Wednesday, December 21, 2011

A thought on a Nick Rowe post

In this post, Nick Rowe proposes that macroeconomics went wrong when we stopped thinking about the trade cycle and started focusing on newly produced output.

There are some senses in which he's right. For example, we know we have reason to worry about the production of investment goods because they are particularly volatile. But of course, the volatility of investment goods is intimately tied to existing investment goods: the capital stock. Likewise, we know that wealth levels are very important to worry about because they too are closely tied to aggregate demand by both consumers and investors. And once again, wealth is a collection of new and old goods, and a lot of wealth is tied up in the trading and retrading of securities: not new production.

So old production is obviously very important, even for we who have gone off track. However, I hesitate to agree with Nick in fully embracing the idea that we even have gone off track. New production is important because new production is very closely related to employment in a way that trade in old goods isn't. For there to be employment there typically must be a buyer, a seller, a medium of exchange, and the exchange of some new production. If we only have a buyer, a seller, a medium of exchange, and the exchange of some old production it's very likely that we'll have little employment (perhaps a broker of some sort... and even he will be producing something new, namely brokerage services).

We care about employment. Why? Because that's how most of us get money. And why is getting money important? Because that's how we get very important goods and services (old or new) in the modern economy.

If you want economics to be a science of how markets and prices and exchanges work and how prices and quantity sold move and change, then Nick is probably right.

If you want economics to be a science of human economic activity, explaining the sorts of economic facts that we care about, there is very good reason for the discipline to have shifted its focus to new output. This is not to say that we can ignore "the trade cycle" as Nick describes it - trade in all goods. We can't ignore that. But if what you care about is employment and human behavior to satisfy human wants, there is very good reason to highlight the role of new output.

Nick is not a friend of Say's Law, but he thinks about the economy in the same way - as a web of exchanges. There's something to that way of looking at the economy. There's something to asking "when is this system balanced and what causes it to become unbalanced?" That's all fine. But even when a balance is struck beetween sales and purchases, there's no guarantee at all that it will be struck at a level that we consider good or healthy.

8 comments:

  1. I rather like the "web of exchanges" view....

    What we have to remember is that employment isn't good for it's own sake, or indeed because it paying for it creates an income for someone.

    Isn't what we're interested in whether the output produced is really worth more than the input?

    There are wealth distribution issues, but aren't they somewhat separate?

    ReplyDelete
  2. I agree, getting money to buy the stuff you want is important. We get money by selling stuff - most of us sell labor - and use it to buy all the other things we want.
    Let's assume a decline in Y without a decline in T. T is the amount of sold goods, new and old. Obviously people earn less money by selling new goods and by selling labor, but they get more money by selling old goods. For a representative agent that doesn't matter. What does matter is total value added. Y is only part of that. If total value added is the same, there's no reason to call that a recession.
    There is of course the distributional problem that many have only labor to sell.


    What we really care about is not Y, but the fulfillment of human needs and wants. We use Y as an indicator for that. That's why we need to examine carefully, whether policy we propose to increase Y does so by driving a wedge between the two or by improving what we care about. And we shouldn't call periods recessions, if all they do is driving a wedge between the two.

    The same is true about unemployment. We don't care about the unemployment rate, we care about the condition of the (potential) unemployed (or jobless, if you want).

    ReplyDelete
  3. Daniel: Thanks for the response.

    "But if what you care about is employment and human behavior to satisfy human wants, there is very good reason to highlight the role of new output."

    Suppose all output were labour services (actually, most output *is* labour services). In a recession, we might all be working just as hard as in a boom, fixing our own cars, cutting our own hair, cooking our own food, doing our own plumbing, etc., but we would be worse off. Because we wouldn't be consuming or producing the things we wanted to produce and consume. In a recession, individuals go part way back towards autarky.

    Recessions are bad because we stop trading as much, and so can't get all the gains from trade. Not because we stop working as much.

    ReplyDelete
  4. Franz - exactly.

    In this:

    http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1898962

    Anthony Evans and I talk a lot about the difference between Y and T. We don't go onto the relationship to Say's law and Keynesianism though.

    ReplyDelete
  5. Nick -
    Right, but the reasons why we can't trade labor in the market as much (which is what a lot of people are really interested in) is closely related to the demand for NEW goods.

    ReplyDelete
  6. Daniel,

    "Right, but the reasons why we can't trade labor in the market as much (which is what a lot of people are really interested in) is closely related to the demand for NEW goods."

    That's what Post-Keynesians and Dinosaur Keynesians mean when they say "labour is a derived demand". And on that they are right, though no exact proportionality can be expected between one and the other.

    But I think we also have to pay attention to why the level of demand for new output is lower. Keynesians such as yourself seem to concentrate on a set of interlinked reasons. Irrational uncertainty is causing a rise in demand for existing assets (such as bank balances, base money and government bonds).

    As I've said before, what I want to know is why you think this is the only reasonable conclusion in any practical situation.

    ReplyDelete
  7. I'm not sure it is the only reasonable conclusion in any practical situation.

    What I think Keynesian insights tell us is that there is no reason to believe we are ever producing at our potential. Why we fluctuate around that sub-optimal level could of course be caused by many things. I'm on record as saying ABCT is one of those things. Certainly real shocks and supply shocks can have an impact too.

    I think you can also say that Keynesian insights about why the economy is not guaranteed to be at full employment can also illuminate certain recessionary episodes that play on factors like liquidity preference and investment demand. The fact that I see Keynesianism as primarily being an explanation of output and employment levels is not to say that I think it's irrelevant to thinking about the business cycle. But I don't think I've ever promoted the idea that it's the only game in town on business cycles (at least I have no business promoting that idea).

    ReplyDelete
  8. Would you say though that we can be far away from the optimal level of output for a long time?

    I've been thinking about this quite a lot and I think it comes down to what's meant by "long-run" in any particular situation.

    The question of expectations is also difficult to ignore and complicated. Because of expectations I don't think the argument leads directly to fiscal stimulus. Consumption is another tricky problem for fiscal stimulus.

    Anyway, I'm quite busy with some economics of my own at present I'll discuss this more with you after Christmas.

    ReplyDelete

All anonymous comments will be deleted. Consistent pseudonyms are fine.