03 · David Ricardo · Thomas Malthus

David Ricardo

David Ricardo (1772–1823) made a fortune as a London stockbroker, read The Wealth of Nations on holiday, and became the most rigorous economist of his generation. His Principles of Political Economy and Taxation (1817) turned Smith's insights into deductive models: comparative advantage, differential rent, and a theory of how output splits between landlords, workers and capitalists. He used them in Parliament to argue against the Corn Laws. His friend and sparring partner Thomas Malthus supplied the population theory that made the whole system look gloomy. Below, the numbers are live: edit the labour table, grow the population, impose the duty, and watch who gains.
Ricardo's chapter 7 example. Portugal needs fewer hours than England to make both cloth and wine. It still pays for Portugal to make only wine and buy its cloth, because a cask of wine costs Portugal less cloth than it costs England. Edit the hours, move the price at which they trade, and watch each country's consumption leave its own production frontier.
Presets

Labour-hours per unit (edit me) · 3,600 hours of labour per country

countryhours / bolt of clothhours / cask of wine1 wine costs
England1.20 cloth → makes cloth
Portugal0.89 cloth → makes wine

Absolute advantage: Portugal is better at both goods. Comparative advantage is about the last column: what a cask of wine costs in cloth given up.

England · possibilities

0010102020303040405050cloth (bolts)wine (casks)autarkyproduceconsume

Portugal · possibilities

0010102020303040405050cloth (bolts)wine (casks)autarkyproduceconsume

Where the terms of trade must fall for both to gain

England 1.20Portugal 0.89p = 1.00
England gains
+9.1%
more of its autarky basket
Portugal gains
+5.9%
more of its autarky basket
world cloth
38.0 → 36.0
autarky → specialised
world wine
37.5 → 45.0
autarky → specialised
Inside the band: both countries end up outside their own frontier.

Grey dot: autarky, labour split evenly. Coloured dot: full specialisation. Green dashed line: what that output can be traded for at price p. Green dot: the same mix of goods as autarky, only more of it.

Try: with Ricardo's numbers, Portugal needs fewer hours for both goods, yet slide p anywhere between 0.89 and 1.20 and both countries gain. Push p past 1.20 and England would rather make its own wine. Then load "Same ratios": Portugal is twice as productive at everything, and the band collapses to a point.

Opportunity cost: the lawyer who types faster than her secretary

ricardo-opportunity-cost
The same logic for one person. A lawyer types twice as fast as anyone she could hire, so she has the absolute advantage at typing too. But every hour she types is an hour she cannot bill. The cost of her typing is the fees she gives up, and that is usually far more than a secretary's wage.

The lawyer's 50-hour week

lawyer50 h
law · 40.0 h
typing 10.0 h
secretary40 h available
not hired
cost of one typed page
lawyer: $25.00 (fees forgone)
secretary: $7.50
who has the advantage?
Absolute, at typing: lawyer
Comparative, at typing: secretary
her weekly net income
$8,000
hiring vs typing herself
+$1,400
per week, by delegating
break-even secretary wage
$100/h
billing rate ÷ typing speed
secretary earns
$0
the other side of the trade

Try: leave the lawyer twice as fast at typing and switch to "hires the secretary": she gains every week even though she is the better typist. Then drop her billing rate below twice the secretary's wage. Now typing herself pays, because the cost of her time is what she gives up, not how fast she is.

Many goods: wages decide where the chain is cut

ricardo-chain-of-comparative-advantagericardo-labor-theory-of-value
With two goods it is obvious who makes what. With many, rank them by how many Portuguese hours each takes relative to English hours. Wages translate hours into money prices: whichever country is cheaper for a good makes it. If England's wages are too high it sells too little abroad, gold flows out and wages fall, until trade balances. Within each country, labour costs set relative prices; between countries they don't, because workers do not move.

Ten goods, ranked by England's comparative advantage

0123PT hours ÷ EN hours3.20Engines2.5h / 8h2.00Clocks3h / 6h1.29Pottery3.5h / 4.5h1.00Cloth4h / 4h0.76Hardware5h / 3.8h0.53Grain6h / 3.2h0.44Leather8h / 3.5h0.33Silk9h / 3h0.25Olive10h / 2.5h0.17Wine12h / 2hrelative wage 1.50

A good is made in England when English hours × English wage ≤ Portuguese hours × Portuguese wage, i.e. when its bar rises above the wage line. Blue: made in England. Orange: made in Portugal.

England makes
2 goods
Portugal makes
8 goods
English exports
20.0
Portugal's spending on English goods
English imports
120.0
England's spending on Portuguese goods
England's trade balance
−100.0
deficit: gold flows out, English wages fall
equilibrium relative wage
0.76×
where the chain is cut

Try: start at a wage of 3×. England is so expensive it only makes engines, runs a deficit, and its wages must fall. Press "Let trade balance": the wage settles where the chain is cut, and the good at the cut is made in both. Now grow the English workforce: the equilibrium wage falls and England takes over more goods. A low-productivity country still trades, at lower wages. This is the reply to the "pauper labour" argument.

Differential rent: the margin sets the price

ricardo-differential-rentricardo-extensive-margin
Ten grades of land, each yielding less corn for the same dose of labour and capital. As population grows, farmers move down to worse land. The price of corn must cover the costs on the worst plot in use, which pays no rent. Every better plot then earns a surplus over that margin, and competition between farmers hands the surplus to the landlord as rent.

Ten grades of land · quarters of corn per dose of labour and capital

margin: 92 qr
A
+8
B
C
D
E
F
G
H
I
J
covers wages + profit (= the margin's yield) rent: surplus over the marginal plot not worth farming yet
plots farmed
2 / 10
marginal: B
corn price
£1.09
£100 per dose ÷ margin yield
total rent
8 qr
rent share
4%
wages+profit: 184 qr

As population grows

rent share %price ×10
0204060100200300400500600

Try: press "Grow population". Each time the best land can no longer feed everyone, a worse plot is brought in. The price of corn rises to cover its costs, and every better plot now earns a rent equal to its advantage over that new margin. Rent does not cause the high price; the high price, set on land that pays no rent, causes the rent.

Ricardo saw rent as a transfer, not a cost of production: it does not make corn dearer, it is the result of corn being dear. Henry George later built the case for a land-value tax on the same reasoning.

The Corn Laws: a tariff for landlords

ricardo-corn-lawsricardo-wages-profits
Britain's Corn Laws kept out cheap foreign grain. Combine them with differential rent and Ricardo's distribution theory. A duty raises the home price, so poorer land comes under the plough and rents rise on all the better land. Wages must still buy bread, so money wages rise too, and the manufacturers' profits that fund new investment are squeezed.

Where the nation's corn comes from

A
B
C
D
E
F
G
H
imports 24

Home plots are farmed while price × yield covers the £1,000 cost of a dose (wages plus capital). The rest is imported at £24 a quarter.

A · 100
£1.4k
B · 92
£1.2k
C · 84
£1.0k
D · 76
£0.8k
E · 68
£0.6k
F · 60
£0.4k
G · 52
£0.2k
H · 44
£0.1k
I · 36
idle
J · 28
idle

Who gets the national income (vs. free trade)

Landlords' rent
£5.8k +£4.9k
Wages (money)
£23.0k +£10.7k
Profits (farm + mill)
£45.0k −£6.7k

Dashed tick = free-trade level. Workers are paid a subsistence wage of 20 quarters of corn, as Ricardo assumed, so money wages rise with the price of bread. Mill owners sell cloth at world prices and absorb that rise out of profits.

price of corn
£24.0
free trade: £14.0
home plots farmed
8
free trade: 4
imports
24 qr
landlords' rent
£5.8k
mill profit
£40.8k
free trade: £48.8k
real wage
20 qr
subsistence, unchanged

Try: with the duty at £10, poor plots E–H are dragged into cultivation, rent balloons and mill profits shrink. Toggle "Repealed": imports flood in, the bad land goes idle, rent collapses and profits recover. Workers' real wage is unchanged either way in Ricardo's model; in practice cheaper bread was the popular core of the Anti-Corn Law League's case.

Malthus's argument as a simulation. Land is fixed, so each extra worker adds less food. When income per head rises above subsistence, population grows; when it falls below, population shrinks. Technology improves slowly. Then, at year 200, let technology speed up and decide whether fertility responds to income.
fertility falls as income rises
Scenarios

Income per head (food per person)

0.00.51.01.52.00100200300400yearsubsistencetakeoff

Population (index, year 0 = 1)

011220100200300400yeartakeoff
year
400
income / head
1.02×
× subsistence
population
1.4×
technology
1.2×

Food = technology × labour½ on fixed land, so each extra worker adds less than the last (diminishing returns). Whenever income rises above subsistence, population grows until income is pushed back down.

Try: in the Malthusian world, technology does improve, but the gain shows up as more people, not richer ones. Raise technology growth with fertility unchanged: people are a little better off, and there are vastly more of them. Income per head takes off only when technology outruns the fastest pace at which population can grow (try 3%), or when fertility falls as people get richer. History delivered both, which Malthus and Ricardo did not foresee.

For most of history the model fits: real wages in England were roughly flat for centuries, and they jumped after the Black Death killed a third of the population. It stopped fitting around 1800, which is roughly when Malthus and Ricardo were writing. Their gloomy predictions of a stationary state never came true, because sustained innovation and falling birth rates were outside their models.

Ricardian equivalence: is a bond just a postponed tax?

ricardo-equivalence
The state spends 100 now. It can tax 100 today, or borrow 100 and tax 100 plus interest later. A household that looks ahead sees that the present value of its tax bill is 100 either way, so it cuts consumption by the same amount and buys the bonds with what it saves. A household living hand to mouth spends whatever is left in its pocket. Mix the two.

The state spends 100 this year. Change in household consumption, year by year

financed by taxes nowfinanced by borrowing
-100-5000510152025yeartax of 163 to repay

Year 0, side by side

Tax now
consumption−53.3
private saving−46.7
government saving0.0
national saving−46.7
Borrow now
consumption−3.3
private saving+3.3
government saving−100.0
national saving−96.7
extra spending from borrowing
+50.0
year-0 consumption, debt minus tax
future tax bill
162.9
= 100 × 1.05^10, present value 100
national saving gap
50.0
debt crowds out more

Forward-looking households treat the bond as a tax bill postponed. They cut consumption by the same annuity either way, and buy the bonds out of the extra saving. Hand-to-mouth households spend whatever is in their pocket this year.

Try: set the Ricardian share to 100%. The two lines lie on top of each other, and borrowing adds nothing to spending. Drop it to 0%: taxing now cuts consumption by the full 100 at once, while borrowing pushes the whole hit to year 10, now grown by interest. Most economists, and Ricardo himself, think real households sit somewhere in between: liquidity constraints, short horizons and uncertainty keep them from fully saving for the future tax.

  • takeawayTrade pays whenever opportunity costs differ, not only when one side is better at something. Being worse at everything doesn't shut a country out; it shows up as lower wages, and trade still raises its real income.
  • takeawayRent is set at the margin. Protecting farmers from imports pushes cultivation onto worse land, raising rents for landowners at the expense of consumers and profits. The Corn Laws were a transfer to landlords dressed up as national self-sufficiency.
  • takeawayDiminishing returns are real for a fixed factor, so Malthus described the pre-industrial world well. The way out was ideas, not land. And whether public borrowing differs from taxation depends on how far ahead people look, which Ricardo himself doubted they do.

Key concepts · 17

David Ricardo · Thomas Malthus
  1. Comparative advantage

    A producer should specialise in what it gives up least to make, not in what it makes most efficiently.

    It shows both sides gain from trade even when one side is more productive at everything. What matters is the difference in relative costs, not in absolute costs. It is arguably the least intuitive and most robust result in economics.

    On the Principles of Political Economy and Taxation (1817), ch. 7

    England cloth / Portugal wineApple designing in California, assembling in China
    ↑ see it in the visualization
  2. Absolute advantage

    Being able to produce a good with fewer resources than someone else.

    Adam Smith's case for trade rested on it. Ricardo showed it is neither necessary nor sufficient. Portugal can be better at both cloth and wine and still gain by buying cloth from England.

    The Wealth of Nations (1776); contrasted in Principles (1817)

    ↑ see it in the visualization
  3. Opportunity cost

    The real cost of anything is the next-best thing given up to get it.

    An hour of a lawyer's typing costs her an hour of fees, however fast she types. Seeing costs as forgone alternatives is what makes comparative advantage click, and it applies to people and firms as much as to nations.

    The lawyer and her secretaryA surgeon who does not mop the operating room
    ↑ see it in the visualization
  4. Production possibility frontier

    The menu of output combinations a country can reach with its labour and technology.

    Without trade, a country can only consume on or inside its own frontier. Specialisation plus trade lets both countries consume outside their frontiers, and that is the gain from trade drawn as a picture.

    ↑ see it in the visualization
  5. Terms of trade

    The rate at which one country's exports exchange for its imports.

    Both countries gain only if the price lies between their two domestic opportunity costs. Where in that band it settles decides how the gain is split, but not whether there is one.

    ↑ see it in the visualization
  6. Labour theory of value

    Ricardo's view that goods exchange, within a country, roughly in proportion to the labour needed to make them.

    It gave him a simple yardstick for his models. Its limits show up in his trade theory: across borders, labour does not move, so prices are not set by labour content. Later economists replaced it with marginal utility and opportunity cost, and Marx took it in another direction.

    On the Principles of Political Economy and Taxation (1817), ch. 1

    ↑ see it in the visualization
  7. Chain of comparative advantage

    With many goods, rank them by relative labour cost; relative wages decide where the chain is cut between the two countries.

    It turns the two-good example into a general rule. A less productive country still exports something by paying lower wages, so trade does not require being good at anything in absolute terms. This answers the 'pauper labour' fear in both directions.

    Dornbusch, Fischer & Samuelson (1977), extending Ricardo

    ↑ see it in the visualization
  8. Differential rent

    Rent is the surplus a better plot of land yields over the worst plot in use.

    As population grows, worse land is farmed, the price of corn rises to cover its costs, and the owners of better land collect the difference. In Ricardo's words, corn is not high because rent is paid; rent is paid because corn is high.

    On the Principles of Political Economy and Taxation (1817), ch. 2

    ↑ see it in the visualization
  9. Margin of cultivation

    The worst land worth farming at current prices, which pays no rent.

    The cost of producing corn on the marginal plot sets the price for all corn. Marginal reasoning of this kind became the core method of economics after the 1870s.

    ↑ see it in the visualization
  10. Diminishing returns

    Adding more labour and capital to a fixed amount of land yields smaller and smaller increments of output.

    It drives both rent theory and the Malthusian trap. It holds for any fixed factor, holding technology constant, which is the assumption that later history broke.

    ↑ see it in the visualization
  11. Corn Laws

    British tariffs and import bans on grain, 1815–1846.

    Ricardo argued they forced cultivation of poor land, raised the price of bread and rents, and squeezed profits, the source of investment and growth. Their repeal in 1846 became the symbol of the free-trade era.

    Essay on the Influence of a Low Price of Corn on the Profits of Stock (1815)

    Importation Act 1815Anti-Corn Law League (Cobden & Bright)Repeal under Peel, 1846
    ↑ see it in the visualization
  12. Wages, profits and rent

    Ricardo's theory of distribution: the national product splits into rent for landlords, wages for workers and profits for capitalists.

    If wages must buy a subsistence basket of corn, dearer corn raises money wages and lowers profits. In Ricardo's account, landlords' interests ran against everyone else's: rising rents came out of profits and growth.

    On the Principles of Political Economy and Taxation (1817), ch. 5–6

    ↑ see it in the visualization
  13. Stationary state

    The end point where profits fall so low that accumulation, and therefore growth, stops.

    With diminishing returns on land and wages at subsistence, Ricardo expected profits to trend down. Free trade in corn and better farming could postpone it. Sustained technological progress turned out to prevent it altogether.

    ↑ see it in the visualization
  14. Principle of population

    Malthus's claim that population tends to grow faster than the food supply, so it is held back by misery or 'moral restraint'.

    It describes most of human history fairly well: gains in productivity became more people rather than higher living standards. It failed as a forecast after about 1800, when technology outran population and fertility fell as people grew richer.

    Thomas Malthus, An Essay on the Principle of Population (1798)

    Pre-industrial England's flat real wagesDemographic transition, 19th–20th c.
    ↑ see it in the visualization
  15. Malthusian trap

    A loop in which any rise in income above subsistence raises population until income is pushed back down.

    Escaping it needs technology to grow faster than population can keep up with. Fertility that falls as income rises makes the escape far easier and lasting, and in practice the two came together.

    Black Death raising wages (1350s)Industrial Revolution takeoff
    ↑ see it in the visualization
  16. Ricardian equivalence

    If households fully anticipate the future taxes needed to repay government debt, borrowing and taxing have the same effect on spending.

    It challenges the idea that deficits stimulate demand. Ricardo set out the logic in his Essay on the Funding System (1820) but doubted that people actually behave this way. Robert Barro revived it in 1974. Most evidence finds partial, not full, equivalence.

    Essay on the Funding System (1820)

    2008 US tax rebates (mostly spent by liquidity-constrained households)Barro, 'Are Government Bonds Net Wealth?' (1974)
    ↑ see it in the visualization
  17. Gains from specialisation

    When each country produces what it is comparatively good at, world output of every good can rise with the same labour.

    Trade is not a zero-sum contest for market share. It is a way of producing more with what exists. The gain comes from imports, which are what a country gets for its exports.

    ↑ see it in the visualization