Lecture notes:Ricardo

0: Introduction

• A. As you will see, Ricardo is fundamentally a theorist--a mathematical economist who doesn't happen to know any mathematics beyond arithmetic.
  • 1. His primary interest is not in the details of the real world, but ...
  • 2. In putting together a logically consistent theory, and ...
  • 3. He is willing to use unrealistic simplification where necessary.
• B. His introduction provides a fine example of courtesy and modesty
  • 1. He says admiring things about Smith and Say--both of whose ideas he is going to demolish various pieces of
  • 2. Concedes the imperfections of what he is doing,
  • 3. Leaves it to some abler man to complete the job.
  • 4. Unfortunately, there weren't any, so we had to wait another eighty years or so until Marshall et. al., building on the framework Ricardo had constructed, gave us modern economics.

Chapter I

I. Value:

• A. Use value vs exchange value
  • 1. Use value a necessary condition to have exchange value.
  • 2. Given that, exchange value depends on
    • a. Scarcity--goods with fixed supply, value determined by demand.
    • b. Quantity of labor required to obtain the good. Most goods.

    3. Ricardo is going to be discussing only the latter category--goods that can be produced, and whose exchange value thus depends on cost of production.

  B. Distinction between labor to produce and labor to earn the money to purchase.

  • 1. In a primitive society, exchangable value is proportional to labor input (Smith).
  • 2. Generalizing this gives us labor input as the measure of value.
  • 3. Smith, however, is using the labor cost of buying an input as his measure of value.
  • 4. The two would be equivalent if wages were always proportional to output, but they are not.
  • 5. Smith's standard--wages or corn--is itself a variable one.
    • a. What does that mean? Variable in terms of what?
    • b. Corn varies in the amount of embedded labor.
    • c. And wages vary with the cost of food.

    6. Suppose the labor cost of producing corn doubles:

    • a. Corn contains twice as much embedded labor.
    • b. In order for it to exchange for twice as much labor, the ratio of wages to corn would have to fall in half.
    • c. But it cannot, because workers were already near subsistence.
    • d. All of which proves, not that Smith was wrong, but only that measuring value by embodied labor gives a different result than measuring it by labor to earn the money to produce.

    7. Ricardo is sneaking up on his rent theory--note "land last taken into production." The fact that the amount of corn produced by a given amount of labor varies with the fertility of the marginal land is going to be key to his analysis.

  • 8. And he, like Smith, is using a "wages determined by subsistence" theory.

  C. Which standard of value is right?

  • 1. Ricardo suggests that if the relative value of commodity A changes relative to B, we should compare both to C-Z. If A's relative value has stayed the same with regard to C, D, E ...Z, it seems reasonable to say that it was B that changed.
  • 2. Especially if we observe that the imputs needed to produce A and C-Z have all stayed the same, but there has been a change in the inputs needed to produce B.
  • 3. So what Ricardo is proposing is not a definition of value used to compare how well off people are in different times and places, but ...
  • 4. One intended to predict relative prices.
  • 5. Which Smith's definition of value makes no claim to do. He has a price theory, based on cost of production including labor, but that is a different part of what he is doing from his definition of value.
  • 6. According to Smith (and Malthus), if the price of gold drops relative to other things because producing it get easier, that is a drop in its value, but ...
  • 7. If the price of labor drops relative to other things because producing it (i.e. growing corn) gets easier, that means everything else has gone up in value.
  • 8. Ricardo points out an inconsistency in Smith, given that laborers do not spend all of their income on corn:
    • a. Laborer gets one bushel/week when corn is 80s/quarter, 1 1/4 when it falls to 40s.
    • b. Consumes 1/2 bushel, exchanges the rest for other things.
    • c. Suppose his remaining 1/2 bushel at the old price buys him more goods than his remaining 3/4 at the new (as it will if other prices don't change)
    • d. His wage has gone up measured in corn, down measured in what he can buy with it.
    • e. But in Smith's sense, the worker's wage is always of the same value, by definition--1.
    • f. Corn comes in only as a useful approximate measure of value, not as a definition of value.
    • g. And Ricardo has merely pointed out a special case where the approximation breaks down.
• II. Different qualities of labor.
  • A. Obviously, some workers have more skill, work harder, etc.
  • B. Over time, relative wages reflect that--one is producing one hour per hour, one two, so to speak (my way of putting it, not Ricardo's).
  • C. Ricardo is interested in explaining relative prices, so ...
    • 1. Provided that the relative productivity of different kinds of labor stays roughly the same over time
    • 2. (And Smithsays it does.)
    • 3. He can ignore it in what he is doing.

• III. Labor applied indirectly as well as directly affects value.
  • A. Even in the primitive state, some capital is used. So the value of deer and beaver is affected by the labor cost of producing the weapons to hunt them.
    • 1. If one weapon was took much more labor to make than the other, that would make that animal more valuable.
    • 2. Or if they were equally costly to make, but one wore out much faster.
    • 3. What if the weapons belonged to different people than the ones using them?
      • a. Supply and demand for capital and labor would affect the division between the two groups of people,
      • b. But not the relative prices of the goods?
      • c. Is that true? We haven't yet gotten to profit as payment for time value of capital.
      • d. And when we do, it won't be true any more.
      • e. But at this point, the cost of using the capital good is merely the amount of embodied labor used up--and employers can hire people to make bows just as cheaply as to hunt deer.

    B. In a more advanced society, tracing out the labor used to produce a final good is more complicated, but the logic of the problem is still the same.

    • 1. As Ricardo claims to show by pointing out that a reduction in the labor used in one stage of production would reduce the exchange value of the good, but ...
    • 2. That would still be true if, as in Smith, exchange value depended on a sum of costs, of which labor was one.

    C. Going back to the primitive society, Ricardo sketches an equilibrium price argument for why relative prices must be equal to relative amounts of embodied labor.

  • D. Note also that, implicitly, wages and profits move in opposite directions in this simple model.
  • E. But it is still unclear what the capitalist is contributing to the process, for which he gets his profit.
  • F. If we had an invariable standard (what does that mean?), it would show us that changes in exchange value always reflected changes in labor input.

• IV. But the conclusion changes if we allow for fixed capital.
  • A. Once we include fixed capital, now relative prices may be changed by changes in the wage rate (not obvious--but true).
  • B. We should distinguish between fixed and circulating capital
    • 1. Noting that the latter has varying production periods.
    • 2. That the distinction is not a sharp one.
    • 3. Since a circulating capital with a long period is like a fixed capital.

    C. If all capital were circulating, and all had the same production period, then exchangable value would be proportional to labor input.

    • 1. Or, in modern terms, if all goods were produced with the same labor/capital ratio, then...
    • 2. All prices would be proportional to labor input.
    • 3. But, of course, under those assumptions labor input is proportional to capital input, so ...
    • 4. All prices are proportional to capital input too!
    • 5. Which is not precisely a labor theory of value, although it looks like one in a bad light.

    D. But if that is not the case, then:

    • 1. If Barley and Oats are produced with the same K/L ratio as each other, and so are cotton goods and cloth
    • 2. Variations of wages will leave the relative prices of Barley and Oats the same, leave the relative prices of cotton goods and cloth the same, but change the relative price of Barley (and Oats) to Cotton goods (and cloth).
    • 3. Ricardo sees the idea of labor capital ratio, but has no words to explain it in:
      • a. Since "capital" is not the amount of capital going into the production but
      • b. Capital years used in production, i.e.
      • c. The amount of interest that must be paid on the capital used to produce a unit of output.
      • d. But he recognizes that amount of fixed capital and slowness of circulation are really the same issue.

      4. Example using machinery

      • a. One man uses 100 workers for a year to grow corn, two others use the same labor to make one machine each.
      • b. Next year, one machine +100 men makes cloth, ditto makes cotton goods, and the farmer uses his 100 men to grow more corn.
      • c. If we looked only at embedded labor, the cloth in the second year costs200 man years=cotton goods=corn crop, since the machine in this example is not wearing out, so none of its embedded labor must be credited to its output, but ...
      • d. The farmer got his capital back after one year, the other two used both their first and second year capital in their second year production (and will use the first and third year capital in the third year's production, and...), so they have to get more profit to compensate, so the price of what they produce must be higher.
      • e. He then runs it through with numbers.
        • i. Wages of £50/annum, so £5000 of capital was employed by each of them; assume a profit rate of 10%
        • ii. Corn, at end of the first year, is worth £5500, second harvest the same.
        • iii.Men with machines have £5500 in the machine at the end of the first year, must receive 10% on that plus their second £5000 of wages, or 10%x£10500=£1050 in profit. Assuming the machine is infinitely durable (which Ricardo forgets to mention) their goods must sell for £6050, even though none of the machine's embedded labor is being used up--and continue to sell for that each year thereafter.

      5. So far, this gives us different ratios of exchangable value to embedded labor, depending on the capital labor ratio.

    • 6. It also implies that relative prices will change, technology held constant, as wages change.
      • a. Note that Ricardo is considering changes in the division of output between the worker and the capitalist
      • b. Since this is a change in wages with technology held fixed.
      • c. So it is just a question of who gets what share of the output.

      7. If wages go up, labor intensive goods rise relative to capital intensive goods. But...

    • 8. This effect is small
      • a. Ricardo guesstimates that plausible variations in the profit rate could not give fluctuation in relative prices of more than about 6 or 7 percent
      • b. From which Stigler gets "the 93% labor theory of value."

      9. But changes in the inputs needed to produce outputs can result in much larger changes in relative prices.

• V. Allowing for fixed capital of variable durability has a similar effect.
  • A. Worked through once for a machine that requires labor input to keep it up--just like labor used directly in production.
  • B. And a second time by observing that a machine that wears out in a year is just like circulating capital with a period of one year.
    • 1. Actually, both Smith and Ricardo routinely ignore the fact that wages are paid out during the year, not all at once.
    • 2. Possibly, in Smith's case, because the harvest comes in once a year and must be stored until eaten. So somebody is paying the interest on that capital until the workers get it.

    C. Consider a machine that lasts only a year, substitutes for 100 men, and costs £5000.

    • a. If wages rise, what happens to its price?
    • b. It must embody less than £5000 in wages, plus some interest, and ...
    • c. If the rise in wages is balanced by a fall in interest, its price will stay the same.
    • d. And that will happen if it is made with the average labor/capital ratio (he is getting ahead of himself here).

    D. Conclusion:

    • 1. Under primitive conditions, prices are proportional to labor.
    • 2. With fixed capital etc., ratio of price to labor input can vary quite a lot.
    • 3. And relative prices can change slightly due to changes in wages and profit, without any change in production technology.
    • 4. With rising wages lowering the relative price of capital intensive goods.

• VI. We need an invariable standard of value.
  • A. It must always require the same labor input--which real monies don't.
  • B. But that would still be a perfect measure of value only for things produced with the same labor/capital ratio as the standard. but...
  • C. Since the change in relative value of goods with different K/L ratios due to changes in wages is small, we can (and he will) ignore that problem.
  • D. And he will write as though money was that invariable standard.
  • E. Note the paradox in Smith's claim that if wages rose all goods must rise in price. What about the price of the money in which price is measured?

  VII. Changes in price due to changes in the value of money

  • A. Do not change distribution between profit and wages
  • B. But changes with value of money held constant do.
  • C. i.e. changes in wages given that the amount of capital and labor used to produce a unit of money is unchanged.
  • D. Note that wages are measured here by labor needed to produce what the worker can buy with them--i.e. they measure the share of output going to the worker, not his welfare. Ricardo abandoned that possibility when he scrapped Smith's version of the labor theory of value.

Chapter II: Rent

• I. We must distinguish rent--the payment for the use of the original and indestructible powers of the soil, from other payments miscalled rent.
  • A. Including the profit on the landowner's investments in improving the land, and ...
  • B. Payment for timber on the land--it would be rent only if you were paying for the right to grow timber and harvest it, not just for the right to harvest the timber already there.
  • C. Rent of coal mines or quarries--where, again, you are removing something permanently.
  • D. Ricardo will use "rent" only in the narrow sense.

• II. The development of rent
  • A. If land, like air, were available in unlimited supply, it would be free.
  • B. It commands rent only because it is not unlimited in extent and uniform in quality--the best land is all in use.
    • 1. When land of the second quality goes into use, there must be rent on land of the first quality
    • 2. And its amount depends on the difference in quality.
    • 3. And similarly with the third.

    C. More precisely, suppose three qualities, yielding, with the same inputs of labor and capital, net produce (after supporting the laborers) of 100, 90 and 80 quarters of corn.

    • 1. While there is plenty of quality 1 land, the farmer will get all of the output to pay his profits.
    • 2. But when quality 2 goes into cultivation, rent appears on quality 1, since ...
    • 3. All corn sells for the same price, and that price must be sufficient to pay for the profits of the farmer farming quality 2 land, leaving the difference as rent for the owner of the quality 1 land--10 quarters.
    • 4. And rent goes to 20 on quality 1 land and 10 on quality 2 land when the third quality comes into cultivation.

    D. But there is an intensive margin as well--before the really bad land is cultivated, it becomes worth applying more labor and capital to the good land to get additional output.

    • 1. Suppose quality 3 is in production, and that doubling the input on quality 1 raises net output from 100 to 185. It is worth doing (cheaper than cultivating quality 3), and rent on quality 1 is now 25 quarters.
    • 2. If quality 3 is not in production, and the second dose on quality 1 is the marginal input, then quality 1 gets 15 quarters rent, quality 2 gets 5.
    • 3. Price equal marginal cost, rent is output where cost is lower than on the margin--does this start to look familiar?
    • 4. The last applied capital pays no rent--i.e. P=MC!

• III. So the reason why the value of corn rises over time is not that the rent goes up, but ...
  • A. Value goes up because the labor cost of producing one more quarter of wheat goes up
  • B. Which causes rent to go up. "Corn is not high because a rent is paid, but a rent is paid because corn is high."
  • C. And Smith was wrong to include rent in cost!
  • D. It is said that land is an especially good thing because it generates rent--but
    • 1. That is a result of the limited supply of land--we would be better off if we had lots more land, hence paid no rent.
    • 2. And worse off if air, water, etc. became more scarce and started to pay rent.
    • 3. If the surplus produce which rent affords is an advantage, we should make our machinery worse each year--giving a rent on the old machinery.

    E. The increase of rent is a symptom of wealth, not a cause.

  • F. The countries in the best position to get wealthy are those that have lots of fertile land, or can import food freely, or have a technology that lets them increase agricultural output per acre readily at constant cost--all circumstances that lead to low rent.
  • G. Smith claims that the high price of corn is the effect of rent, but ...
    • 1. The price of corn is determined by the cost of production on marginal land
    • 2. Which pays no rent.
    • 3. Hence the existence of rent does not affect exchange ratios.
    • 4. (but the existence of the circumstance that leads to rent--scarcity of first quality land--does affect exchange ratios--via its effect on labor cost of producing on marginal land)

    H. A reduction in national capital, and consequent reduction in population, would push the margin towards more fertile lands, reducing rent.

  • I. And an improvement in agricultural technology which made land more productive, permitting us to take the less productive land out of production, might make rent go down, in the short run (capital held constant).
    • 1. At which point wages as a share of output would fall (iron law)
    • 2. Profits would increase
    • 3. Accumulation of stock would increase
    • 4. Eventually, the old marginal land would be back in production, and rent back up.
• IV. Distinguish between improvements in agricultural technology that give more output and those that reduce needed inputs (of labor).
  • A. Improvements such as better crop rotation give you a larger yield from any piece of land, labor held constant.
    • 1. If the result is to increase output, but keep the difference in output between successive doses of capital and labor the same ...
    • 2. Then the marginal dose comes with less capital and labor, and rent is lower. (p. 55)

    B. Improvement such as better plows give us the same output from the land at lower cost in labor (and capital)

    • 1. The same amount of land is cultivated
    • 2. If the difference in output from each dose of labor + capital remains the same, while the doses get smaller, then rent stays the same.

    C. Note that we are measuring rent as proportion of output, not as exchangable value.

    • 1. But things that push us to produce on worse land both raise rent and ...
    • 2. Raise the exchangeable value of corn
    • 3. Thus doubly benefitting the landowner.
    • 4. This assumes that the increased difficulty of production is due to shifting the margin of production, not to something that decreases output everywhere (Japanese beetles, say). That might raise or lower rent, depending on how it decreased output.

Chapter III

• I. In the case of mines as well, high rent reflects the high value of the produce, not the other way around:
  • A. If there were an unlimited supply of equally fertile mines
    • 1. the value of the produce would be the amount of labor necessary to produce it, and ...
    • 2. No rent would be paid.

    B. With mines of varying fertility

    • 1. The marginal mine must cover the cost of labor and capital
    • 2. So the marginal mine's costs regulate the price
    • 3. And all better mines pay rent.

    C. If the labor cost of producing gold from marginal mines was always the same, gold would be as nearly invariable a measure of value as is possible. Ricardo will write as if this were true--i.e. use "gold" to mean an imaginary standard having this characteristic.

• IV. Note that throughout this discussion, Ricardo is ignoring his earlier point that revenue from an extractive industry is not really rent.
  • A. If a mine eventually runs out, part of the cost of getting gold now is not getting it later.
  • B. If mines have unlimited amounts of ore, why don't you produce everything you want from the most fertile mine?
  • C. The implicit model seems to be a mine that can produce at a certain rate forever, with a fixed input required for each unit of output--the lower the ratio of input to output, themore fertile mine.
  • D. Which lets Ricardo apply is analysis of rent to mines, but isn't ver realistic.
  • E. On the other hand, doing it right probably requires Hotelling's analysis of depletable resources, which isn't going to be written for another century or so.

Chapter IV

• I. Of course, market price is not always equal to natural price.
  • A. Ricardo repeats Smith's argument for how equilibrium in the capital market moves market prices towards natural prices.
    • 1. Goes through a little more detail--borrowed capital shifting from one manufacturer to another.
    • 2. Observes that it works better than might be supposed.
    • 3. And notes non-pecuniary as well as pecuniary costs and benefits.

  II. And refers the reader back to Smith.

Chapter V

• I. The ability of the laborer to support himself and bring up the children needed to maintain the working poulation depends not on his wages but ...
  • A. on what his wages will purchase.
    • 1. So the natural price of labor depends on the price of foods, necessaries, and customary conveniences.
    • 2. And with progress, tends to rise, because corn gets more costly as we push to worse marginal land, but ...
    • 3. The effect might be temporarily reduced by improvements in agricultural technology
    • 4. As happened, temporarily, for 180 years after Ricardo published!

    B. The natural price of goods other than raw produce and labor tends to fall over time

    • 1. Since the increase in the natural price of the raw material that goes into them
    • 2. Is outweighed by improvements in technology, etc.
    • 3. This is apparently an empirical claim.

    C. The market price of labor, as of other goods, can deviate from its natural price, but tends to conform.

    • 1. When market price is above natural, laborers are well off.
    • 2. When below, laborers are wretched.

    D. Market rate could be above natural rate in an improving society for an indefinite period of time

    • 1. Since capital might keep accumulating
    • 2. As in fact happened from then till now.

    E. In talking of capital accumulating, one should distinguish between

    • 1. An increase in the value of capital--which might happen because the things making it up became more valuable (i.e. required more labor to reproduce)
    • 2 And an increase in the quantity of capital.
    • 3. If capital increases in quantity and value, the natural price of labor rises, due to increased value of the stuff needed to support it.
    • 4. Capital might increase while its value held constant or fell, in which case the natural price of labor would be constant or declining.
    • 5. In both cases, market price of labor would be above natural price.
      • a. First case, only by a little, so soon back to natural price
      • b. Second by a lot, since his wages have gone up and the cost of subsistence is constant or falling. So it may take a long time for market wage to get back down to natural wage.

  II. The natural price of labour, measured in food etc., is not fixed

  • A. It varies across time and space
  • B. because it depends on the habits of the working population
  • C. The more they require to feel they are doing well enough to afford to marry and have children, the higher the natural wage will be.
• III. Wages are affected by two factors
  • A. Supply and demand of labor
  • B. Price of what laborers buy.
  • C. If lots of fertile land is available, capital may accumulate faster than labor.
    • 1. Mimimum doubling time for population has been estimated at 25 years, but ...
    • 2. Capital might double faster than that under favorable circumstances
    • 3. In which case wages would go up.

    D. In new settlements, this happens--until they get pushed onto worse land.

  • E. Contrast countries with lots of fertile land but bad institutions with countries which are fully developed.
    • 1. In the first case, introduce good government and capital will rapidly accumulate, making people better off.
    • 2. In the other, accumulation of capital will simply be matched by new population.

    F. The friends of humanity cannot but wish ... that workers should have luxurious tastes.

• IV. As capital accumulates, what happens to wages?
  • A. If we look only at the supply/demand side, we observe that capital will accumulate more and more slowly (he hasn't shown this yet, but will), so wages will sink--i.e. be closer and closer to the wage that gives constant population.
  • B. But the prices of necessaries are rising (because they are being produced on worse and worse marginal land), so the wage (measured in Ricardo's imaginary money--i.e. value) that gives constant population is rising.
  • C. So money (i.e. value) wages rise, but real wages (in our sense--what the laborer can buy) fall.
  • D. But the rise in money wages will lower the profit rate--as we will see later.
  • E. So both rent and wages increase as capital and population accumulate, but ...
    • 1. The increase in rent involves the landlord getting an increasing share of agricultural output--and selling it at a higher price per unit
    • 2. The increase in wages involves a decreasing amount of corn--at a higher price.
    • 3. The difference between the two situations being that the landowner is selling corn, the laborer is buying it, and the value of corn is rising.

    F. All of these calculations are done in terms of Ricardo's imaginary gold of fixed value, but the argument does not depend on that.

    • 1. One might think (Smith apparently does) that an increase in wages would raise the price of all commodities.
    • 2. But gold is a commodity--and prices are measured in it. Its price cannot rise.
    • 3. To put it differently, if prices of goods all go up we will need more gold to circulate our goods, but ...
    • 4. If prices of all goods in gold go up, the price we are paying for gold in goods has gone down, so how can we import gold?
    • 5. So the argument works for prices in real gold as well.

    G. Digression on the poor laws, citing Malthus:

    • 1. Describes the bad consequence, but doesn't run through the argument--number of people on welfare (in our terms) would rise until it absorbed all of the country's net revenue save that used by the government
    • 2. But immediate abolition would have terrible consequences.
    • 3. Hence gradual abolition the solution.
    • 4. Sound familiar?
    • 5. Shifting from a local to a national poor law (which is suggested by Smith's attack on the settlement laws, although I don't think he actually proposes it) would be a bad idea
      • a. Because local administration and payment gives each parish an incentive to collect the money as efficiently as possible, and
      • b. Spend it as sparingly as possible.
      • c. Which is why welfare is not yet destroying the country

      6. If everyone could live comfortably on welfare, the principle of gravitation is not more certain than the tendency of such laws to convert wealth into misery.

    • 7. Presumably the part of the argument we would not see in modern discussions of the problem of welfare is a population argument--by supporting the unemployed poor we keep up the laboring population, thus depress the wages of the employed poor.

Chapter VI

• I. What determines the average rate of profit and interest?
  • A. The higher wages (measured in labor--hence share of output) are, the lower profit is.
    • 1. Fairly obvious for the manufacturer, what about the farmer?
    • 2. He is paying more money for his labor, but selling his corn for more money, but ...
    • 3. He is either producing less corn per unit of labor (on marginal land) or paying more rent (on more fertile than marginal land).
    • 4. Since money is assumed of constant value, the amount for which you can sell the output of an hour's labor on marginal land is always the same. Ricardo goes through some numbers on this.
      • a. The farmer on marginal land is getting the same amount of money, paying more in wages, so has less profit.
      • b. And the farmer on non-marginal land is paying the difference between his output and the output of the farmer on marginal land to the landowner as rent.

      5. Note that all of this increase in wages is with technology fixed--output per worker is the same, save that we are moving to worse marginal land, so output of corn per worker on marginal land is less, but ...

    • 6. That assumption is not essential, because Ricardo is measuring everything in a money whose value (labor input) stays constant, so measured in that money output per worker is always the same.

    B. So the farmer has an interest in keeping rent low

    • 1. Not because he pays it--he really doesn't (in value terms--go through Ricardo's analysis)
    • 2. But because high rent means high value of corn means high (value of) wages means low profits.
    • 3. And even lower profit measured as a rate of return rather than a value, since higher value of agricultural output means higher value per unit of his capital stock, means more value per worker employed.
    • 4. So he is getting fewer £s of profit on more £s of stock.

    C. Effect on other goods:

    • 1. Most goods have some agricultural output as inputs, so will rise in value along with the increase in the value of corn, although not by as much.
    • 2. But they are rising because they have more labor expended on them, not because the labor is more expensive.
    • 3. Since, after all, the money they are measured in (by assumption) is also being made with more expensive labor.
    • 4. And, measured in that money, profit falls as wages rise.
    • 5. All of this assumes that when corn becomes more expensive, wages rise
      • a. Which doesn't have to happen in the short run, both because it takes time for population to adjust and because the workers might be making above the natural wage to start with, but ...
      • b. In the long run ... .

    D. The same thing would happen if there was an increase in the price of other goods consumed by the laborer--profits would fall.

    • 1. But not if there was an increase in the price of luxury goods.
    • 2. Which labourers do not consume.

    E. All of this describes equilibrium.

    • 1. In the short term, some profits might rise even while profits in general were falling, but...
    • 2. They would be brought back down as capital moved into the more profitable field.

    F. Very long term story:

    • 1. Over time, we push onto worse and worse land, so the value of corn rises ...
    • 2. Checked from time to time by improvements in machinery for producing necessaries and in agricultural science.
    • 3. Eventually we reach the point where the workers on marginal land just produce enough to pay their wages.
    • 4. So the rate of return on capital is zero, so it stops accumulating.
    • 5. And in fact, accumulation will stop short of that point.
    • 6. As accumulation goes on, total amount of profits keeps rising for a while, but eventually the reduction in the profit rate with further accumulation outweighs the increase in the amount of capital that rate is earned on, so ...
    • 7. Total of profit starts to go down.
    • 8. But the total of value is still going up.
      • a. As we move to worse land, we are still farming the better land, so ...
      • b. The increase in capital means more total output, and ...
      • c. The (agricultural) output has greater value/unit, so ...
      • d. Total value of agricultural output is increasing.
      • e. Indeed, total value is increasing--more labor (which produces a value divided between labor and capital) and more rent.

      9. So the shift in value is away from profit and into labor and rent--the latter get a larger fraction of the total, and eventually the former starts to fall absolutely as well as relatively.

Chapter VII: On Foreign Trade

• I. Trade does not directly increase the value annually produced by a country, although it may increase the amount, usefulness, etc. of the goods the country can consume.
  • A. Because if the ability to import wine cheaply increases its quantity and drops its price, it also drops its value
    • 1. Wine is now being produced by using English labor to produce (say) cloth and trading that for Portuguese wine
    • 2. And it takes less English labor to produce wine that way than directly, so ...
    • 3. It has less value--in Ricardo's sense.

    B. It has been argued by high authority (Smith?) that when capital is shifted into foreign commerce, the result will be to raise the rate of profit in general. But ...

    • 1. The purchase of foreign commodities will employ the same, more, or less of the produce of England's land and labor (as producing them would have before?).
    • 2. If the same, then the same amount available for everything else.
    • 3. If less, then more money available to purchase domestic commidities--and more capital to produce them.
    • 4. If more, then proportionally less demand for domestic commodities, and less capital to produce them.
    • 5. Putting it differently, foreign trade is simply a new technology for producing goods, and in Ricardo's system, new technologies (save in agriculture or necessaries for workers) do not affect the profit rate.
    • 6. But note that Ricardo has assumed away the affects of changing labor/capital ratios--which might be relevant if, as Smith argues, foreign trade is a capital intensive industry.

    C. The effect of trade is to get us more usefulness for the same value.

    • 1. Which might result in more capital accumulation, if people who don't need to spend as much money to get the same goods save what is left, instead of buying more goods with it.
    • 2. And it could increase profits if it decreased the cost of food and necessaries, and thus wages.
    • 3. Ditto for improvements in manufacture, inland trade, etc.

    D. Profits depend only on wages (measured in value terms!), Prices are independent of wages (rise in wage compensated by drop in profits) but depend on productivity.

    • 1. So an improvement in productivity benefits everyone
    • 2. A fall in wages benefits only the owners of capital.
• II. The theory of value that explains prices within a country does not explain prices across countries:
  • A. Because labor and capital are mobile within a country but relatively immobile across countries.
  • B. So profit rates and wages tend to be equal within a country but not across countries.
  • C. So exchange ratios in international trade are not determined by ratio of embodied labor.
    • 1. It might be profitable for Portugal to import from England cloth that cost 90 man years to produce in Portugal and 100 man years in England
    • 2. Because it would be sending wine, which costs 80 in Portugal, 120 in England.
    • 3. Principle of comparative advantage first rears its head.
    • 4. If capital and labor were perfectly mobile, this wouldn't happen--because they would be getting more in Portugal, and would move there.

    D. Gold and silver distribute themselves among countries in such a way as to make profitable in money terms those trades that would be profitable in barter terms.

    • 1. The cloth must sell for more in Portugal, or it won't be imported, so ...
    • 2. Wages paid 90 men in Portugal must be more than those paid 100 men in England (actually, cost of wages plus associated capital costs)
    • 3. Which means that the value of gold in Portugal, measured in Portuguese labor, is lower than the value of gold in England, measured in English labor.
    • 4. If it were not, gold would flow into Portugal (exporting both cloth and wine, importing nothing) until it was.
    • 5. The specie flow mechanism for foreign trade equilibrium.
    • 6. If England discovered a new way of making wine that cost less than 80 man years, England would export cloth, import nothing, until enough gold accumulated to drive the price of one of the goods in England above that in Portugal (This version is actually Ricardo's example, the previous was my addition).
    • 7. If there is a trade imbalance, then people who buy bills of exchange in one country on another know they may not be able to trade for a bill the other direction, so discount to allow for the cost of transporting the gold to pay the bill.
    • 8. Oddly enough, if England could make both goods cheaper (in labor), gold flowing into England and out of Portugal would make prices in general rise in England, fall in Portugal (until a new equilibrium was reached).
• III. Explanation of varying value of money in different countries:
  • A. If a country improves its ability to produce goods that are readily traded, gold flows in, prices measured in gold (real gold, not Ricardo's imaginary money) are higher.
  • B. Rate of profits may be the same, wages the same, but everything measured in money is higher.
  • C. So real money, unlike Ricardo's imaginary money, varies in value across countries for this reason.
  • D. But this does not imply a difference in the profit rate, which is a ratio of two amounts of money, across countries.
  • E. In the early state of society, when goods are bulky, the value of money depends mainly on distance from the mines (think of the labor cost of growing corn, transporting it to the mine, trading for gold, bringing the gold back).
  • F. But once one country improves its manufactures (of high value to weight goods), now it is relative ability in manufacturing those that determines the value of gold at home.
  • G. The higher value of money will not be indicated by the exchange.
    • 1 That must be at par as long as you can freely import and export money, and doing so does not cost very much.
    • 2. A country that could prohibit the export of money could push up its domestic prices--and push the exchange against it. Must give 11 ounces in that country to get 10 ounces abroad.
    • 3. Similarly with paper money not freely convertible into gold.
    • 4. And an exchange against England is evidence that the currency is depreciated--otherwise you would just export money, melt it down, and have it coined as foreign money.

Chapter VIII On Taxes

• I. What does Ricardo mean by a tax "on X"? What do we mean?
  • A. Suppose you have an income £1000, spend £800 on consumption, so £200 a year goes to capital accumulation.
    • 1. Tax of £100 is imposed.
    • 2. If you reduce your expenditure to £700, the tax is "on revenue"
    • 3. If you reduce your savings instead, it is "on capital"
    • 4. So what Ricardo is asking is what use of your income is cut by the amount of the tax-- "where does the money come from."
    • 5. Relevant if you are asking "where will the lack of that money be felt." Or "what are the consequences of having paid the money?"

    B. Contrast that to the modern sense.

    • 1. Compare a consumption tax to a tax on interest.
    • 2. A tax on interest might result in people cutting consumption, so as to keep up their savings.
    • 3. A tax on consumption might result in people consuming just as much and spending down capital.

    C. We are interested in two different questions, neither of which is Ricardo's.

  • D. First question:
    • 1. What determines how much tax you owe, hence what incentives does the tax produce?
    • 2. If I am taxed on interest, that lowers the after-tax rate of return, making investment less attractive.
    • 3. If I am taxed on consumption, I can avoid the tax, at least for a while, by saving and investing.
    • 4. More generally, a tax changes relative return from different activities, thus gives me an incentive to change behavior.

    E. Second question is incidence--when you impose a tax, who (or what factor) really pays it?

    • 1. But that doesn't mean, as in Ricardo, what factor is diminished as a result.
    • 2. It means what factor determines how much you owe.
    • 3. Ricardo is interested in the issue of how prices adjust to taxes, and thus what incidence is in our sense--but ...
    • 4. He is looking at it primarily as "effect on how much capital etc. exists" not as "effect on your incentive to avoid the tax by not doing the activity the tax really falls on.

    F. Part of the reason for the difference is that Ricardo is much less interested than we are in the effect of incentives.

    • 1. In particular, his implicit model has a constant labor/capital ratio, so ...
    • 2. Changing the relative price of labor and capital doesn't automatically shift the ratio--as it would in our models. He doesn't have a "marginal product of capital" picture.
    • 3. Although he does recognize that the interest rate is an incentive to accumulate, and ...
    • 4. The wage rate an incentive to procreate.
• II. Taxes on transfers
  • A. Smith merely objects to them on Ricardian grounds--that they reduce the capital stock
  • B. Ricardo also objects to them on Coaseian grounds--that we want to move property to its highest valued use.

   

Chapter IX: Taxes on Raw Produce

• I. What is the effect on prices of a tax on raw produce?
  • A. It raises the cost for producing on marginal land, hence raises the price of produce.
  • B. Corn rent will fall, money rent stay the same, because:
    • 1. Difference in corn output between lands of varying fertility remains the same, so ...
    • 2. Corn rent gross of tax remains the same, but ...
    • 3. The more corn produced, the more tax paid, so corn rent net of tax falls, but ...
    • 4. This is precisely offset by the increase in the price of corn, so money rent stays the same.
    • 5. It must be precisely offset by the increase, because the after-tax money return to the farmer on marginal land must stay the same.

    C. So wages will go up.

  • D. So profits will go down.
  • E. So the landowner, and the stockholder (i.e. someone who has invested at a fixed interest rate before the tax went on) are unaffected, but other owners of capital pay the tax indirectly in lower profits as well as directly in their own food.
• II. Possible objections to such a tax are ...
  • A. An unequal tax, but ...
    • 1. That could be remedied by combining it with taxes on rent and interest, thus ...
    • 2. Producing the effect of an income tax, without...
    • 3. An intolerable inquiry into what individuals' income actually is.

    B. During the interval between imposing the tax and adjusting the population, workers would suffer.

    • 1. Ricardo is not sure there is such an interval, and thinks in the general case it depends on why food is expensive.
    • 2. If the reason is a bad harvest, then high wages aren't going to solve the problem
      • a. The food simply isn't there, so unless you somehow provide more food from somewhere, the workers are simply going to go hungry.
      • b. And if you do adjust wage upward, that simply bids corn still higher, transferring money to the producers.

      3. If the reason is increased demand, that is a consequence of high wages--which resulted in the population increase that is increasing the demand.

    • 4. If the reason is inflation (fall in the value of money), then the inflation will affect wages just as it affects the price of corn, and simultaneously.
    • 5. What if the reason is a tax on corn?
      • a. Does not reduce the amount of corn? Is that clear?
      • b. Does not reduce demand vs supply of labor.
      • c. If wages did not rise in proportion to price of corn, would there not be an excess supply of corn, resulting in price dropping?
      • d. But if price dropped capital would be withdrawn from agriculture, and ...
      • e. Why can't we have a new (short run) equilibrium, with less corn produced, either on the same land, price having risen and wages not, or ...
      • f. On better marginal land, with tax partly compensated for by a drop in the value of corn--and, as a result, a fall in rent?

      6. Defense of Ricardo's claim that wages would rise immediately when you tax corn.

      • a. The existing stock increases in value when you impose the tax--because it consists in part of a certain quantity of corn, and the value per quantity of corn has just gone up due to increased cost of production (since cost of production includes the tax).
      • b. Imagine that the stock was all corn and all wages were corn. You would bid corn wages up to their old level
      • c. You should get the same result if half the stock is corn and half the worker's income goes for corn. When corn rises 10% wages rise 5%, letting the worker consume the same corn and the same amount of other goods as before.

    C. Raising wages and thus lowering profits discourages accumulation.

    • 1. Ricardo answers that any tax must be paid from somewhere, and savings can be made from anywhere, so ...
    • 2. A tax on profits has no more tendency to decrease savings than ...
    • 3. A tax on consumption would
    • 4. But this misses the incentive effect--lowering profits reduces the gain to you from future savings.
    • 5. Ricardo is looking only at the flow of money, and not at the incentives that determine where it will go.
    • 6. Ricardo thinks that a tax that reduces profit is better than a tax on consumption, because everyone getting income from profit must pay it--including the miser. So if combined with other taxes on other sources of income, everyone contributes according to his means.
    • 7. Ricardo thinks in terms of the question of whether a tax "presses unequally on that class which accumulates and saves" rather than whether it presses unequally on income from savings, which is not at all the same thing.
    • 8. He ought to see this, since he does recognize that, in the process of accumulation, savings eventually stops when the interest rate gets low enough.

    D. Because the price of all commodities that contain raw produce has been raised, England will be unable to compete in foreign trade.

    • 1. But the price of goods that do not contain raw produce, or contain very little, will fall, measured in real gold.
    • 2. Since if the first effect is to raise the price of some goods and leave others the same, the result will be an outflow of gold and a general price drop, until trade again balances.
    • 3. But there is an inefficiency due to the fact that the relative price of goods with differing amounts of raw material in them no longer equals the relative cost of production, so foreign trade no longer quite maximizes net gains.
    • 4. Preventing the "very best distribution of the capital of the whole world."

Chapter X: Taxes on Rent

• I. A tax on rent cannot raise the price of raw produce
  • A. Since that is determined by cost of production on land that pays no rent.
  • B. Hence the tax is paid entirely by the landowner.
  • C. But ...

  II. This assumes we are really taxing economic rent, and not also taxing the profit on the landlowner's investment in buildings etc.

  • A. A tax on rent as currently defined would raise the price of corn, since it would fall in part on capital provided by the landlord on marginal land.
  • B. But if rent were taxed, landlords would soon find a way to separate economic rent from other payments, so as not to pay the tax on the latter.
  • C. How? Ricardo seems to skip over this problem. Possibilities include:
    • 1. The landlord bills separately--but then he will decide that almost none of what he is getting should be classified as economic rent, since that way he doesn't have to pay taxes on it.
    • 2. All improvements are made by the tenant--but that has obvious problems, since the tenant has no immediate incentive to be concerned with gains after the term of his lease, and ...
    • 3. If the improvements are attached to the land, they will revert to the landowner when the lease terminates--and be taxed as rent in the next round of leases.

Chapter XI: Tithes

• I. A tax, paid to the church, consisting of a fixed fraction of agricultural output--a tenth.
  • A. Like the tax on raw produce, it raises the price of corn.
  • B. If the price was £4/qtr and we have a tithe, Ricardo says price goes to £4 8s=88s
    • 1. But ten percent is now 8.8s, not 8s, so what is left after tax is less than £4, so
    • 2. price should go just a little higher-- 80/.9=88.888...

    C. The only difference between a tithe and a tax on raw produce is that the tithe is a fixed percent rather than a fixed amount of money per quarter of corn

    • 1. So if the value of corn is falling (either in a retrograde state or when agricultural technology is improving), a tithe is a decreasing tax in value terms
    • 2. But if the value of corn is rising (advancing state, pushing onto worse and worse land), a tithe is an increasing tax in value terms.

    D. Just like the tax on raw produce, it changes the corn rent but not the money rent.

    • 1. Corn rent is the difference in output between the more fertile land and the marginal land
    • 2. Which is still the same (unless the margin moves, which Ricardo thinks it will not).
    • 3. But 1/10th of that additional output goes to the church, reducing the corn rent to .9 of its old value.
    • 4. But the price of corn goes up by 1/.9, so the money rent stays the same.

    E. But it does encourage the import of corn, since we buy it with outputs that do not pay such a tax.

    • 1. That effect would be eliminated if we taxed imported corn at the same rate.
    • 2. In which case it would be a fair tax if paid to the state and used to diminish other taxes, but...
    • 3. If it goes to the church, it increases the amount of unproductive labor.

    F. More generally, taxing one industry and not another results in an inefficient pattern of imports and exports.

    • 1. Still more generally, although Ricardo doesn't mention it, it results in an inefficient substitution between taxed and untaxed goods
    • 2. For example, if the tithe is on raw produce, there is an incentive to be inefficiently careful to avoid wastage in milling grain, storing etc., since the saving from that do not pay tithe.

Chapter XII: Land Tax

• I. If it is imposed in proportion to rent, varying as rent varies, it is simply a tax on rent, already discussed.
• II. But a tax on all cultivated land, independent of rent (a fixed amount per acre, presumably), is a tax on produce, so raises its price
  • A. It will raise the price of corn in proportion to the tax paid on marginal land
  • B. Thus more than compensating the owners of better land, who are paying less tax per unit of output
  • C. So some of what the consumers lose through the tax goes not to the government but to the landlords who are now collecting more rent than before.
  • D. Consider the virtues of the FDA from the standpoint of drug companies.
    • 1. It makes rules that make it more expensive to introduce new drugs
    • 2. Which drives up the price of drugs
    • 3. The drug companies pay the "tax" (higher cost) on their new drugs, and receive the higher price
    • 4. But they also have an inventory of old drugs, already produced, and since they compete with new drugs, the rules drive up their price too
    • 5. So the drug companies might be better off as a result of the rules.
• III. Smith thinks any tax on land really comes out of rent
  • A. Because farmers calculate the taxes they will have to pay in deciding how much rent they are willing to offer.
  • B. Which is true, but ignores the possibility that the tax might raise the price of corn, thus making farmers willing to pay more in rent+tax than before.

  IV. Such taxes discourage cultivation--but for the same reason as all taxes.

  • A. Part of the labor of the country is diverted to the state.
  • B. If this is large enough it may not be worth people continuing to save.
  • C. But Ricardo treats this as if it were an all or nothing effect, rather than a continuous reduction in the rate of accumulation as the profit rate falls.
• V. Ricardo goes through Say's analysis, to show that it is mistaken.

Chapter XIII: Taxes on Gold

• I. Suppose gold were produced domestically, and its production was taxed.
  • A. In the short run, its quantity stays the same, and so does its price.
  • B. Over time, it wears out, or the demand increases, and its price rises.
    • 1. While this is happening, the owners of the gold mines pay the tax out of decreased rents.
    • 2. When the process is complete, the higher price just balances the tax--on the marginal ounce of gold.

    C. The total value of all money would stay about the same, independent of its quantity, because

    • 1. Demand is precisely proportioned to value, since...
    • 2. What people want is not a certain weight of coin but a certain purchasing power.

    D. On the gold used for money, although a large tax was received, nobody would pay it:

    • 1. The savings would be paid out of a reduced input on mining gold
    • 2. With the same value--and usefulness--of gold produced.
    • 3. This works perfectly only if there is an unlimited supply of mines of equal fertility, so that there are no rents on gold mines.
    • 4. Otherwise the reduced output leads to a loss of rent on the more fertile mine
    • 5. So their owners pay part of the tax, the rest nobody pays.

    E. Quang Ng article: Diamonds are a Government's Best Friend

    • 1. Ng published an article in the AER, which argued:
      • a. That there were some goods for which the utility the consumer gets depends on the value of the good, not its quantity.
      • b. For example, diamonds; what you want is not a one carat diamond but a thousand dollar diamond, so as to show off your wealth.
      • c. So if you taxed the production of such a good, the tax would be paid, in effect, out of the savings of mining fewer diamonds
      • d. Thus giving you a burden free tax--not only no excess burden, but no burden at all.

      2. This is Ricardo's argument, except that he is applying it to gold used for money, where there is a much clearer reason why usefulness depends on value not weight.

    • 3.And Ricardo's version points the way towards fiat money--the limiting case as the quantity of gold approaches zero.
    • 4. Evidence both that Ricardo was a very smart guy, and ...
    • 5. That neither Ng nor the referrees nor the editors had read him.

XIV: Taxes on Houses

• I. Tax on rent of houses
  • A.Will diminish the demand for houses, and ...
  • B. Since the supply is very inelastic in the short term
  • C. The result will be a fall in rent, so part of the tax is paid by the landlord.
  • D. In the long term, the building rent will have to go up to its normal level--i.e. the same profit rate as anything else--so the tax falls on the ground rent.
  • E. And on the occupier of the house, presumably because he is paying a higher building rent
  • F. Ricardo concedes Smith's argument in favor of taxing ground rents--perfectly inelastic supply ( "no discouragement... to any class of industry"), but ...
  • G. Argues that it is unjust to the owners, and ...
  • H. Coaseian point.
    • 1. Land value becomes very uncertain if it depends on the level of government expenditure--as it will if it is the sole source of taxes to finance such expenditure, so ...
    • 2. Land will be held by gamblers, not
    • 3. The sort of men to make the best use of it.

XV: Taxes on Profits

• I. Incidence of taxes
  • A. Tax on a luxury only falls on the consumer, but ...
  • B. Tax on a necessity also falls on profits via increased wages.
  • C. Tax on profits in any particular field merely raises prices in that field.
  • D. Tax on profits in all fields:
    • 1. Raises all prices, if money is not taxed.
      • a. In which case all producers are compensated in the higher price for the tax, meaning that theirmoney profit is unaffected, but ...
      • b. Pay the tax in the higher prices they pay for the goods they consume.

      2. Leaves all prices (in money) the same if money is taxed too (i.e. tax on the profit from mining gold)

      • a. In which case producers receive the same profit before tax, and pay the tax out of their profits
      • b. The only difference in the cases is that Ricardo has assumed a £100 tax, which represents different amounts going to the government in the two cases, since the value of money is different.

      3. Ricardo now drops his implicit assumption that all goods are produced with the same labor/capital ratio, and notes that a tax on profits will raise the price of capital intensive goods relative to labor intensive goods.

    E. Does a change in the money supply change relative prices?

    • 1. If goods are not taxed, no.
    • 2. If goods are taxed, no longer true.
      • a. If some goods are taxed and some not, and the tax is a fixed amount of money, then a price rise will lower the relative price of the taxed goods.
      • b. Since the tax rate, as a proportion of price, has gone done.

    F. Suppose we tax profits of everyone except the farmer (money untaxed)

    • 1. All goods except raw produce rise in money value, giving producers the same after tax money income as before.
    • 2. Farmer has the same corn and money income as before
    • 3. So pays the tax in the higher prices
    • 4. And the same is true if money is taxed, or changes its value for some other reason, although what the farmer pays then takes the form of a drop in corn prices.
    • 5. Ditto for the landlord.

    G. Suppose we tax everyone's profits.

    • 1. Farmer is in the same situation as everyone else, paying via higher prices.
    • 2. Landlord, on the other hand, gets more rent due to the tax on the marginal output, so ...
    • 3. He is fully compensated for the price rise, and remains untaxed.

    H. Stock holder has a fixed income, so it matters to him whether you do or don't tax money.

    • 1. If you don't, its value stays the same, everything else rises, he contributes to the tax.
    • 2. If you do, its value goes up to, he does not contribute to the tax.
    • 3. This assumes that dividends are not covered by a profit tax.

    I. In fact, with real money, imported from abroad ...

    • A. Prices of all goods cannot rise when you tax profits, since ...
    • B. If they did, gold would flow out until they came back down again.
    • C. So the situation is equivalent to the one where you tax the production of money.

XVI: Taxes on Wages

• I. Will raise wages, and so diminish profits.
  • A. Like a tax on necessities, except that ...
  • B. Stockholders, landlords, etc. pay no tax directly (they would with a tax on necessities, since they eat too).
  • C. Hence a tax on wages is wholly a tax on profits, while ...
  • D. A tax on necessaries is partly on profits, partly on rich consumers (of necessaries).

  II. Buchanan attacks Smith's claim that a higher price of provisions leads to higher wages. Ricardo responds:

  • A. A high price of provisions might reflect inadequate supply, in which case the shortage will have to be born--higher wages won't increase the amount of corn. But ...
  • B. There might be a high price with a high supply--if corn is expensive to grow (i.e. agriculture has moved onto bad marginal land) and there is a large demand.
  • C. Indeed, while short supply may create high market price, it is also true that high natural price reflects large supply, because
  • D. The high price means we are producing the marginal corn very far out on the margin (very bad land), so lots of land, including the good land, must already be in production, producing lots of corn.
  • E. Roughly speaking, in our terms, high price might reflect unusually low supply or unusually high demand.
  • F. Which is what we now call the identification problem-the reason you cannot simply deduce a demand curve from a series of historical quantities and prices (or a supply curve either).
  • G. In the long run, the population argument implies that wages must rise. What about the short run?
    • 1. Employers have the same capital as before--and the government is now also bidding for workers with the money they collected as taxes, so ...
    • 2. Wages must rise immediately.
    • 3. To prevent that, government would have to take the tax and spend it entirely abroad, which ...
    • 4. Would indeed lower corn wages--as would any tax that did the same thing.

    H. Suppose the tax were simply handed over to employers.

    • 1. They now have more money to hire laborers, so demand goes up, so price goes up
    • 2. So no ultimate effect at all--workers pay the tax out of the higher wage.
    • 3. Right answer--precisely the Social Security point (that it doesn't matter whether the tax is "paid" by employer or employee--the effect on both is the same), but ...
    • 4. Ricardo is getting it in a different way--he doesn't explicitly assume that payment to employers is proportional to the wages they pay.

    I. But taxes are wastefully spent, so tend ultimately to diminish the capital stock and lower wages.

  • J. Smith thinks a tax on wages will show up as higher prices of everything but agricultural produce and will lower rents.
    • 1. If this were true, if all wages and profits went up proportionally to the tax, then the manufacturers would be able to consume the same amount of each others' goods as before.
    • 2. So the tax would be paid entirely out of rent.
    • 3. And the manufacturers would actually benefit in being able to buy more corn than before.
    • 4. But the total tax collected is only a little more than the total rise in the price of all commodities, so to pay all the tax in the form of higher prices the landlords must be consuming practially all of the manufactured commodities--which they are not.
    • 5. Indeed, there is an infinite series buried here: when the price of commodities rises because of the higher wages, that raises the cost of living to workers, so wages go up more, so ... .
    • 6. Ricardo may not realize that infinite series may have finite sums.

    K. Landlords as such do not pay the tax, but as consumers of labor (hiring servants etc) they do.

  • L. Conclusion--taxes on the labouring poor are ultimately paid out of profits not out of rent.
  • M. Smith thinks that taxes are multiply accumulated
    • 1. Correct if it means that the rich taxpayer is also paying the taxes of the poor, but not ...
    • 2. If it means total amount paid is more than total collected.

    N. Ricardo argues that excess burden is impossible, but ... remember he is calculating in value not usefulness.

    • 1. When capital shifts out of taxed salt
    • 2. People are getting something else less useful.
    • 3. A point he sees in the international trade context.
    • 4. But they are getting the same value in his sense, since whatever is produced instead has the same input cost, hence the same value.

    O. Ricardo points out the argument for taxing the export of goods where you have a national but not individual monoposony.

  • P. And points out that elasticity exists for the total even if most people won't change consumption at any given price point.

Chapter XVII: Taxes on Commodities other than Raw Produce

• I. A tax on any commodity raises its price to compensate, because of equilibrium on the capital market.
  • A. If the commodity is a necessity for workers, the increase raises wages, just as a tax on corn would.
  • B. If it is a luxury, there is no effect on wages, so the tax is paid by the consumers.

  II. The economics of debt finance

  • A. If the government borrows £20 million to pay for a war, i.e. to support unproductive workers, ...
    • 1. The real cost to the capital plus consumption of the society is the amount spent (£20 million)
    • 2. not merely the interest on the debt.
    • 3. Indeed, the interest is merely a transfer from the taxpayers to the people who lent the government the money, and not a net cost at all.

    B. Suppose the government had collected the whole £20 million in taxes the year it was spent.

    • 1. Individual taxpayers could still borrow their share and pay out the interest annually.
    • 2. So far as whether the taxpayer then paid his creditor, the government should be indifferent, aside from ...
    • 3. A general interest in the performance of contracts, and ...
    • 4. Which of the two could use the money in a more productive way--which could, of course, go in either direction.
    • 5. This basic point--that it doesn't really matter whether you finance expenditure by debt or taxes, provided the amount spent stays the same--is called the "Ricardo Theorem" and still a matter of interest.

    C. When you get the same effect, but with the government doing the borrowing, the logic of the situation is the same.

    • 1. Justice and good faith demand that the debt be paid.
    • 2. What about the question of who can use the money better?
      • a. Repudiating the debt and lowering taxes accordingly.
      • b. Will make one man poorer (the creditor) and one richer (the taxpayer)
      • c. Leaving taxable income, capital, etc. the same after as before.
      • d. Or in other words, repudiating the debt doesn't actually relieve the country of a burden and make it better off--just redistributes (unjustly).

    D. Debt finance of war is a bad idea, because

    • 1. It makes the cost look less than if we paid for the war by current taxes, even though the real cost is in fact the same.
    • 2. The individual taxpayer sees his income reduced only by the interest on the loan, so only reduces his expenditure by that amount, with the result ...
    • 3. That the total capital of the society goes down--government is shifting £20 million out of productive activity, but consumers are reducing their consumption by only £1 million in the first year, so the capital stock is falling by £19 million.
    • 4. Furthermore, once a country has accumulated a large debt, it is in the interest of citizens to get out of paying their share by emigrating.
    • 5. So a country that has accumulated a large debt should try to pay it off quickly. It sounds rather as though Ricardo is advocating a once and for all capital levy to eliminate the debt, although that is not entirely clear.

• III. Taxation and monopoly prices:
  • A. A monopoly price applies to a good in fixed supply--perfectly inelastic supply curve in our terminology.
    • 1. Hence the price is entirely determined by demand.
    • 2. Which may vary from year to year, causing the price to vary as well.
    • 3. An example would be Smith's wines, produced on special land.
    • 4. It does not seem to occur to Ricardo that even special vinyards have an intensive margin, and that the quantity thus can be varied.

    B. Raw produce is not at a monopoly price

    • 1. Because more can be produced
    • 2. The only difference between the case of raw produce (i.e. agricultural output) and manufactured goods is that
      • a. The price of raw produce is determined by its cost of production on marginal land.
      • b. The price of manufactured goods is determined by their cost of production--which is the same for every unit.
      • c. In other words, in our language, P=MC, MC=AC for manufactured goods, which Ricardo assumes are produced under conditions of constant cost, but MC>AC for agriculture.

    C. It would be at a monopoly price if the land were so thoroughly cultivated that no more could be produced--output on both the extensive and intensive margin <=cost of production.

    • 1. In which case all land would pay rent, and ...
    • 2. Any tax on corn will come out of rent, since ...
    • 3. Supply curve is perfectly inelastic (our language, not his)
    • 4. Of course, this assumes the tax is not enough to drive the rent on the least productive land below zero.

    D. Argument with Mr. Buchanan (and Smith) on whether raw produce is at a monopoly price:

    • 1. Buchanan thinks they are because they yield a rent, and infers that all taxes on raw produce must fall on the landlord not the consumer.
    • 2. But, as Ricardo has shown, it isn't true--marginal land does not yield rent, and it is cost of production on marginal land that determines price.
    • 3. So a tax on raw produce is paid in higher prices.

    E. Inconsistency between the position Smith and Buchanan take here and their position on the subject of a tax on malt.

    • 1. See p. 176 for an elegant example of how to politely point out that someone is inconsistent--by quoting one side of his position as the ideal statement of your argument.
    • 2. Tax on malt does not lower the price of barley, because it must pay the same return as other crops that could be raised on the same land.
    • 3. So it must raise the price of malt.
    • 4. But Buchanan thinks that a tax on barley (form which malt is made--malt is barley that has gone through one stage of the processing that converts it into beer) will not raise the price of barley but be paid out of rent.
    • 5. But Buchanan asserts that a tax on bread will be paid out of rent
      • a. Which seems inconsistent, since bread is one of the things made out of wheat
      • b. Just as malt is one of the things made out of barley
      • c. So taxing bread should have the same effect as taxing malt.
      • d. Perhaps Buchanan believes that bread is a necessity and beer isn't (heretical view, of course), and that that affects the argument.

      6. Say argues that agricultural output is unaffected by tax, so market price is unaffected, so the landlord must pay the tax out of rent.

      • a. This would be right if what was taxed was rent, but ...
      • b. If what is taxed is output, the result is that farmers on the less fertile land could no longer cover their costs, even at zero rent, so supply would go down.
      • c. Does "net produce" mean "net of the farmer's input costs," in which case it is rent plus farmer's profit, or "net of all production costs" in which case it is rent, or what?
      • d. I don't know--the term is associated with the single tax proposed by the physiocrats, French pre-Smithian economists--and I have been unable so far to find a clear definition.

Chapter XVIII: Poor Rates

• I. Poor rates are the taxes that pay for the poor law--the English welfare system.
  • A. Nominally levied on the farmer in proportion to his rent
    • 1. If so, they are a tax on rent, and ...
    • 2. Fall entirely on the landlord.
    • 3. But in fact, the tax is proportioned to the annual value of the land--which, as Ricardo pointed out earlier, is in part economic rent and in part return on capital invested in the land--improvements, buildings, fertilizing, etc.
    • 4. Which means that it is in part a tax on such improvements, hence on marginal output if the way you get the marginal bushel of corn is to devote capital to either improving existing farm land or bringing new land into cultivation.
    • 5. So it raises the price of corn.
    • 6. And this is true whether the improvements are paid for by the farmer (renting the land on a very long lease, so that he gets much of the gain from the improvements) or the landlord.

    B. Under current application of the law, the farmer is paying a much larger fraction of his profits than the manufacturer, since

    • 1. The farmer is rated according to his production (?not explained?)
    • 2. The manufacturer is rated only according to the value of his buildings, not the value of the rest of his capital, fixed and circulating

    C. In an advancing society (where more land is being taken into production, so the cost of making bad land productive is part of the marginal cost of producing food) poor rates are largely a tax on corn,

    • 1. and get paid in part by capitalists in lower profits (due to higher wages), in part by consumers of raw produce in higher prices.
    • 2. So might even benefit landlords, by giving them more in the higher price of corn than they pay in other forms.
    • 3. Like a tax on corn, discussed earlier.

    D. But in a stationary or retrograde society

    • 1. The land under cultivation isn't expanding, so improving the land isn't part of the marginal cost of corn, so the price doesn't rise.
    • 2. So the tax is paid in the short run by the farmer, on renewal of the lease by the landlord.
    • 3. The reason for this is that the landlord (or farmer's) capital at this point is incorporated into the land, and cannot be withdrawn, even though if it could that would reduce the tax, making the tax again a cost of production.
    • 4. And in his footnote, Ricardo anticipates Marshall's concept of quasi-rent (and the more familiar concept of sunk costs) about eighty years early, noting
      • a. That improvements incorporated with the land (sunk costs) henceforth are subject to the same economic principles as land, but ...
      • b. Will not be made in the first place unless they are expected to at least pay for themselves.

Chapter XX:Value and Riches, Their Distinctive Properties

• I. Value and riches are quite different things:
  • A. Value is cost of production, so technological improvements that increase output do not increase value
    • 1. If we can produce twice the output with the same inputs, then ...
    • 2. Value of output per unit falls in half, so ...
    • 3. Total value being produced stays the same.

    B. Riches (aka Smith's use value, roughly our utility) are increased by technological improvements.

  • C. It is true that if I possess a scarce commodity, its scarcity makes me richer--because its scarcity makes other people willing to give me lots of good things for it
    • 1. But its scarcity doesn't make the society as a whole richer--my gain due to its scarcity is at the expense of the people who trade lots of good things for some of my scarce commodity
    • 2. Indeed, its scarcity makes the society as a whole poorer-one useful thing we have less of.

    D. Smith's labor commanded definition of value doesn't make sense for riches, since increased productivity gives us more riches without more command over labor.

  • E. Wealth may be increased either by having more labor and capital used to produce or by having the existing inputs produce more per unit--i.e. by technological improvement
    • 1. In the first case, the country becomes rich and the value of its output goes up--more input. And it gets that way by parsimony--by giving up current consumption in order to increase its capital stock.
    • 2. In the second case, wealth increases but not value--and nothing has to be given up.
    • 3. So the second way is obviously better if you can manage it.

    F. Long discussion on Say's inconsistent use of the terms.

  • G. Say argues that Smith is wrong to count only labor as giving value.
    • 1. But Smith does not deny that nature provides use value.
    • 2. And Ricardo does deny that natural forces freely available give exchange value.

Chapter XXI: Effects of Accumulation on Profits and Interest

• I. Profits can fall only because of a rise in wages
  • A. If capital accumulates, and if time is permitted for the wage force to grow accordingly, profits would stay the same, save that
    • 1. Agriculture would be pushed onto worse marginal land
    • 2. Making wages rise and profits fall.

    B. But Smith attributes the fall of profits to the accumulation of capital and the resulting competition.

    • 1. But the competition for labor is only a short run phenomenon, since (assuming plenty of land) the higher wage will result in the labor force increasing until the wage is back down to its natural level
    • 2. Smith is here confusing disequilibrium--profits low because wages have been bid above their natural level--with the long term effect of accumulating capital.

    C. As Say has shown (still called "Say's Law") production creates its own demand,

    • 1. You produce something in order to trade it for other things
    • 2. People's wants are insatiable--there is always something they want more of
    • 3. And money saved is spent.
    • 4. So the whole economy can simply scale up and up, in size and output, as long as there are no fixed factors of production (for Ricardo, fixed quantity of fertile land).
    • 5. So whether increased output causes a fall in profit depends only on wages, which depend on the cost of producing food.

    D. Smith thinks that merchants invest in foreign trade because there is nothing left to do with their capital at home, but ...

    • 1. Although demand for necessities may be limited
    • 2. Demand for other things is not, and
    • 3. It is only because foreign trade gets us things better that we do it.
    • 4. If we could not engage in foreign trade, we would invest our capital at home.
    • 5. Presumably making (less of, or lower quality of) the goods we now get by trading for them.

    E. What does real world evidence tell us?

    • 1. Not a lot. Recorded interest rate figures (available to Smith and Ricardo for past times) are mostly of the legal maximum rate.
    • 2. People borrowed at more than that, but since they did it covertly we don't have good data on what the real market rate was.
    • 3. The price of funded property (roughly speaking, the market value of a government bond that pays a fixed interest) is not a reliable measure of the interest rate, because
      • a. In war time, the government keeps borrowing, so the market doesn't have time to reach equilibrium.
      • b. In peace, some people prefer government securities to other investments, so the interest rate calculated from the price of securities is below the market interest rate for other loans.
      • c. different sorts of securities are not perfect substitutes for each other

      4. In a footnote, Ricardo points out that Say is missing the logic of risk premiums, in assuming that because the riskiness of a government loan (in a country whose government cannot be trusted to repay them) requires the government to pay a higher interest rate, that will force up the general interest rate--presumably for safer investments.

Chapter XXII: Bounties on Exportation and Prohibitions of Importation

• I. Bounty on exporting corn lowers its price abroad, leaves it the same at home.
  • A. If corn sells for less abroad than it costs at home (Ricardo is ignoring transport costs)
    • 1. It can only be sold if there is an export bounty
    • 2. In which case, the total demand for British corn incrreases
    • 3. And in the short run, this raises the price in England (and make the reduction in price abroad less than it would otherwise have been)
    • 4. But since this does not affect the cost of production, the price in England would fall back to cost as more capital shifted into agriculture.
    • 5. All of this assumes, although Ricardo doesn't say so, that the increased demand doesn't push us onto worse marginal land.

    B. Someone (not named) argues that a bounty will increase the money price of corn

    • 1. and thus wages,
    • 2. and thus the price of all other commodities
    • 3. So a bounty will make the money price of everything higher.
    • 4. Which seems right in terms of the specie flow mechanism, since
      • a. The bounty increases exports
      • b. Pulling in gold
      • c. And raising the general price level to a new and higher equilibirum.
      • d. And Ricardo mentions this point later in the chapter

    C. Ricardo replies that in the short run, there is a scarcity of corn, which will not be solved by higher wages,

    • 1. and in the long run additional corn will be produced at the old cost
    • 2. Unless the increase pushes us onto worse marginal land
    • 3. Thus increasing the cost of production of corn.

    D. Ricardo disagrees with Smith's analysis of bounties on corn exports (while referring to his "justly celebrated work.")

    • 1. Smith assumes there is no effect on output (remember "in the state of the tillage?")
    • 2. His argument was that the increased price of corn had no effect on how much was grown, since the real price is not increasing, since wages are going up with the price of corn.
    • 3. But, Ricardo points out, corn is not all that laborers consume,
    • 4. so their wages, even if they adjusted instantly, would go up by only part of the increase in the price of corn.
    • 5. So growing corn would be more profitable, so more would be grown.
    • 6. But in any case, the increase in wages must be paid by the manufacturer as well as the farmer, and only the farmer is getting the increased price due to the bounty, so ...
    • 7. Capital should shift from manufacturing to farming until relative prices are back to their old level.
    • 8. At which point the price of manufactured goods would not have risen because, although we are producing less of them (capital has moved out), we are importing more, in exchange for the exported corn.

    E. Smith's error is believing that all prices rise or fall in proportion to the price of corn.

    • 1. Because the price of labor, and everything produced by land or labor, rises or falls with the price of corn.
    • 2. But he is forgetting that an increased cost of labor might be balanced by a decrease in profits rather than an increase in price.
    • 3. And in fact his system is inconsistent because, as Ricardo has just pointed out, it starts with some special factor raising the price of corn, but doesn't notice that any resulting increase in other prices due to higher wages must affect corn as well via the wages of farm laborers, so its relative price will remain high.
    • 4. Ricardo dodges a lot of the difficulty of thinking this through by the trick of analyzing the problem mainly in terms of value rather than money price--thus ultimately in terms of relative prices.
    • 5. And points out that you can see the argument by going back to barter:
      • a. Can corn rise in exchangable value with other things?
      • b. Obviously yes, if its cost of production is changed by changing marginal land or varying agricultural technology
      • c. In which case it cannot be true that its price regulates all other prices (in the strong sense in which Smith uses the term--i.e. not only that the price of corn affects other prices, but that all other prices scale in proportion to the price of corn).

    F. Smith argued that bounties and tariffs really benefitted manfacturers, but only appeared to benefit landowners.

    • 1. Ricardo replies that if the bounty pushes agriculture onto worse land, it permanently benefits the landowners in the form of higher rents, while
    • 2. A bounty on the export of manufactured goods, or a tariff on the import, only benefits manufacturers of those goods temporarily, until enough capital has shifted into the industry to bring profits down to their normal level.
    • 3. Note that they are reaching their opposite conclusions from different arguments
      • a. Smith bases his argument on the natural wage relation between corn and wages.
      • b. Ricardo bases his on the special nature of agriculture--it uses an input in fixed supply (land).
      • c. And Ricardo, of course, is right.

      4. What is wrong with bounties and tariffs is that they cause capital to be misallocated among nations

      • a. If Corn can be produced in France for £3 15s and in England for £4, a 10s bounty will ultimately result in a price of £3 10s in France, £4 in England
      • b. And the French will gain 5s in less expensive corn for every 10s the English taxpayers lose.

    G. The bad effects of trade restrictions

    • 1. The reduce the total quantity of commodies we can consume
    • 2. But they do not benefit the protected manufacturers (in the long run) at the cost of landlords etc.
      • a. Protected goods are more expensive because they cost more to produce here
      • b. So the higher price does not represent a long term benefit to the producer of protected goods.
      • c. after all, the return on capital producing protected goods is driven down to the same level as on all other capital
      • d. So manufacturers pay the cost of tariffs (in the higher price of the goods they buy) just as everyone else does.

      3. This point is important, since landowners argue, on the authority of Smith, that the should have import duties on corn, to balance the effect onthem of the import duties on manufactured goods.

      4. The right solution is not to balance one bad law by adding another, but to repeal the first.

Chapter XXIII: On Bounties on Productions

• I. Subsidies to producers
  • A. On corn, paid for by a tax on everything else.
    • 1. None of the money goes to government, so the nation is neither richer nor poorer?
    • 2. Price of commodities goes up, corn down
    • 3. Corn rent goes up, corn income stays the same, just as with a tax on corn (signed reversed). Land lord gains in buying cheaper corn, loses in buying more expensive commodities.
    • 4. Wages go down if the fall in the cost of corn costs the laborer more than the rise in the price of taxed commodities
    • 5. If so, the employers pay lower wages--but have more of the tax on the commodities left to pay (since the laborers are not paying enough of it to cover the subsidy on their corn).
    • 6. Ricardo thinks there would be no effect on the amount of corn and commodities produced, because ...
    • 7. He apparently assumes a perfectly ineastic demand for corn. That sounds right, on the theory that the demand for corn is the population times the number of bushels per year to feed someone, but ...
    • 8. is implausible--consider the choice between eating agricultural output and feeding it to animals in order to produce meat, for example., or the decision of how much labor to spend keeping down wastage at the flour mill.

    B. Contrast the effect of a fall in the price of corn due to a bounty with the effect of a fall in the value of corn due to producing it with less labor.

    • 1. In the latter case, the capitalists are better off after the fall than before, since their profits are greater and prices have not risen.
    • 2. In the former case, their profits may go up, but so do the prices of the manufactured goods they buy.
    • 3. In general, taxes that change wages--for instance, taxes on necessaries--reduce profits, but only by the amount needed to transfer the tax from the worker, who cannot pay it (since he is at subsistence), to the employer who can.
    • 4. So it does not much matter whether you tax necessaries or profits--the same amount is being paid, and mostly by the same people.
    • 5. Note that Ricardo is very insensitive to substitution effects, so looks at the effect of a tax almost entirely in terms of who pays a tax, not how the price change due to the tax alters incentives and thus behavior.

    C. Suppose we reverse the bounty--tax corn and subsidize manufactured goods.

    • 1. Same outcome, signs reversed.
    • 2. If the laborer consumed more than the average fraction of corn (as, of course, he does), he would lose more than he gained in the price changes, so wages would rise, and ...
    • 3. The capitalists would pay the worker's share of the tax.
    • 4. Again, a complicated measure producing no result at all
    • 5. Again, because Ricardo assumes perfectly inelastic demand for corn.

    D. So far we have assumed no foreign trade.

    • 1. Add that in, and the subsidy to corn is subsidizing the food of foreigners at our expense
    • 2. Or the subsidy on manufactures subsidizes foreigners' consumption of those
    • 3. While the matching tax creates an incentive to import that good.

Chapter XXXI: On Machinery

• I. Ricardo's opinion was once that the introduction of machinery benefits landlords, capitalists, and workers, but ...

• II. He now believes it might injure the workers.
  • A. Because the development of machinery might increase the net income of the country, while ...
  • B. Decreasing the gross income, ...
  • C. Where the "cost" that is being eliminated is workers.
  • D. In modern terms, if new discoveries greatly raise the marginal product of fixed capital, capitalists will shift from circulating capital to fixed capital, reducing the demand for labor.

• III. Demonstration in a simple model
  • A. A capitalist has £20,000 with which he both farms and manufactures necessities, of which
    • 1. £7000 is fixed capital, and
    • 2.£13,000 is circulating capital
    • 3. And he makes 10% profit, all of which he spends on himself.

    B. Somebody invents a new machine which can do the same work, in either his farm or his manufacture, with much less labor (Ricardo omits this step, but there must be some reason why the capitalist changes from his old pattern of production)

  • C. The capitalist uses half his labor for the next year building the machine.
  • D. He is now left with half as much circulating capital--since half his workers were working on the machine instead of growing food and producing necessaries to replace the circulating capital that they consumed--but a correspondingly greater amount of fixed capital (i.e. the machine)
  • E. So he fires half his workers and produces as much as before with the other half plus the machine.
  • F. So net produce is still the same--£2000/year of profit for the capitalist--but gross produce is less, since he is no longer producing (and spending on wages) the output that went to the workers he is no longer employing.
  • G. And if enough of this happens, the demand for labor falls, and we are in a declining state so far as labor is concerned.
  • H. This argument is closely related to Smith's argument about different employments of capital
    • 1. Ricardo is considering the shift from a one year production period (circulating capital) to a very long production period (fixed capital, which takes many years to wear out and have to be produced)
    • 2. Smith was considering the shift from a one year production period (agriculture) to a several year production period (foreign trade).

• IV. The machinery has lowered the value of commodities, so everyone can live as comfortably as before at a lower cost measured in value.
  • A. So the capitalists can accumulate faster than before.
  • B. Since some of their accumulation goes to circulating capital, that gives a countervailing effect--an increase in the demand for labor.
  • C. So Ricardo cannot sign the net effect--machinery might make the workers better or worse off.
  • D. Note that the argument does not depend on the simple example--substitution of machinery for people could happen industry by industry.
  • E. Note also that Ricardo, unlike modern writers with similar worries, does not model the problem in terms of unemployment (even though he talks about redundant workers)
    • 1. In his model, wages fall
    • 2. And the labor force declines.

    F. Ricardo's conclusion

    • 1. Worries about the effect of machinery on the working classes are not simply bad economics
    • 2. And the result could go either way, since ...
    • 3. Workers are injured by the shift of capital from circulating to fixed, but ...
    • 4. Benefitted by the increased demand for servants by rich people, who now have more riches for a given value of income, increased capital accumulation, and lower cost of what they consume.

    G. Note that Ricardo does not (here) note the feedback from wages to capital/labor ratio

    • 1. As wages fall and profits rise (in his scenario)
    • 2. The new machine becomes less and less attractive
    • 3. Since it substitutes capital, which has now become more costly, for labor, which has become less.

• V. More generally, how people spend their money affects how well off workers will be.
  • A. If a rich man spends his money for (labor intensive) servants rather than for consumer durables and the like (carriages, furniture)
    • 1. His capital continues to circulate, employing workers year by year, instead of ...
    • 2. Being fixed and yielding a flow of services to him, but no wages
    • 3. And the former case results in a higher demand for labor, hence higher wages.

    B. Similarly, a country engaged in war with lots of people employed as soldiers and sailors increases the demand for labor, since ...

    • 1. A wealthy man who is taxed to pay the wages of the soldiers
    • 2. Might otherwise spend the money on durable consumer goods
    • 3. Which would employ workers once, to produce them
    • 4. but as taxes, it pays workers to produce what supports the soldiers--and takes the soldiers out of the work force.
    • 5. This assumes that the tax money comes out of revenue, not out of capital.
    • 6. Why doesn't the same argument work for my spending the money on furniture, thus on the clothes and food of the man who makes it?

• VI. Consider also the case of substituting horses for men--a lower technology version of the same problem.
  • A. So far as the theory is concerned, workers are simply machines to convert corn into labor
  • B. And so are horses.

• VII. Ricardo now notes the relation between changes in wages and profits and the employment of machinery
  • A. As wages go up and profits down, that gives a reason to substitute machinery
  • B. But wages are going up only because population is rising
  • C. So this slows the rate of increase of demand for labor, but cannot reverse it.

• VIII. Discouraging the adoption of machinery would be a mistake because ...
  • A. Foreign countries would permit it
  • B. Giving capital a greater return abroad
  • C. So British capital would flow out
  • D. Reducing the demand for British labor by more than the introduction of machinery would have.
  • E. And, of course, since you would be producing less for an hour of labor, you would get less goods in trade for an hour of labor.

• IX. These are issues that people have been worrying about for the past two centuries--usually in the form of concerns that "automation" of production would lead to mass unemployment.
  • A. The standard form of the argument is wrong, because decreased demand for labor will reduce wages, bringing quantity demanded back up.
  • B. And in fact, there has been no long run trend towards unemployment.
  • C. It is wrong a second time, to the extent that it sees the problem as "insufficient demand"--an argument Ricardo (and Smith) have already answered.
  • D. But it could in principle be right, if production functions shifted in a way that made capital much more productive relative to labor
  • E. Thus shifting the distribution of income in favor of capital
  • F. Conceivably by enough to outweigh the increase in total income--from the standpoint of the workers.
  • G. But note that in our world, a lot of the capital is human capital, so workers get income from both labor and capital, so ...
  • H. It is really an argument about unskilled workers vs everyone else.