Home > Uncategorized > Signals require realizations, or; Why Austrian-economics-types need to stop worrying and learn to love the inflation

Signals require realizations, or; Why Austrian-economics-types need to stop worrying and learn to love the inflation

Matt Yglesias has been on a roll lately and this morning’s installment doesn’t disappoint.   Go read the post.

I’d like to make a broader but related point; to sum up, a signal is a useful piece of information only because someone somewhere will act on it.   Granted this all sounds a bit Zen, so let me rephrase that before I get to my argument: if no one acts on a signal, then no one will have reason to think that anyone else will act, and so no one will do so.   This is what I mean when I say that “signals require realizations”.

The point is broadly applicable.   For example, take inflation, which is the subject of Matt’s post.   Austrian/hard money types like to point out that inflation doesn’t create jobs.   Matt counters, correctly, that jobs create inflation.   I agree.   In fact, not only do I agree, but (somewhat obliquely) so does Ben Bernanke.   The picture above is the “virtuous circle” from Frank and Bernanke “Principles of Macro” textbook.   The point I want to emphasize is that box in the lower left hand corner.

What Bernanke is trying to say with this picture, is that if you can keep wages rising only slowly (which is a direct consequence of slow hiring) than you can keep inflation low.   If you keep inflation low, than no one is expecting prices to rise and so no one is expecting wages to rise–therefore they do not.   This is an example of a signal that needs realizations.   The central bank would like to signal low inflation (in this example–there may be times when signaling high inflation would be beneficial… like now), to do this someone must refrain from raising prices, but if those same someones must pay higher wages to their workforce than they have no other choice but to charge higher prices.   Since a given firm is small compared to the economy as a whole, it takes the prices it can charge and wages it must pay as given.   Which means that sort of spiral can only be interrupted by making sure that the wage increase didn’t happen, but why wouldn’t they?   The answer is obvious, it’s called unemployment.

Prices as Information

The singular contribution of the Nobelist F. Hayek is the idea that prices (broadly understood to include input prices, like wages) reflect all the information necessary for production decisions.   Hayek got a lot of things wrong in his career, but this point is generally correct and really important (although I would argue misunderstood by the Austrian school and Hayek himself–I’ll get to that).    Why would firms produce more?   Because the price of their outputs have increased (at constant volume of sales), while the price of their inputs have not.

The point I am making, is that they will respond to these prices, but they must first see the price change (or equivalently, at constant price, they need to observe shortages).   The realization of the price/demand change is the signal to change production plans.

Change is Hard

There is a huge informational problem at the heart of any economic system (see here… I know I’ve linked to this article before, and for no particular reason, but seriously read it, its awesome).   To be in a “classical” equilibrium, would require all the firms, households, aliens etc to solve a massive system of simultaneous equations and then adjust prices and production accordingly–and constantly.   This doesn’t happen.   In the real world, firms are just responding to the prices/demand that affect them directly–they use local information, not global information.

If prices could adjust instantly, this would be no problem.   But why would we expect that?   If they did, then the “classical” economy would be the one we see.    But they don’t adjust; not necessarily.   There are lots of possible reasons, but not really a consensus.   For one, there are prices, like wages, that are set by long-run contracts.   Or there can be “menu costs” associated with changing pricing plans.   To me, it doesn’t matter much, prices do not adjust; the dollar menu at McDonald’s doesn’t become the $1.05 menu one day and then the $0.95 menu the next.

As I see it, though, all the usual focus on why prices don’t change is the wrong one.   Why would they change?   If you are a business owner, and you see that the demand for, say, Big Macs, is lower one day than it was on the previous; what is your first reaction?  Not to lower the price, almost surely.   Instead, your first reaction is that lower demand today is most likely just noise.   If the problem persists, then you need to think about changes, not before.

Inflation and Unemployment as Coordinating Devices

If no (other) prices are changing, being the only one who changes prices is costly.   If all other prices are changing, then not changing prices is costly.   Up to a point, inflation coordinates decisions to raise prices all up and down the production chain.    For car manufacturers to charge the “right” price, the steel mill must charge the right price as well and they must all pay their workers the “right” wage.   If prices are too low, then there will be shortages; inflation is the realization of the signal that there is a shortage somewhere.

In unemployment is high, then few workers will demand higher wages–they have much more competition for each opening.   If wages are not rising, then the most important input cost for firms is not rising.   Moreover, since products are paid for with wages (generally) then more unemployment is unlikely to mean more demand.   Output prices will not rise and production will likely be scaled back.   Unemployment is the signal to stop producing.   Eventually labor markets should clear, but wages would generally have to fall more than output prices; and why should they do that in any reasonable time frame?   (There is at least one price, the price of servicing debt, that never adjusts to changing circumstances–or it changes in an unhelpful direction–since it is set exclusively with long-run contracts)  Still, unemployment is a signal to everyone involved to start cutting prices.

Other Applications

“Signals as realizations” is a helpful idea in other contexts as well.   One point I like to bring up is that, in the real world, avoiding “moral hazard” requires someone to get hurt.

Consider food stamps.   Moral hazard?   If you mean that people are likely to purchase more food because they are on food stamps, then yes, there is moral hazard.   But what is the “realization” of the “signal” that no help purchasing food will be given to the poor?   Someone must starve.

There is a strange tendency to forget the reason we have these programs.   In our models, there are no deviations; everyone acts “on equilibrium path”.   In the real world, we only know what the consequence of deviating because someone has been punished for doing so.

Categories: Uncategorized
  1. nemi
    June 26, 2012 at 10:09 pm

    Concerning the linked article:

    Is a market maker a central planer or a part of the market economy? Does it matter whether the market maker is working for the government or not?

    Is a CEO commanding a firm with market power (i.e., any firm) a central planner or not? After all, he
    is determining the price of their products. Does it matter whether he is working for the government or not?

    Is the politician determining the price of living within a geographic area a central planer or not. Does it matter whether he is working for a private company that owns the land?

    Is it really inconceivable that there would be some benefit of some of these CEO´s and market makers would talk to each other instead of simply communicate through price changes (oh wait, I think that they already do that). Maybe there could be some additional benefit of sitting close to each other (oh wait, they already tend to move their head quarters to regional centers).

    Wouldn’t having the employers of the central planning authority being stationed exactly as in the present and forming price decisions based on the same criteria while receiving the same compensation be one of many possible ways to compute prices?

    Isn´t the requirement that “central planning” would have to find some optimal prices through an algorithm like saying that if medicine can’t give us eternal life – to hell with it.

    (Not that I can see any benefits of more centralization – I simply note that the pretended considerations of alternative regimes usually are strawmen^14)

    PS: Sorry for the rant – it just kept on growing. Love your writing by the way.

    • BSEconomist
      June 27, 2012 at 10:55 am

      Those are really good questions. Hopefully I can answer all of them in a way that makes sense.

      Trying to answer as many as I can at once, I would have to say: No, it (theoretically) doesn’t matter how the information is transmitted through the economy, with the exception that information transmitted through collusion would move the economy away from its efficient frontier, so this kind of collusion would benefit the CEOs while harming everyone else (a market economy is “manipulable”).

      But, for practical reasons, it certainly does matter how information is transmitted. It is difficult or impossible, in practice, for local knowledge to be transmitted to the central planner and that knowledge turns out to be necessary. Also, the “market” is (in effect) a powerful massively parallel computer running what is in effect a reasonably efficient algorithm–my view is that this is the best intuition for understanding the benefits of markets–while it would take a really really powerful computer for a central planner to do the same.

      The point is that central planning fails on informational and practical grounds, but markets will require price changes to transmit information to market participants.

  1. June 22, 2012 at 10:33 am
  2. July 5, 2012 at 3:07 pm
  3. July 5, 2012 at 3:11 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: