in construction of factories, machine-making, storage, transportation, plus normal profits,

The assumptions we have made tell us that the world cotton textiles takes a very simple form. It is, with P price and Q being the quantity of cotton textiles produced worl

P = wA

If the price of cotton textiles were higher, growers would earn more than wA and British would earn more than wB, more would enter the and cotton-processing businesses, and production would the price of cotton textiles were lower, American would earn less than wA and British cotton-processors

4

less than wB, some would exit the cotton-growing and cotton-

processing businesses, and production would contract. The world

supply of cotton textiles is perfectly elastic at the price P =

wA+wB. The industry will supply whatever quantity Q is

demanded at that price. The supply curve is a flat, horizontal line.

Figure 1

World Supply of Textiles

(No-Slavery World)

In a world without slavery and with constant returns to scle in

growing and processing cotton, the world supply of cotton

textiles is a perfectly elastic horizontal line, with the price of

cotton textiles equal to the sum of the American cotton-

growing wage wA and the British cotton-processing wage wB.

Let’s also assume a particularly simple form for world demand for

machine-made cotton textiles. Let the demand curve be:

P = P0 - βQ

less than wB, some would exit the cotton-growing processing businesses, and production would contract. supply of cotton textiles is perfectly elastic at the wA-FwB. The industry will supply whatever quantity demanded at that price. The supply curve is a flat,

Price

wA+93

W

Figure World Supply

(No-Slavery

world sup p ty (no-slavery

In a world without slavery and with constant returns to growing and processing cotton, the world supply textiles is a perfectly elastic horizontal line, with the cotton textiles equal to the sum of the growing wage w,and the British cotton-processing

P P o

4

Quantity

Let's also assume a particularly simple form for world machine-made cotton textiles. Let the demand

5

If production of machine-made cotton textiles were very small,

then the (few) purchasers would be willing to pay some maximum

price P0: that’s the value of machine-made cotton textiles to those

who want them the most. As production expands, in order to sell

all the textiles produced they need to be sold to those who value

them less and less, with each unit increase in quantity produced

reducing the market price by the demand parameter β.

Figure 2

World Demand and Supply of Textiles

(No-Slavery World)

The equilibrium quantity—the amount of machine-made cotton

textiles produced—will be that which makes the price along the

demand curve equal to the supply price:

P0 – βQ = P = wA + wB

And so the quantity QF of cotton textiles produced in the no-

slavery world in which cotton is grown by free labor is:

I f production of machine-made cotton textiles were then the (few) purchasers would be willing to pay price Po: that's the value of machine-made cotton textiles who want them the most. As production expands, in order all the textiles produced they need to be sold to those them less and less, with each unit increase in reducing the market price by the demand

wA-1-vt W

Figure World Demand and Supply

(No-Slavery Price

•

world demand tor textiles tno -p=176•11

world trio-Mai/

Po - I3Q = P W A

ly ot '

Quantity

The equilibrium quantity—the amount of machine-textiles produced—will be that which makes the price demand curve equal to the

And so the quantity QF of cotton textiles produced in slavery world in which cotton is grown by free

6

QF = (P0 – wA – wB)/β

Now let’s calculate who gets the surplus from the cotton-growing

and cotton-textile industry in this no-slavery world. American

cotton growers earn the wage wA—but they could earn that same

wage working in New York or growing wheat in the Midwest: they

aren’t made better off by the fact that the cotton-textile exists.

(This is surely too strong a conclusion: their wage wA is probably a

small premium above the wage they would earn elsewhere. But the

major point is that they gain little, and that is surely true.) British

cotton processors earn the wage wB—but they could earn the same

wage elsewhere: the British economy is full of opportunities and

has a very flexible labor market. (Once again, the conclusion is too

strong, but the message—that British workers had ample other

opportunities and would have done almost as well without the

option of working in spinning mills—is almost surely true.)

Figure 3

Consumer Surplus in Textiles

(No-Slavery World)

QF = (PO - WA -

Now let's calculate who gets the surplus from the and cotton-textile industry in this no-slavery cotton growers earn the wage wA—but they could earn wage working in New York or growing wheat in the aren't made better off by the fact that the cotton-(This is surely too strong a conclusion: their wage WA is small premium above the wage they would earn elsewhere. major point is that they gain little, and that is surely cotton processors earn the wage wB—but they could earn wage elsewhere: the British economy is full of has a very flexible labor market. (Once again, the conclusion strong, but the message—that British workers had opportunities and would have done almost as well option of working in spinning mills—is almost

wA+w

Figure Consumer Surplus

(No-Slavery Price

*Odd dornarKi for 'MMUS ( 3 -siave p m i t y p

6

7

Now let’s look at consumers of cotton textiles. They get lots of

surplus. The first consumer gets (P0 – wA – wB) in surplus. The last

purchaser gets 0. The average purchaser gets (P0 – wA – wB)/2.

There are (P0 – wA – wB)/β purchasers. So total consumer surplus

is:

€

P0− w

A− w

B( )2

2β

Which you can also think of as the shaded green area in Figure 3

above.

Now let’s switch gears, and look at the world as it actually

was—the world with American slavery.

The World as It Was, with SlaveryNow consider the world as it was, with slavery. North Amerian

slaves—and there were, with the closing of the slave trade, a

relatively fixed number of North American slaves—could grow

enough cotton to produce a quantity QS of textiles. Assume that all

the slaves who could be deployed to cotton were, and that the total

production of cotton textiles with slavery was QS. Then the price of

cotton textiles will be the most that that quantity can be sold for:

PS = P0 - βQS

That’s the price at which British textile manufacturers will sell

their textiles. Because they are a competitive industry on the

buying side as well as the selling side, and because the quantity of

cotton that slaves can grow is fixed and limited, they will pay as

much as they can for scarce cotton imports. How much can they

afford to pay? Well, if they pay more than PS – wB they will lose

money and go bankrupt. So slaveholders will earn

Now let's look at consumers of cotton textiles. They get surplus. The first consumer gets (Po – wA – wB) in surplus. purchaser gets O. The average purchaser gets (Po – wA – There are (Po – wA – wB)/(3 purchasers. So total i

(P() - 142 A W

2

Which you can also think of as the shaded green area in abov

Now let's switch gears, and look at the world as was—the world with

The World as It Was, Now consider the world as it was, with slavery. slaves a n d there were, with the closing of the slave relatively fixed number of North American slaves—enough cotton to produce a quantity Qs of textiles. Assume the slaves who could be deployed to cotton were, and that production of cotton textiles with slavery was Q. Then the cotton textiles will be the most that that quantity can be

Ps = Po

That's the price at which British textile manufacturers their textiles. Because they are a competitive industry buying side as well as the selling side, and because the cotton that slaves can grow is fixed and limited, they will much as they can for scarce cotton imports. How much afford to pay? Well, i f they pay more than Ps – wB they money and go bankrupt. So slaveholders

7

8

PS – wB

For each unit of cotton they sell.

Figure 4

World Demand and Supply of Textiles

(Actual, Slavery World)

How is this PS – wB per unit of cotton divided up? Some—an

amount wS—goes to maintain the slaves. Some—an amount t—is

collected as tariffs on imports of goods from Britain that

slaveholders buy and goes to reduce the taxes of the United

States’s northern and western free-state citizens. And the rest:

PS – wB - wS – t

goes as profits to the slaveholders.

Or does it? The supply of slaves is fixed. Slaves are a durable

asset. If you get (PS – wB - wS – t) for each bale of cotton you

For each unit of cotton

PS –

Figure World Demand and Supply

(Actual, Slavery Price wond demand tor p = k 3 - 0 wor

supply textileI.

os Quantity

How is this Ps – wB per unit of cotton divided up? Someamount wS —goes to maintain the slaves. Some—an amount tcollected as tariffs on imports of goods from slaveholders buy and goes to reduce the taxes of States's northern and western free-state citizens. And

P S - W B - W S

goes as profits to the

Or does it? The supply of slaves is fixed. Slaves are asset. I f you get (Ps – wB - ws – t) for each bale of

g

9

extract from a slave, people who want those profits will bid up the

price of slaves. How far will they bid them up? Until the price of

slaves is such that it is no longer an extraordinarily good deal to

buy slaves and put them to work in the cotton fields. If r is the

appropriate rate of return on capital in the American economy, the

price of slaves will rise to:

(PS – wB - wS – t )/r

The capitalized value of the profits from slaveholding is thus

captured by those who owned slaves when slave prices rose, and

slave prices rose when people learned more about just how much

surplus there was from growing cotton and shipping it to Industrial

Revolution Britain. It was those who held slaves then who

benefited the most, not those who bought slaves at high prices and

put them to work later on.

So northern and western taxpayers gained as tariffs on imports sold

to slaveholders reduced their taxes. Those who owned slaves when

it became clear that Cotton would be King gained. How about

consumers—the buyers of cotton textiles?

Well, as long as PS is lower than wA + wB (and we know it isn’t: if

it were growing cotton with free labor would have been cheaper

than growing cotton with slave labor), consumers gain because the

consumer surplus triangle for the actual, slavery world is bigger

than the consumer surplus triangle for the counterfactual, no

slavery world. The difference is:

(wA + wB – PS)(QF+QS)/2

and can be seen to be the shaded area in Figure 5.

extract from a slave, people who want those profits will bid price of slaves. How far will they bid them up? Until the slaves is such that it is no longer an extraordinarily good buy slaves and put them to work in the cotton fields. I f r appropriate rate of return on capital in the American price of slaves will

(Ps w B - ws –

The capitalized value of the profits from slaveholding captured by those who owned slaves when slave prices slave prices rose when people learned more about just surplus there was from growing cotton and shipping it Revolution Britain. It was those who held slaves benefited the most, not those who bought slaves at high put them to work

So northern and western taxpayers gained as tariffs on to slaveholders reduced their taxes. Those who owned it became clear that Cotton would be King gained. consumers—the buyers of

Well, as long as Ps is lower than wA + wB (and we know it it were growing cotton with free labor would have than growing cotton with slave labor), consumers gain consumer surplus triangle for the actual, slavery world than the consumer surplus triangle for the slavery world. The

(WA + wB – Ps)

and can be seen to be the shaded area in

10

Figure 5

Gain to Consumers From Slavery

Now, as I said above, all these conclusions are too strong. British

wages were not completely unaffected by the existence of the

cotton textile industry. British manufacturers gained, the landlords

of Manchester gained, and British workers pulled into the booming

textile factories gained too. But the British textile-processing

industry (and the American cotton-growing industry) were nearly

competitive. And competitive industries do not capture economic

surplus—they pass it along, forward to consumers and backward to

the owners (in this case, initial slaveholders) of the factors of

production they must be.

So economists would say that the standard answer—that

slaveholders benefited from cotton-growing slavery—is largely

wrong. Initial slaveholders when it became clear how big a deal

cotton would be gained, but world consumers gained as well, and

so did those who benefited from lower American domestic taxes

Figure Gain to Consumers

Price world •Jomard

Fo

i

wostir tem

LaNr

Now, as I said above, all these conclusions are too wages were not completely unaffected by the existence cotton textile industry. British manufacturers gained, of Manchester gained, and British workers pulled into textile factories gained too. But the British industry (and the American cotton-growing industry) competitive. And competitive industries do not surplus—they pass it along, forward to consumers and the owners (in this case, initial slaveholders) of the production they

So economists would say that the standard answerslaveholders benefited from cotton-growing slavery—wrong. Initial slaveholders when it became clear how big cotton would be gained, but world consumers gained as so did those who benefited from lower American

11

made possible by high tariffs on imports financed by cotton

exports.

This is why we economists want to make every sociologist,

historian, and political science at Berkeley—and beyond!—to take

microeconomics, and then to take it again, and again. If you are

going to evaluate what’s going on in the economy, you need to be

able to make these “incidence” calculations. And only

microeconomic theory gives you the tools you need to do so.

You may complain: I’ve thrown a lot of algebra around (simple

algebra) but no real numbers. How important was all this, really?

Ah. You’re going to have to answer that. That’s what the next

problem set is about…

JBD

20050215

made possible by high tariffs on imports financed export

This is why we economists want to make every historian, and political science at Berkeley—and beyond! —microeconomics, and then to take it again, and again. I f going to evaluate what's going on in the economy, you need able to make these "incidence" calculations. microeconomic theory gives you the tools you need to

You may complain: I've thrown a lot of algebra around algebra) but no real numbers. How important was all Ah. You're going to have to answer that. That's what problem set

1

JE200502

121