Marshall Lecture Notes

• Where Marshall fits in:
  • A. Unlike the other books, this is the economics you have already studied
  • B. At its birth
  • C. With a thoughtful discussion of how to justify it.
    • 1. You may want to think about the arguments he is answering
    • 2. Especially if you have argued politics and economics with people.

    D. If you have the time, you may want to read more of the book than I am assigning.

• Chapter I (not assigned)
  • A. Economics is the study of man in the ordinary business of life.
    • 1. Study of wealth, but also ...
    • 2. Of man
    • 3. Since everyday work is one of the two great forces molding people
    • 4. Religion being the other.

    B. Economics is relevant to mankind both

    • 1. Because people spend much of their time earning and spending money, and ..
    • 2. Because your income has a large effect on the life you live.
    • 3. Especially if you are very poor.
    • 4. Hence the studyof the causes of poverty is of great importance.

    C. One important question is whether it is necessary to have lower classes, that is:

    • 1. "whether there need be large numbers of people doomed from their birth to hard work in order to provide for others the requisites of a refined and cultured life; while they themselves are prevented by their poverty and toil from having any share or part of that life."
    • 2. Do any such people exist in the U.S. at present? Our "residuum" consists mostly of people who are poor because they are not working--and most of them have televisions.
    • 3. Marshall argues that the trend in the 19th century is towards eliminating such,
      • a. with steam engines eliminating much of the exhausting and degrading toil.
      • b. Higer wages, more education,
      • c. Railway and printing press permitting better communication
      • d. Rapid increase in skilled labor, now outnumbers unskilled.
      • e. And many skilled laborers now live better than much of the upper class a century ago.

    D. Economics is a late development and still inadequate, because

    • 1. One reason is that people think of it as a base study of wealth.
    • 2. Another is that economic conditions of modern life are more definite, even if more complex, than in the past, meaning ...
    • 3. It is easier to make sense of the logic of the economy now, since business is more clearly separated from other things, rights more clearly defined, custom less important, etc.
    • 4. Although the same substance may have existed in a less obvious form in the past, so economics is relevant to past societies, just harder to invent if you are living in one of them.

    E. Sometimes it is said that modern industrial life is more competitive than life in the past; what does that mean?

    • 1. More competition between people? Not a fundamental feature of modern life.
    • 2. The real change from the past is that more is determined by rational choice, less by custom.
    • 3. The past has less deliberate selfishness and less deliberate unselfishness
    • 4. i. e. Less deliberation. That, not selfishness, is the characteristic of the modern age.
    • 5. Primitive people deal more fairly with their neighbors--but less fairly with strangers.
    • 6. On the whole, people are better off now, so we ought not to use a term like "competition" which has negative connotations to define modern industry.

    F. Why "competition" is often a pejorative term.

    • 1. Traders like to malign their competitors--although the customers are being benefitted.
    • 2. And "regulation of competition" is often a euphemism for suppressing competitors at the cost of the consumers.
    • 3. Doubtless if the alternative is "unselfish work for the public good," that is better still--but not an option for most of the work of the world at present.
    • 4. And the history of socialist ventures suggests that they only work for small bands of religious enthusiasts.

     

  Chapter II: The Substance of Economics

  • A. What are people's objectives?
    • 1. Obviously not just money--inventors et. al. are stimulated by a noble emulation, others by conceptions of duty, etc., but ...
    • 2. The steadiest motive to ordinary work is the desire to get paid for it.
      • a. The money may be used for selfish ends
      • b. Or unselfish ones.

      3. but the motive is supplied by an amount of money, and ...

    • 4. That makes it possible to quantify it, and thus ...
    • 5. Lets economics become more scientific than any other branch of the study of man, although
    • 6. Less than physical sciences.

    B. Quantifying incentive

    • 1. The force of motives can be approximately measured by the sum which someone will just give up to get something, or ...
    • 2. Must be paid to do something.
    • 3. Note that it is the force of the motives, not the motives, that is being measured.
      • a. We cannot measure and compare directly mental states
      • b. Some affections belong to higher, some lower nature, hence different in kind
      • c. But comparable in force--as incentives to action.

      4. So if someone is just balanced between two alternaties it makes sense to say that he expects from them equal pleasures.

    • 5. So they provide equal incentives to action
    • 6. Which is what the economist is interested in
      • a. The economist does not weigh the real value of different objectives
      • b. Love of virtue against love of food.

      7. Note the argument he is answering: That no two pleasures are really equal in all their characteristics, so economists cannot compare them.

    C. Problems in quantifying motive by money measures.

    • 1. A given amount of money represents different amounts of pleasure etc. to different people, or the same person under different circumstances.
      • a. Two people with the same income might get different benefit from its use, different pain from losing some of it. In other words, different people have different utility functions.
      • b. But if we average over many people, the money which people of equal income will give is a good measure of benefit or injury.
      • c. On average, a shilling is worth less the more money you have.
      • d. But if we average over many people, this difference too will average out, provided ...
      • e. We have no reason to believe that one group of people are on average richer than another.
      • f. Most events economics deals with affect all classes of society in about the same proportion, so it is legitimate to take money measures as proxies for aggregate happines measures.

      2. We need not assume that all actions are deliberate and calculated.

      • a. Economic behavior tends to be more so than behavior in general, and ...
      • b. When people act without immediate calculation, it is likely to be on habits and customs that have arisen from past calculation, observation of how different behavior turns out.
      • c. In backward countries, there may still be many customs that survive although not functional, but ...
      • d. In modern business such habits rapidly get eliminated.

    D. Money or income or purchasing power or ... is the center that economic science clusters around, not because it is the main aim of human effort but because it is the one convenient way of measuring human motive on a large scale.

    • 1. Hence the attacks of Carlyle and Ruskin on economics were misdirected,
    • 2. Since they were supposing that economists assumed that the only motive that mattered was the desire for wealth, or sordid selfishness.
    • 3. And indeed, economics takes much account of non-pecuniary costs and benefits (consider Smith on equal net advantage in different occupations).
    • 4. To the extent that economists ignore some motives, such as (usually) altruism, it is not the lack of will but the lack of power
      • a. It is hard to reduce such motives to law and measure them, although not wholly impossible
      • b. (and Becker has been working on the problem).

    E. Considering individual or society.

    • 1. Perhaps early economists looked too much to motives of individual action.
    • 2. But individuals are part of a social organism, etc.
    • 3. And economics should be concerned with collective ownership, collective persuit of aims, etc.
    • 4. And Marshall thinks that improvements in society, intelligence, communication, etc. are widening the scope of collective action for the public good
    • 5. And that the spread of co-op movement etc., under motivfes other than pecuniary gain, are opening up the possibility of measuring such motives.

  III. Chapter III: Economic Generalizations or Laws

  • A. Division between experimentalist and theorist.
    • 1. In some contexts learning facts is crucial,
    • 2. In others figuring out how to fit them together.
    • 3. Hence a role for both kinds of workers.

    B. Successful physical sciences are not all exact sciences, but they aim at exactness--i.e. generating predictions that can be tested by experiment.

  • C. Nature of economic laws
    • 1. Tendencies--if nothing interfers with them.
    • 2. In astronomy, the tendencies generate very precise predictions, but ...
    • 3. In predicting the tides, much less so.
    • 4. Economic laws are more like the laws of the tides
      • a. Human behavior is very complicated, so the best we can do is inexact and faulty
      • b. Which is no reason not to do it.
      • c. As carefully as we can.

      5. What obeys human laws is called "legal;" what obeys scientific laws is called "normal."

      • a. Any use of "normal" implies the predominance of certain tendencies which are likely to be persistant over those which are relatively exceptional
      • b. And depends on context--illness is abnormal, but so is a long life with no illness.
      • c. And so for normal economic action--it must be qualified by the context within which it is normal.

      6. The term should not be limited to free competition, since

      • a. We may want to apply it to circumstances where free competition cannot exist (the normal behavior of a natural monopoly, for instance--my example, not Marshall's.)
      • b. And lots of economics depends on background norms that are not themselves part of free competition, such as norms of honesty.
      • c. Whose existence affects what economic behavior is normal

      7. Nor does normal mean "morally right."

      • a. There may be normal results of economic laws that we disapprove of and want to change, such as poverty, just as ...
      • b. There are normal consequences to swallowing poison.

     

• Book III Chapter III: Gradations of Consumer Demand
  • A. All goods are ultimately destined for consumption, so it is the ultimate consumer's demand that drives demand, even though some immediate demand is for intermediate goods.
  • B. Diminishing marginal utility of any one good.
    • 1. Introduces the idea of marginal utility--utility of the marginal purchase (the analogy to Ricardo's marginal land is still implicit at this point)
    • 2. Assumes we are comparing the utility of different quantities at the same instant, so ...
    • 3. Not contradicted by the possibility that consumption now might increase utility of future consumption, as with Music etc.
    • 4. Marshall notes that the law does not strictly hold--marginal utility could conceivably increase for a while with increasing quantity--and explicitly draws the parallel with production inthe footnote.

    C. Implication for price:

    • 1. Individual buys up to the point where the marginal utility of the last bit purchased equals its price.
    • 2. This holds the marginal utility of money fixed--since we are considering alternatives for the same person at the same time.
    • 3. Varying incomes, the same price represents a greater marginal utility when you are poor than when you are rich
    • 4. But in each case, the marginal utility at the quantity consumed is the utility of the price.

    D. Invention of the demand curve. (?)

    • 1. We could represent how much someone would buy at any price by a demand schedule.
    • 2. And must in order to have a clear concept of "how much he wants it."
    • 3. To say that someone's demand for something increases means that the curve shifts out.
    • 4. Even if demand by one individual is discontinuous (one car, or two cars, but not 1.7 cars), the aggregate demand schedule is still almost continuous.
    • 5. And even if there is a large random element in individual demand, it averages out in aggregate demand.
    • 6. So in a large market, every fall in price results in an increase in quantity demanded.
    • 7. Which is the law of demand--demand curves slope down
    • 8. Although we have no general theory to tell us by how much.

    E. Digression on defining commodities:

    • 1. A drop in price of all teas might reduce quantity demanded of some varieties, because people substitute others.
    • 2. The line dividing commodities is drawn for convenience.

     

  Chapter IV:

  • A. Elasticity of demand explained.
    • 1. The idea
    • 2. Applied to aggregate demand, but ...
    • 3. The pattern of demand is likely to depend on income, so it is useful to think of the demand of one social class at a time, then add them up.

    B. Pattern

    • 1. Usual pattern is that demand become inelastic as price becomes so low that people approach satiety.
    • 2. A possible exception is status goods.
    • 3. And we would expect that "so low" would be higher for richer people.
    • 4. Examples:
      • a. Salt is cheap, and demand inelastic, for almost everyone
      • b. Meat is cheap for the rich, not for the working and lower middle class--which greatly outnumber the rich, so demand is still pretty elastic.
      • c. Sugar has fallen to the point where demand is inelastic even for the working class.

      5. Marshall notes again the question of defining the commodity:

      • a. Demand for beef is very elastic, if we hold the price of mutton fixed, since they are close substitutes.
      • b. But demand for meat is not as elastic--since that means looking at the effect as we change prices of all meats together.

      6. Necessaries--wheat, assuming it is the cheapest food--may have quite inelastic demand throughout the price range.

    C. Water:

    • 1. Discussion of where the elasticity comes from at various incomes.
    • 2. What about for us?
    • 3. Note that the demand for water is itself a sum of demands for different uses, each elastic in a different price range.

    D. Problem in actually observing demand curves.

    • 1. Things are not constant over time.
    • 2. Purchasing power of money changes--but that can be observed and controlled for ("1970 dollars")
    • 3. Incomes fluctuate from good years to bad.
    • 4. Tastes change, new substitutes become available, ...
    • 5. And there are time lags in people's response to changed circumstances. So we might misattribute to the later fall in price the lagged response to the earlier: increasing use of petroleum (kerosene) lamps as people get use to them (after an initial fall in the price of kerosene) might be misattributed to the continuing fall in price.
    • 6. And substitution over time, so a high price may mean postponing purchase of a durable item until the price falls again. Consider the response to the"Cotton famine" of the1860's. In the short run, people postponed the purchase of clothes--but they couldn't keep postponing it forever (at least, not in the English climate). So eventually, quantity demanded went back up.

    E. Additional statistical problems

    • 1. Lots of transactions are not observed
    • 2. Markets are not well defined.
    • 3. Data tends to show purchases by retailers, not by consumers.
    • 4. Quality changes may cause the appearance of price changes--paying more for what is really a different and better good is not a price rise.

     

• Chapter V:
  • A. Allocation of a good among alternative uses:
    • 1. You want to distribute it in such a way that no reallocation would make you better off, which means
    • 2. Keep adjusting until its marginal utility in each use is the same.
    • 3. The equimarginal principle.
    • 4. Applies both to intermediate goods (yarn) and to consumed goods (water).

    B. Trade:

    • 1. With self-sufficient households, one may have so much yarn that the marginal utility is low, but not much wood, so MU of wool is low, of wood is high, while another household has the opposite situation
    • 2. Creating the opportunity for gains from trade.
    • 3. Which can be done by barter in a simple economy.
    • 4. Money in a complicated one, at which point

    C. The same argument implies that marginal utility per shilling will be the same for all goods consumed.

    • 1. Which is how people adjust their expenditures, why they keep household accounts, etc.
    • 2. Marshall notes that household production is important as well--the French housewife can produce better food from cheap ingredients than the British or American.

    D. Allocation is not only among present uses of income but also between present and future.

    • 1. Here again, we optimize by making the marginal utility of present and future expenditures the same, but...
    • 2. we must allow for the uncertainty of the future outcome--I might die before buying the G3 powerbook I am saving for, or Apple might go out of business.
    • 3. And the preference for present over future pleasures.
    • 4. Which is a function of individual taste, and varies from one person to another.

    E. Analysis of time discounting

    • 1. If individuals were indifferent between present and future pleasures, they would spend roughly the same amount per year.
    • 2. But most prefer present to future consumption.
    • 3. Which influences both saving/borrowing behavior and ...
    • 4. Choices between durable and less durable consumer goods.
    • 5. Also, some people get pleasure from ownership
      • a. For its own sake
      • b. or for distinction
      • c. So that land may yield a low pecuniary return in equilibrium.
      • d. My japanese weapon story. I should have bought the nagamaki, because even if it was only so-so as an investment, and not quite worth the price as a source of a flow of utility from ownership, the sum of those two values was more than the cost.

      6. We can observe how someone discounts future money amounts but not how he discounts future pleasures, since the same amount of money might produce different amounts of pleasures at different times, circumstances, etc.

    • 7. But we can estimate his discount rate on utility if we assume
      • a. That he expects to be about as rich in the future as now, and ...
      • b. To have about the same ability to convert money into pleasure
      • c. In which case his discount rate on money is roughly his discount rate on utility.

      8. His discount rate on money will be the market interest rate--because of the equimarginal principle applied through time.

    • 9. But that is not automatically his discount rate on utility, since he may expect changes in MU of income over time.
    • 10. You will find a more extensive explanation of all of this in my (webbed) Price Theory book.

     

• Chapter VI: Value and utility
  • A. What is the relation between price paid and benefit conferred?
    • 1. We never pay more than something is worth to us, but often less, in which case the difference is consumer's surplus.
    • 2. A benefit derived from your opportunities, or your environment.
    • 3. Marshall shows why consumer surplus from buying up to P=MV is the area above the demand curve.
    • 4. If tea were not available, he would be worse off by the lost consumer surplus.
    • 5. What about the consumer surplus on what he would buy instead? He would spend a little more on each other good--and marginal purchases yield no consumer surplus (just as marginal land pays no rent).
    • 6. So if we ignore differences in income, we could aggregate consumer surplus for the whole market to measure the benefit from the availability of tea.

    B. Consumer surplus seems a difficult and artificial concept, but ...

    • 1. It merely makes explicit and clear what was implicit in the way we thought about gains from consumption.
    • 2. Things being worth more than they cost.
    • 3. Note the earlier step linking the demand curve (how much you buy at any price) with the marginal value curve (how much a little more is worth to you, as a function of quantity), which makes consumer surplus observable.
    • 5. Just as earlier, Marshall justifies aggregating values by arguing that most of the issues economists are interested in affect large and heterogenous groups with similar composition.
    • 6. A further complication is that as you buy more of something, your MU of money rises, so in adding up the consumer surplus rectangles (the benefit measured in money equivalent from consuming five bags of tea instead of four, four instead of three, ...) we are aggregating not quite comparable units
      • a. But in practice this is a small effect
      • b. Which is made explicit in the modern distinctions between income compensated and uncompensated demand curves.
      • c. An exception might occasionally occur, as with a Giffen good.

     

• Book V, Chapter 3: Equilibrium of demand and supply
  • A. The distinction between cost of production and expenses of production
    • 1. Cost is social cost--how much leisure is given up, goods used for this purpose that would be valuable in other uses, etc. "efforts and sacrifices."
    • 2. Expenses of production are the actual expenditures that must be made by the producer in order to get the good produced.
    • 3. Note that these are conceptually quite different things, and might be actually different
      • a. For instance, if the work is done by slaves or conscripts
      • b. Or if there are externalities
      • c. Or if some inputs are sold by a monopoly, for a price higher than their cost of production or value in any alternative use

      4. And they are easily confused in economic analysis:

      • a. For instance, if we argue that a firm will produce its output in the lowest cost way possible, thus giving an efficient result
      • b. The premise is appropriate for expenses, but ...
      • c. The conclusion is about costs.

      5. Making one important issue the question of when the two are actually equal.

      • a. An important step in demonstrating the efficiency of competitive equilibrium
      • b. Via price equal to marginal cost and marginal value on every margin
      • c. For details, see my (webbed) Price Theory book.

    B. Production cost will depend on quantity being produced, and might increase or decrease with quantity.

    • 1. Increase because you need to bring inputs from farther away to get more, etc.
    • 2. Decrease because of machinery made practical by large scale production.

    C. And note that a firm will choose that set of inputs that minimizes the expenses of producing what they produce, so as to maximize their profits.

  • D. Marshall's current assumptions:
    • 1. Competitive market--no cartel agreements
    • 2. Everyone is well informed about market price, so only one price for one commodity at one time.
    • 3. He assumes that the supply price of a commodity will be equal to the expenses of production in a representative firm--P=MC. This is an assumption because he hasn't gotten to analyzing the supply side of the market yet.
    • 4. Supply curve is rising or horizontal--he'll discuss the complications of a falling supply curve later.
      • a. Why is a falling supply curve an odd case?
      • b. Because competition requires multiple firms, but if there are scale economies all the way out to the size of the market there should only be one firm.
      • c. So horizontal or rising supply curves fit naturally with a competitive market, falling supply curve is a puzzle.

    E. Equilibrium price:

    • 1. Both demand price and supply price depend on quantity
    • 2. If quantity is such that demand price is higher than supply price, producers have an incentive to produce more, if lower, less.
    • 3. So quantity moves to where demand price=supply price
    • 4. Which is the intersection of supply and demand on a conventional diagram.
    • 5. And it is a stable equilibrium--shifts away from the equilibrium quantity result in forces moving the system back to the equilibrium quantity.

    F. Is price determined by supply or demand?

    • 1. Which blade of the scissors cuts the paper?
    • 2. If quantity is very inelastic (fish in the fish market, once deliveries have been made, or a rare book sold at auction), then changes in demand are responsible for changes in price.
    • 3. If quantity is very elastic (constant cost of production--Ricardo's normal case), then long run price is always the cost of production, whatever demand is.
    • 4. As a general rule, demand changes more in the short run than supply, so if we are considering short time intervals it makes sense to think of price as mostly determined by the effect of varying demand on constant supply.

     

  Chapter VIII: Marginal Cost in relation to values

  • A. Principle of substitution --> equimarginal principle in production
    • 1. Producer tries to find the least expensive way of producing his output
    • 2. Which implies that, on the margin, net output per shilling of input should be the same for all inputs employed
      • a. Application to horse vs steam power
      • b. Skilled vs unskilled labor

      3. He also has the option of increasing one input, holding others constant (or changing them to allow for the new one)

      • a. His gain is what he sells the new output for, minus the cost of any other additional inputs
      • b. His cost the cost of the expanded input
      • c. So as long as the former is larger than the latter, he will use more of the input, so
      • d. At the optimum, marginal productivity of each input times price of output=marginal revenue product of that input=price of that input.

    B. Note that the principle of diminishing return has at least two related but distinct meanings/contexts

    • 1. Using too much of one input relative to another, when you can buy as much as you want of each--suboptimal production strategy.
    • 2. Ricardo's sense--diminishing returns in agriculture due to limited amount of fertile land.
    • 3. But mathematically, these are really the same--2 would be 1 if there were more equally fertile land available, so that we could increase all inputs together--including land.
    • 4. One could imagine an industry where total return increased as you scaled up all inputs, but diminishing return still applied if you held one fixed and increased another.

    C. Confusion with regard to the implications of the equimarginal principle in production: Marginal revenue product=input price.

    • 1. It is not that the marginal use determines the price
      • a. Suppose the marginal use of water is for irrigation
      • b. A new innovation makes irrigation twice as productive as before
      • c. Does the price of water double?
      • d. No--more water is used for irrigation, the price rises, you are now irrigating less valuable crops than before (on the margin). The margin has moved.
      • e. Input price still equals MRP, but is not necessarily twice the old value.

      2. But rather that you look at what is happening on the margin in order to understand what is determining price--namely supply and demand equilibrium, i.e. solving a set of equations, one of which is P=MRP.

    • 3. If you took some water away from some use other than the marginal one, the ultimate result would be exactly the same--because water would be shifted from the marginal use to the one which now had a higher than normal MRP.
    • 4. Similarly, a loss of steam from a steam engine has the same effect on pressure, whether or not it comes out through the safety valve-but the safety valve happens to be where the steam will come out from, just as the marginal use happens to be the one that will be affected by changes in the quantity of the input used.

    D. Machinery vs land--determinants of price

    • A. Machinery can be produced, wears out, so price usually near cost of production
    • B. But land, or things representing sunk costs, yield quasi-rent, where floating capital yields interest.
    • C. With no sharp line of division.

     

  Chapter IX

  • A. The use of taxation as a device to illustrate economic arguments (as in Ricardo)
    • 1. A tax on production tends to check production, which raises price, which shifts some of the tax forward to the consumers and a little back to the producers of the inputs used in producing the taxed commodity, and similarly
    • 2. A tax on consumption tends to reduce demand, which lowers price, which shifts some of the tax backwards to the producers (and a little farther backwards to input producers, although Marshall doesn't say so).
    • 3. How much shifts which way depends on elasticities: Suppose a tax on printing
      • a. Specialized resources, such as printing presses, would have low prices for some time.
      • b. Less specialized resources, such as steam engines and clerks, could shift to something else when their price/wage fell, so it wouldn't fall very far.

      4. Suppose the tax was only a local one:

      • a. If there were no compensating effect (how is the tax money being spent?), then inputs that were specialized but mobile would move out
      • b. Leaving the burden to be born by immobile resources, and ...
      • c. Local merchants.

      5. Suppose it were on printing presses

      • a. It would be paid from the quasi-rents of the presses in the short run, assuming there were no presses so bad as to be worth scrapping given the tax
      • b. But over time, presses would wear out and not be replaced, until eventually ...
      • c. The tax was fully transferred to printers, customers, etc., via a higher price of presses.

     B. Suppose a fixed supply of meteoric stones harder than diamonds falls:

    • 1. Total supply is fixed, but ...
      • a. We can expand on the intensive margin--run the factory night and day
      • b. At an additional cost, so we observe declining returns on the intensive margin.

      2. Their price:

      • a. At any instant, equal to expected future revenue from owning one, and at that price you can buy as many as you want, but ...
      • b. Once you buy them, revenue might turn out more or less than expected.

      3. Next suppose some are scattered, can be found with searching

      • a. We now have an extensive margin too.
      • b. And will go out it until the additional search cost of finding one more just equals its value.

      4. Finally, make them brittle, assume an inexhaustible supply to be found at constant search price, and ...

    • 5. We have gone all the way from the pure land extreme to the pure produced good extreme.
    • 6. Or in other words, Marshall has used his imaginary meteoric stones to illustrate that Ricardo's analysis of land and Ricardo's analysis of produced goods simply represent the two extremes cases in a continuum of possibilities.

    C. Actual goods do not represent pure forms of "land," "capital," "labor," ...

    • 1. The rent on land is in part the return from capital investments in it (as Ricardo pointed out)
    • 2. Wages include an element of rent (on scarce human abilities, presumably, although Marshall doesn't explain)
    • 3. Economists have learned to separate out the different pure elements mixed in the actual things we observe
    • 4. And to analyze what happens to actual things by combining the effects on the elments.
    • 5. Thus (my example, not Marshall's) an increase in long run interest rates will eventually raise the wages of professions that require long training, since part of that wage is a return on capital.

    D. Scarcity rents and Differential rents

    • 1. There is no fundamental difference
      • a. Ricardian rent on land is equal to the difference between the productivity of the good land and of the marginal land, but ...
      • b. The marginal land would not be in use if the good land were not scarce
      • c. If there were no "marginal land," we would could still calculate differential rents by considering marginal production on the good land--i.e. the intensive margin.

      2. Marshall goes through the logic of thinking of the problems both ways for his meteoric stones.

    • 3. The idea that the existence of the inferior land tends to raise the rent of the superior is exactly backwards
      • a. If we eliminated the inferior land, we would have to push production even farther out on the intensive margin of the good land
      • b. Making rents higher, not lower.
      • c. It isn't the existence of the inferior land that causes rent, but the nonexistence of (more of) the superior land (my point, not Marshall's).

  Chapter XII: Increasing Return

  • A. Background: [from Chapter 2, not read]: "The law of increasing return may be worded thus: -an increase of labour and capital leads generally to improved organization, which increases the efficiency of the work of labour and capital"
    • 1. If this effect were true within the firm, it would lead to a natural monopoly
    • 2. And in places, Marshall seems to be imagining what we now call monopolistic competition--an industry made up of many firms, each of which is to some degree a monopoly in its market niche--geographical or defined by product characteristics.
    • 3. But the easy case is where the economy of scale occurs outside the firm but inside the industry. The bigger the industry, the more opportunity there is for people to invent specialized machines to sell the firms, to learn new things about how to produce the goods, etc.
    • 4. Making it a special case of what we now call a positive externality.

    B. Imagine a sudden fashion for wrist barometers.

    • 1. In the short run, the price goes up, but ...
    • 2. It would go back down again, as more workers learned how to make them, and ...
    • 3. Might eventually go lower than it started, as the scale of production led to greater division of labor, lower cost, etc.

    C. Note the assymetry between demand and supply

    • 1. Demand curves always slope down (unless you count the case where a low price makes something no longer a status good), supply curves may slope up or down.
    • 2. And the elasticity of demand curves is usually similar in the short run and the long run (save, presumably, the very short run in which people can put off a purchase until the price comes back down)
    • 3. But while an increase in price always increases supply, the long run effect might be (in the case of increasing return, i.e. a falling supply curve, will be) for the increased quantity supplied to decrease price below its original level.
    • 4. So elasticity of supply works like elasticity of demand in the short run (and that is the analysis you have gotten in your price theory course) but may be quite different in the long run.

    D. Note also the difference between what happens to a firm and to an industry.

    • 1. Firms have a life cycle, associated with the life of the founder, although it may be prolonged if he manages to find a competent successor.
    • 2. But the industry is the sum of lots of firms in different parts of their life cycle at any given time.
    • 3. And the pattern of cost for a firm may be quite different than for an industry
      • a. In the standard modern picture of perfect competition, for example, we can get an industry with a horizontal supply curve out of a lot of firms all of which have rising supply curves--because additional demand pulls more firms into existence.
      • b. And Marshall offers a different set of examples.

      4. Note also that "supplementary costs" (roughly speaking, our "fixed costs") are likely to be large relative to "prime cost" (roughly speaking, our variable cost) for industries with increasing return, since they are industries with lots of fixed capital etc.

    • 5. As total output goes up, Marshall imagines the size and efficiency of the typical firm rising.
      • a. Possibly because it is in monopolistic competition?
      • b. Or because a larger industry means more specialization, specialized tools available, etc., hence larger economies of scale at the firm level.

  Chapter XIII: Relation to the doctrine of maximum satisfaction

  • A. The issue here is to what degree the market outcome is in some sense optimal--maximizes the total satisfaction of the people involved.
    • 1. An issue still of great interest
    • 2. And which we now describe in terms of the "efficiency of competitive equilibrium."

    B. The analysis must distinguish among goods with constant, rising, and falling supply curves.

    • 1. One possible implication of the falling case is that we could benefit by a protective tariff,
      • a. since it would increase domestic demand, thus quantity, thus decrease price in the long run.
      • b. This could work in principle, but note that ...
      • c. The tariff decreases demand for other domestic industries (the ones producing the export goods that were exchanged for the import competing goods), so if they have declining cost curves too, you are getting the opposite effect there.
      • d. And in practice, infant industries have fewer votes, less political influence, than mature industries, so you are likely to get a perverse pattern of tariffs.

      2. Consider the efficiency of the market outcome, with each kind of supply curve, by imagining the effect of a tax or subsidy on such an industry

      • a. If there are kinds of industries where a tax costs less in consumer surplus than it brings in in revenue
      • b. And kinds where a subsidy produces more increase in consumer surplus than it costs in expenditure
      • c. Then taxing the former and using the money to subsidize the latter will produce a net benefit, so ...
      • d. The market outcome is not optimal, and can be improved by government intervention.
      • e. (Why is Marshall only considering consumer surplus, and not producer surplus, at this point?)

      3. For an industry with a horizontal supply curve, the effect of a tax or subsidy is a net loss, for the standard excess burden reasons--consumer surplus with a tax falls by more than the amount collected, consumer surplus with a subsidy rises by less than the amount spent.

      • a. Marshall provides both a verbal version of the argument
      • b. And a graphical version in the footnotes.

      4. With a rising cost curve, a subsidy is even worse, but a tax lowers cost, so it is logically possible for consumer surplus to fall by less than the amount collected.

      • a. At this point Marshall is ignoring producer surplus
      • b. Which must exist if the supply curve is rising, and is decreased due to the tax
      • c. Which eliminates this argument, as it turns out.

      5. With a falling cost curve, a tax is worse than in the constant cost case, but a subsidy might produce a net benefit, by pushing quantity up and thus cost down.

      • a. This is a correct argument in modern terms
      • b. And corresponds to what we would describe as a negative Pigouvian tax to encourage production of something that produces a positive externality.

    C. The argument for maximum satisfaction is in part right, in part wrong:

    • 1. Right in pointing out that the market transactions that occur between individuals are precisely the ones that produce a net gain for each. I only sell you something if it is worth more to you than to me--in which case we both gain by the transaction
    • 2. But it does not follow that competitive equilibrium maximizes net satisfaction
      • a. First, because utility per dollar of gain might be different for the different parties, so a $2 gain for one combined with a $3 loss for the other might be a net gain in satisfaction.
      • b. Thus if producers were much poorer than consumers ...
      • c. More generally, redistribution from rich to poor can increase total satisfaction.
      • d. (Note that, as a matter of theory, this is less clear than Marshall supposes)
        • i. If rich and poor have the same tastes but different opportunities, then declining marginal utility implies that the rich person has a lower marginal utility of income than the poor--same utility function, different income.
        • ii. But if rich and poor have the same opportunities but different tastes, we get the opposite result: Rich people are rich because they very much enjoy income, therefore work very hard to get it.
        • iii. And the marginal utility of income is actually higher for the rich.
        • iv. Version i seems more plausible than version ii, but both are logical possibilties.

      3. And in the case of increasing return (falling supply curves), it is possible that a suitable combination of taxes and subsidies would produce a net gain for all concerned.

    • 4. On the other hand, there are a lot of practical problems, such as "the danger that in the trade which had got a bounty and in other trades which hoped to get one, people would divert their energies from managing their own businesses to managing those persons who control the bounties."
    • 5. So the conclusion that letting each person act in his own interest is best for all must be qualified twice
      • a. There is a net gain if someone spends his money in ways that benefit poor rather than rich, even if doing so does not maximize his benefit.
      • b. And if he spends his money on industries with falling supply curves.

      6. So we might want to tax commodities with declining return, subsidize ones with increasing return...

    • 7. Or perhaps not-Marshall has raised a possibility, but is not sure of its real world relevance. Further work remains to be done.

   Chapter XV: Summary

  • A. In the very short run (higgling in the market), production cost is almost irrelevant since nothing gets produced that fast, so price is determined by quantity available and demand (i.e., there is an almost perfectly inelastic supply curve in the very short run)
    • 1. Although this slightly oversimplifies, since sellers can carry goods over to tomorrow, so production expected to occur tomorrow is not wholly irrelevant
    • 2. But pretty nearly so, if goods are very perishable--for example a fish market.

    B. In a perfectly stable equilibrium, long run and short run costs are the same, and equal to price.

    • 1. It is useful to think of the length of period as divided into two categories
    • 2. long periods, for which any temporary supply shortages have had time to work themselves out, and ...
    • 3. Short periods, in which the supply of machinery, inputs, etc. is fixed, although the producers have time to adjust their behavior to those supplies, demand, etc.
    • 4. Roughly speaking, our "long run" and "short run"

    C. Marshall discussed (in chapters we did not read) the joint consumption, joint production problem--Smith's wool and mutton, for example, or cars and gasoline.

  • D. Also in chapters we did not read, he discussed producer surplus--which is a generalization of Ricardian rent.
    • 1. Price must cover marginal cost of production.
    • 2. People who own resources that give an advantage in production thus get income above cost.
    • 3. Which is a rent on their scarce productive assets--like rent.

    E. Relation between production cost of productive facilities and value of what they produce--Ricardian rent generalized

    • 1. Price must cover production cost on the worse facilities that are in use (Ricardo)
    • 2. In the long run, sunk capital's return should (ex ante) cover its cost
    • 3. But in the short run it acts like rent.

    E. Analysis of increasing return

  • F. And the question of taxes and net satisfaction.
    • 1. The argument to show that laissez-faire produces the maximum total of human satisfaction is weaker than previous economists thought
    • 2. Both because of the effect of differing marginal utility of income between rich and poor, and
    • 3. Because it may be possible to increase aggregate satisfaction by taxing industries with rising cost curves in order to subsidize those with falling cost curves.
    • 4. An early version, in part, of Pigou's argument for subsidizing products with positive externalities.