Tuesday, April 28, 2009

 

Mankiw Doesn't Realize When He Has Been Beaten...

...and portrays me as an unrealistic ideologue. (In fairness, I implied Mankiw was crazy in my original Mises.org critique to which he was responding, so he handled it with class.) Mankiw first quotes me, saying that future inflation is not necessary to clear the market, because another solution would be for current prices to fall. (Thus, even if it's true that the "equilibrium real interest rate" is very negative, butting up against the nominal zero rate barrier, then you still get market clearing because the large drop in present prices gives you room for a steep price inflation back up to "normal" levels in the future.) To this Mankiw replies:
I think this analysis is correct, under the maintained assumption that prices (including wages) are completely and instantaneously flexible. But if prices are sticky, then the immediate deflation and concurrent increase in expected inflation won't occur painlessly. Instead, it would take a while for the price level to fall, and as we wait, the economy would suffer through a period of depressed economic activity.

According to conventional new Keynesian analysis, sticky prices are the ultimate market imperfection that makes aggregate demand matter. If you deny that prices are sticky and assume they can instantaneously jump downward to new equilibrium levels, many macroeconomic problems become much easier to solve. Indeed, you don't need to solve them at all, as the market would do it.

I wish we lived in the world that Mr Murphy describes, but my reading of the evidence is that we don't.
[Bold in original]

OK hang on a second. I didn't say prices would "instantaneously" drop down to new equilibrium levels (where "equilibrium" isn't even really a good term because Mankiw means something different from what Hayek or Garrison would in this context). Rather, I said that the market isn't so impotent because of the zero-nominal-interest-rate barrier as Mankiw and others seem to think. The market is trying to adjust to the shocking realizations that have set in after the housing collapse--and this includes the huge increase in the demand to hold cash--but Bernanke won't let it. So as usual, we have Mankiw et al. blaming the market for something that their own prior interventions are causing.

You don't need to take my word for it. I will quote someone whom Mankiw presumably respects to make my case:

In this post the economist gives a "Deflation Alert." So presumably deflation is possible. Oh darnit, on second thought that's not really such a decisive point in my favor, because people who worry about deflation might mean long-term deflation. In contrast, I was arguing that prices could fall fairly rapidly in order for the market to clear, without the need for Mankiw's own suggestion of promised massive future inflation.

Ah, but that's why this blog post from last November is relevant. The author reports with worry that, "The CPI drops 1 percent. Even the core CPI is falling." Now this is quite interesting. From September to October 2008, general consumer prices fell one percent--and that's the actual fall, mind you, not an annualized figure (which would have been in excess of 12 percent deflation per year). What's even more extraordinary is that this price drop of 1 percent occurred while the stock of money held by the public (M1) rose by more than 2 percent (and again, that's the monthly figure).

Hang in there folks, we're almost there: Now if prices fell by (more than) 1 percent in one month, at the same time that the quantity of money rose by (more than) 2 percent, we're looking at a 3 percent monthly drop had Bernanke merely held the money stock constant. If the CNBC pundits are right, and what the economy "needs" right now is a negative 5 percent real rate of interest, and if nominal interest rates are zero, then that means it would take the market economy all of 50 days to achieve the necessary price deflation to clear. (Again, this calculation assumes that Mankiw's diagnosis of the problem, and his prescription of negative real interest rates, is correct.) The economy has officially been in recession since December 2007, meaning that the interventionist approach has not fixed things in at least 16 months. Note that 50 days << 16 months.

To sum up, in case my cutesy-ness has lost some readers: Mankiw admits that my proposal of allowing prices to fall would work, if only prices in a market economy could quickly fall. But alas, in the real world, market prices don't fall quickly enough, and so that's why Mankiw thinks Bernanke needs to promise to dump massive amounts of inflation on us in the future.

Yet I am claiming that the only reason prices haven't been falling very rapidly is that Bernanke has been pumping in money like there's no tomorrow. We all know he's increased the monetary base at an absurd pace, but even the M1 money stock rose 17% during 2008. So if the overall CPI was roughly flat for the year, while the money stock rose by 17%, imagine what would have happened had Bernanke merely pumped in enough base just to maintain M1, let alone had Bernanke truly done nothing and let things play out.

I will leave you with this final quote from Mankiw. Again, does the following sound like a guy who thinks that the market can't give us price deflation in a timely enough fashion?
The credibility of the promise is paramount. To get long-term real interest rates down, the Fed needs to convince markets that it will vigorously combat deflation, and that if deflation happens in the short run, the Fed will reverse it by subsequently producing extra inflation. A credible promise of subsequent price reversal after any deflation ensures that long-term expected inflation stays close to the inflation rate implied by the Fed's target price path.

So there you have it, folks. The Fed needs to credibly promise to inflate in the short run, in order to prevent prices from falling. And then the Fed needs to credibly promise to inflate in the long run, because in the imperfect market economy, prices don't fall in the short run.

UPDATE: I realized I had left one gaping hole in my response. Mankiw might say, "Sure, gasoline and stock prices can fall very rapidly, but the one really sticky price is the wage rate. So you might get housing markets and commodity markets clearing, but you'd still have incredibly high unemployment if we followed your crazy advice and let the CPI drop, say, 17% in 2008."

It's true that wages are stickier than many other prices, but that's largely due to other interventions in the market (unions etc.). But fine, let's take all those as given. It still doesn't follow that the market is as fragile as Mankiw believes. It's not that people in particular jobs would have to take a 20% pay cut, but rather that laid off workers would need to settle for much lower pay in order to get hired someplace else. So if a former quant at a hedge fund that blew up used to make $200,000 a year, and now is driving a cab for $50,000 (I don't know what cabbies make), that's a 75% wage cut right there. Were it not for Paulson and Bernanke's bailouts, there would be plenty of huge cuts like that to be thrown into the national averages.



Comments:
Isn't someone losing their job in the same category of overall economy wage deflation. How about losing a 3% match to your 401k. What about receiving lower quality health insurance that requires you to foot another grand or two per year. This one might be pushing it - real wages haven't been sticking to inflation raises over time.

Wages are sticky? I'd respectfully disagree that wages didn't turn on a dime for 99% of us that aren't union protected professors.
 
Of the middle class individuals I work with - the only ones whose wages are sticky are union (for the medium term, depending on their industry) and state subsidized industry managers / professors / professionals / doctors. Everyone else I meet (outside of entreprenuers) has taken pay cuts
 
Is contradicting oneself and refusing to admit it mandatory for Keynesian economists? It seems they do it a little bit too often for it to be a random fluke ....
 
The first Anonymous beat me too it. I obviously haven't done any broad study of the issue, but I would agree with him/her that wages are a lot less downward-sticky than believed. It's just that they won't cut your direct, monetary compensation. They will, however,

-Reduce/eliminate 401k match (as 1st Anon said)
-Increase health insurance charges and copays
-Cut back janitorial services
-Reduce the budget for things not directly related to projects for clients.

And I don't think you can blame the government for this -- it seems like more of a sociological issue, where people overestimate the impact of money-wage cuts compared to the other alternatives. Like with price controls, the result is not to keep the price from changing, but to make people pay a different way, that they probably think is even worse!
 
hpx, you noticed that well. To modern 'liberals', consistency is the hobgoblin of the small mind, and to prove that they have great minds they dispense with such mundane things as logical rigor.

"Liberal" policy is firmly grounded in the expedience of today.
 
Just wanted to provide one more swath of anecdotal evidence against the wage stickiness -- I work in IT for a manufacturer (non-union) and our company took pay cuts across the board. It was an interesting thing to witness. People rationalized taking the cut because other prices (namely gas) were falling and because going out into a job market with high unemployment seemed untenable. Some of our best employees did leave and found better-paying jobs with other companies who needed their skill level more. Now I know this is just one little Midwest factory, but seeing that adjustment and hearing people's thoughts on it throughout the process really made me doubt the claims that wages are sticky downward. At least in this case, things seemed to move toward equilibrium pretty quickly and smoothly.
 
We also eliminated our 33% 401k match temporarily with little reaction. I think everyone understands the implicit contract more than we give them credit for. People know that if you don't like the terms of your employment, you're free to pursue work elsewhere. And given that choice, they will weigh the options and act according to their best interest.
 
I look forward to the "painless" inflation of the coming years rather than that painful, painful deflation.
 
One way around 'sticky' wages is simple: cut staff, close plants. Reducing costs by any other name is still reducing costs.

Just like UAW buying Chrysler is effectively deunionization. "Taking one for the team" just sounds nicer than pay-cut.

You can ignore the market, but the market won't ignore you. :)
 
Bob,

You may win many battles (all of them, to be more precise) against Mankiw, but you will never win the war of ego. That's something he has in spades over you.

I mean, come on! The man is a HARVARD economist. And he was a special economic adviser to the PRESIDENT OF THE UNITED STATES... the freest country in the entire universe!

You've got nothing on this guy, even if you are right.
 
sorry to burst a bauble here but the stickiness of wages is one of those things that has been thoroughly empirically checked over the years.
there are plenty of none uniun reasons for employers to avoid fairing expensive workers-
1)lose of orgenisation memory
2)transaction cost for finding and rehiring replacements after the crisis is over.
3)fear of company secrets leaking.
 
I'm no economist (as I said before, I work in IT), but wouldn't wage price stickiness depend largely on cultural norms? If so, might these change over time, place, and market sector?

The factors Avner referred to are all valid reasons why an employer might be hesitant to cut wages. However, that doesn't mean that prices are necessarily sticky downward any more than other prices; it just implies that the existing labor commands a premium over alternatives -- the risk and transaction costs are reflected in the price. In that case, the fact that wages may not fall as fast should actually facilitate a better reflection of relative prices between labor options. There is no way the empirical work controls for that. Is this the very definition / explanation of wage stickiness or is there a subtle difference here?
 
(I should have posted this comment on the first Mankiw post.)

If the exact suggestion of destroying a percentage of the FRNs is used to create credible inflation, then won't people just significantly curtail their use of physical paper money? There are almost no circumstances where I have to use FRNs, and even businesses that prefer cash transactions may perceive the risk involved and start requesting that customers use credit/debit instead.
 
It doesn't seem that Mankiw specifically says that wages are sticky downwards - is that what he means? Because if wages truly are just sticky (up or down), then he has once again ignored any sense of ethics or humanity.

For the average person to have to deal with inflation of consumer goods prices while simultaneously not having an increase in their wages, this outcome seems at least as painful as Mankiw seems to think deflation will be.
 
Bob -- It's probably always a mistake to call the other guy crazy, unless you are talking about Paul Krugman. ;-)
 
Isn't it interesting that Mankiw's blog does not allow comments and Murphy's blog does?

By the way, I just ordered the Human Action study guide from mises.org and I'm looking forward to spending my vacation in its pages. Yes, I have no life, but at least I will have knowledge!
 
It seems to me that the major reason for sticky prices (or specifically wages) is uncertainty. Avner touched on this, "transaction cost for finding and rehiring replacements after the crisis is over." If employers don't know the future it seems safer to keep current employees. If the crisis is over in a year, it would be cheaper to keep the current staff than loosing staff now and hiring and training a new staff in a year.

If deflation is allowed to occur, it will not be uniform (e.g. cars will likely fall more than cereal). The reason that it takes time for prices(wages) to adjust is that nobody knows what the new prices should be. There is uncertainty and it takes entrepreneurs a while to figure out the best new arrangement of prices and resources.

Bob, you mentioned that when the Fed causes inflation it increases some prices more than others (General Motors vs General Mills). In effect the Fed say, "we're not going to let you take a little time and figure out the best configuration of resources, we are going to force a certain configuration with a preference to interest rate sensitive endeavors."
 
Post a Comment

Subscribe to Post Comments [Atom]





<< Home

This page is powered by Blogger. Isn't yours?

Subscribe to Posts [Atom]