Aggregate Supply and Aggregate Demand

Sometimes we see people who view themselves as "pro-Say's Law" arguing that "Keynesians are wrong about growth being demand driver: supply has to come before demand."

This is odd, to me, because, while I think it is questionable whether Say's Law always holds (unless one states it as a tautology, in which case, it will always "hold," but trivially so), it certainly is based on a genuine insight: people supply commodities on the market in order to demand others: supply IS demand, just seen from a different perspective. But once one gets that, then it is clear that neither supply nor demand can come first, since they are simply different views of a single action: offering something in exchange for something else.

So, why even separate them in macroeconomic analysis? I am mulling this over, but my current suspicion is that problems may arise with one or another aspect of this single action. For instance, if taxes on income over $10,00 were raised to 99.99%, and the economy slumped, we would point to a supply-side problem, in the sense that everybody knows everybody wants more stuff, but the effort of producing doesn't seem worthwhile at such a punitive tax rate. On the other hand, if the economy tanks even though there are few barriers to supply, except for the fact that eight out of every ten people has become a Buddhist monk and taken a vow of poverty, we can notionally separate aggregate demand and aggregate supply and say, "This is a demand-driven slump."

Well, that is what I think today, at least.


  1. Stockpiles?
    Says law certainly holds for some things. Sex. It seems like it ought to hold for conversation, but then people learned to stockpile their words -- books -- and suddenly it got more complicated even for that.

    If you can stockpile sex, and convince Salma Hayek to sell some of her surplus, you'll be rich.

    1. Ken, you've now got me thinking of the Hayekian triangle.

    2. Anonymous4:33 AM

      Now that was funny.

    3. Anonymous4:33 AM

      Now that was funny. And dirty, too.

  2. I think supply and demand analysis mostly exists because of money; to determine money prices.

  3. They're not really have two *curves* a supply curve and a demand curve. The quantity supplied at equilibrium is the same, which is why both curves say (Y=something depending on your model)
    Since they both equal Y, they're necessarily equal.

    When we talk about supply or demand causing something, we're talking about a shock that moves one curve or the other. We call a tax shock a supply side shock because we tend to model it as moving the aggregate supply curve, not the aggregate demand curve.

    A demand drive slump on the other hand is a shift in the aggregate demand curve. Usually the because we're saying that the marginal propensity to consume is declining.(and thus we get into the paradox of thrift)

    1. Unknown, I'm not sure you're getting the point that the demand curve IS a supply curve and the supply curve IS a demand curve.

    2. If that is your point, I'm pretty sure that it is incorrect.

      Say's Law says that the quantity supplied equals the quantity demanded at equilibrium. Not that the two curves(which each define a different set of possible worlds) are equal.

      Maybe you can elaborate further on what you mean? Because my understanding is pretty much the exactly the opposite of what you just said. How can they be the same? One has a positive slope, the other has a negative slope. (or maybe zero and undefined, definitely not equal...)

    3. "If that is your point, I'm pretty sure that it is incorrect."

      No, it is a standard part of economic education: I am not saying anything novel here.

      Look: my supply curve for labor IS my demand curve for money. My demand curve for the goods I buy IS my supply curve for money.

      And since money is just an intermediary I and others use to demand real goods, it means my supply of labor IS my demand for all the goods I consume, and other people's supply of those goods IS their demand for my labor and the other goods they consume.

      And if you're worried about slopes, just switch the axes: to change my supply of labor curve into a demand for money curve, put money on the x axis: bingo, the curve slopes downward!

    4. Take two continuous functions.
      One has an ascending slope.(dy/dx is positive at all points)
      The other has a descending slope.(dy/dx is negative at all points)

      They will be equal at at most one point. Period. Fact. Negating the axes doesn't fix the problem, because it flips both curves.

      I'm not disputing that one person's income from the labor that they supply is a factor in their demand, but it is a mistake to assert identity when you mean relation.

      Moreover, we're talking about macro, not micro. Your lead post is titled "aggregate supply and aggregate demand" In your explanation, however, you've switched contexts from macro to micro. To equate the curves in macroeconomic context is sweep the whole model under the rug.

      The *curves* are not equal because they identify a set of possible worlds. Only one is consistent with the real world, the one at which our equilibrium condition is met. ie the intersection. At that point and that point only will they be equal.

    5. "One has an ascending slope.(dy/dx is positive at all points)
      "The other has a descending slope.(dy/dx is negative at all points)

      "They will be equal at at most one point. Period. Fact. "

      Well, duuuuuh. Do you think I'm a moron? Do you realize I teach macro and draw such curves on the board every day? That I teach my students to look for the one point they intersect? Are you next going to tell me that adding two positive numbers together results in a positive number?

      "Negating the axes doesn't fix the problem, because it flips both curves."

      Did you perhaps mean reversing the axes? Yes, both curves will flip: and you will see how the supply of goods curve is at the same time a demand for money curve, and it will slope downward, and you will see how the demand for goods curve is at the same time a supply curve of money, and it now slopes upward.

      "it is a mistake to assert identity when you mean relation"

      And it is a mistake to think of something as a relationship when it is actually just the same thing seen from two different perspectives.

      Really, you are arguing against a standard insight people usually are expected to learn in their study of economics, and treating me like a moron because I have gotten some grasp of it. Here is Sheldon Richman making the same point. Here is Roger Sandilands making it. Here is Henry George.

      And this is a quote from the great economist Frank Fetter: "Demand is supply in another aspect."

      I am saying something very standard. It's you who ain't gettin' it, kid.

  4. I don't know the history of thought, but I'd think that it has something to do with money; i.e. price determination theory.

    1. Jonathan, check out my latest post: I try to show why money is not the fundamental "loose joint."


Post a Comment

Popular posts from this blog

Central Planning Works!

Fiat Currency

More college diversity and tolerance