Home > Macro/Monetary Theory > Peter Schiff on Deflation

Peter Schiff on Deflation

A while back I sort of blasted Selgin for his position that deflation is not necessarily bad.  That was a bit unfair of me since he actually has a fairly nuanced and not totally unreasonable point.  If we had a monetary policy regime which caused deflation to not be bad (in other words, if we had one entirely different than the one we have) then it wouldn’t be bad.  But my reaction was sort of a knee-jerk response to a point of view that I run across often in libertarian circles that drives me nuts.

Here is the guy I was really arguing with.  So let me take out my frustration on him.  There is a lot that is wrong with this so I will try to knock off the really obvious but less important points first and work my way up to the big important stuff.  First of all, Schiff discusses an article on Bloomberg which he provides no link to.  IMO this represents a breach of blogging etiquette (there’s not a single link in the piece, it’s almost like he doesn’t expect the reader to question any of the bold empirical claims he is making).  Here is the link.

Postponing Consumption

Here is the argument.

 . . . there is now a nearly universal belief that deflation is an economic poison that works its mischief by convincing consumers to delay purchases. For example, in a scenario of 1% deflation, a consumer who wants a $1,000 refrigerator will postpone her purchase if she expects it will cost only $990 in a year. Presumably she will just make do with her old fridge, or simply refrain from buying perishable items for a year to lock in that $10 savings. If she expects the cost of the refrigerator to decline another 1% in the following year, the purchase will be again put off. If deflation persists indefinitely they argue that she will put off the purchase indefinitely, perhaps living exclusively on dried foods while waiting for refrigerator prices to hit zero.

That is a perfectly good debunking of a position that I don’t think anybody actually holds.  I dug out my intermediate macro text and I think I have identified what he is arguing with.  Here is Blanchard on the subject.

When inflation decreases in response to low output, there are two effects: (1) The real money stock increases, leading the LM curve to shift down, and (2) expected inflation decreases, leading the IS curve to shift to the left.  The result may be a further decrease in output.

We have just looked at what happens at the start of the adjustment process.  It is easy to describe a scenario in which things go from bad to worse over time.  The decrease in output from Y to Y” leads to a further decrease in inflation and, so, to a further decrease in expected inflation.  This leads to a further increase in the real interest rate, which further decreases output, and so on.  In other words, the initial recession can turn into a full-fledged depression, with output continuing to decline rather than returning to the natural level of output.  The stabilizing mechanism we described in earlier chapters simply breaks down.

Now I actually have a lot of issues with this theory.  For instance, it assumes irrational inflation expectations.  Also it assumes an exogenous change in output which causes inflation to fall.  I am actually inspired to explore these issues in a later post (this is not to say that economists are unaware of them).  But nowhere is it assumed that people, expecting prices to fall slightly, will decide to not consume at all.  There is a much more complex argument underlying this.  It involves a feedback loop between prices, inflation expectations and consumer demand which is certainly questionable.  But Schiff is not even scratching the surface of the actually questionable parts.  He is just seeing a result that he thinks is questionable (and is) and he is imagining that it is the result of a ridiculous formulation of the underlying consumer demand function (which it is not) and he is pointing out how ridiculous that demand function would be.  If this were what any economist had in mind, then he would have an excellent point.  However, I don’t think that is the case.

Sticky Wages (and “a change in demand is different from a change in quantity demanded”)

This is an area where I sort of agree with him.  The role of sticky wages tends to be exaggerated with a conspicuous lack of attention put on artificial (government) sources of stickiness like unions and minimum wage and price controls.  But he is going too far by claiming that these are the only source of price stickiness.  Certainly, at the very least, we can acknowledge that long-term contracts exist.  Plus there’s the whole debt thing that I’m always talking about (more on that later).  But what really betrays a failure to comprehend the sticky-wage argument is this.

However, inflation allows employers to do an end run around these obstacles. In an inflationary environment, rising prices compensate for falling sales. The added revenue allows employers to hold nominal wage costs steady, even when the raw amount of goods or services they sell declines.

This misses the point entirely.  He seems to be taking a decrease in the quantity sold for granted (reasoning from a quantity change?) and then treats the inflation as a completely independent phenomenon that just puts extra money in sellers’ pockets in spite of this decrease in order to allow them to keep wages high.

The actual argument behind sticky wages is not that the monetary authority has to cause inflation when the quantity of goods produced falls in order to keep wages from falling.  It is that they have to prevent prices from falling because wages can’t fall and that would cause the quantity of goods produced to fall.  This being due to the fact that employers could not continue to employ the efficient number of workers at the new, higher real wage.

So if we don’t start our reckoning by assuming that the thing we are trying to avoid happens exogenously (for no apparent reason), and instead we imagine that aggregate demand drops, perhaps due to unexpectedly tight monetary policy, we can make better sense of things.  In this case, demand for all goods would decrease.  This means that sellers would have to either reduce their prices or sell less or do both.  If all prices (including wages) were not sticky at all, they would simply lower their prices and wages until they were at the same quantity of sales and employment but at lower prices and wages.  But if they can’t lower wages, it will not be optimal for them to keep producing the same quantity at a lower price and the same wage.  This will cause them to cut back production (by laying off some workers) and lower prices until they are maximizing profits given the new lower demand and inefficiently high, sticky wages.

This, of course, can be avoided if the monetary authority avoids tightening, or in the case of aggregate demand (which is not the same as quantity) dropping for some other reason, by adopting a more accommodative policy to push it back up.  Is this a compelling argument for having a monetary authority manipulate the price level in the first place?  Maybe not, you decide.  But it is not a nonsensical argument and arguing with a distorted, straw-man version of it doesn’t get us anywhere.

And while we’re on the topic of Peter Schiff not really understanding what “demand” means, let’s deal with this:

Economists extrapolate this to conclude that deflation will destroy aggregate demand and force the economy into recession. Despite the absurdity of this argument (people actually tend to buy more when prices fall), . . .

There are two problems here.  One is that deflation doesn’t destroy aggregate demand, falling aggregate demand causes deflation.  I can forgive him for this because the textbook argument I cited above does actually give the impression that the causation goes both ways.  I think this is a problem with that exposition.  So I would be inclined to agree with his criticism if he didn’t blow it with the parenthetical statement.

This is econ 101 stuff.  The quantity demanded by a given consumer (or all consumers) is larger when price is lower and other things are constant.  This is different from a fall in demand!  Market prices are determined by supply and demand.  When demand falls, both price and quantity fall.  Aggregate supply and demand mean something a little different but the basic intuition is the same.  If your argument against the deflationary spiral is that “people buy more when prices fall” you are several layers of reasoning short of understanding the thing you are arguing with.

Debtors and Creditors

This is where it starts to get near and dear to my heart.  Schiff points out that inflation helps debtors but hurts creditors, giving the impression that it is simply a sterile transfer from one group to another with the implication that it goes on because the beneficiaries are more politically connected than those on the other side.

While it is true that inflation (by which I mean more than expected inflation) helps debtors and hurts creditors, this does not make the net effect neutral.  This is because people in aggregate are net debtors vis-à-vis the banking system.  So it is possible for us to all go broke together.  This is where my crackpot theories diverge from both the Schiff/Rothbard crackpot theories and from the “mainstream” (crackpot) theories.  So going into that in detail here would take us far afield but if you’re interested click here, here and here.

The Zero Inflation Boundary

Schiff gives the impression that, until this Bloomberg article, economists were only worried about negative inflation and that they only advocated low, positive levels of inflation because this provides a buffer against accidentally falling into deflation but that so long as this is avoided, there is no problem.  I don’t think this is an accurate characterization of the stance of most mainstream economists.

There are two separate issues here.  One is the question of the appropriate long-run policy regime.  This is where Schiff probably got this idea.  Most macro models have some version of money neutrality in the long run.  Often, this takes the form of superneutrality which means that the monetary authority can grow the money supply (and therefore the price level) at any rate they want without affecting the real economy in the long run. 

This leaves policy-makers (assuming they control the money supply) with the issue of selecting a long-run growth path for money and prices.  A natural argument, if you believe in sticky prices/wages, is then to say that a positive but moderate inflation target might be best because it allows for some cyclical fluctuation around the target without triggering the sticky-wage problem.  This, admittedly, is a lot like Schiff’s characterization of the mainstream position.  But this is not really the issue that most economists are concerned about when they warn about inflation being too low.

It is one thing to argue that the monetary authority could follow a zero-inflation long-run policy path and it would be no better or worse than a 2% inflation or 10% or -2% inflation path.  It is something else entirely when the monetary authority sets a 2% inflation target and then produces 1/2%.  It’s not that there is something magic about crossing the zero-lower-bound on inflation.  The problem is when inflation runs below expectations.

When people make long-term financial decisions, they do so with some expectation about future prices.  For instance, if you invest in producing a good, you have to predict the price of that good in the future.  If the price turns out to be lower than you thought, your calculations will be wrong.  There is not some eternal magic number that the price must be and the central bank has to make it hit that price or else cause a lot of problems.  It’s just that they have to not screw up peoples’ calculations by causing them to expect a certain price level and then causing a different level to occur.

This is particularly problematic when it goes on for an extended period of time without any attempt at filling in the gap.  The bigger that gap gets the bigger the difference between the actual price level and the level people were expecting when they initiated long-term financial positions in the past.  Plus if they go on like this for a while without changing their target, it starts to become unclear what to expect from them in the future.

This is mostly important for decisions involving nominal debt because, as I like to go on and on about, debt contracts are long-lived and nominally denominated so your obligation in nominal terms does not fluctuate with the nominal value of other things like your labor or your house or the output of your business.  And remember, aggregately, we are net debtors so when the price level grows more slowly than we were expecting, this causes a systematic weakening of the financial position of the economy as a whole.

This is my special way of characterizing the issue at the heart of the inflation question.  There are different ways of characterizing these issues and different models for trying to understand them but almost all of them rely in an essential way on the role of inflation expectations.  Peter Schiff is not even scratching the surface of these issues.


P.S. This is the most unimportant point and I originally put it at the beginning but it is probably also the most inflammatory and I figured some people might not make it past it to the more important stuff so I cut it out and stuck it here at the end.


Definition of “Inflation.”

This is a common sticking point for Rothbardians (which is what I am now using to refer to a particular popular sect often known as “Austrians” on the advice of commenter John S).  Economists almost universally use this word to refer to the change in the aggregate price level.  Rothbardians commonly assert that this is the “wrong” definition in the sense that it was not the original definition.  This is a completely pointless debate.

Usually this argument is brought up for one of two reasons.  One is to generally discredit mainstream economists and imply that they don’t know what they are talking about because they don’t even use the “right” definitions of words.  The other is to weasel out of hyperinflation predictions by saying “well, okay, prices haven’t really risen that much but that’s not the actual definition of inflation…”

I am no expert on the evolution of this term and I don’t care about it enough to research it but just for fun here is my educated guess of how things went down.  Originally inflation meant what the Rothbardians say it means because at that time, that was the most useful definition for thinking about the economy.  Eventually the state of economics evolved in such a way that someone had to give a name to the rate of change in the aggregate price level and they called it the “inflation rate” because they figured prices tended to be positively correlated (perhaps proportional) to the size of the money supply, other things being equal.  As the science continued to evolve, economists found themselves being very interested in the rate of change of prices and not so much in the money supply, except to the extent that it affected prices.

To see why this is reasonable consider the following thought experiment.  You need to make some investment/consumption decisions.  Maybe this is buying a house, or investing in the stock market or taking a new job or whatever, it doesn’t matter what it is.  An angel comes down from heaven and offers to do you one of two favors.  He will either tell you what the size of the money base will be in ten years (and not the price level) or he will tell you what the price level (somehow defined) will be and not the size of the money base.  Which would you prefer?

Before 2008, you probably could have at least argued that the choice was trivial because you could extrapolate pretty easily from one to the other but that notion should be dead now.  Of course, Rothbardians realize that what really matters is prices, their arguments are always actually about the price level.  They never say “we’re going to have hyperinflation of the money supply and you should be really worried about that even though prices will never rise more than 2% per year.”  When they make their inflation arguments, they are always talking about the price level.  It is only when those predictions don’t come true that they step into their time machine and act like the original argument that took place two years ago was really a hundred years ago and “inflation” didn’t really mean what you thought it meant.  And by the way, they weren’t really making a prediction because Austrians don’t make predictions so if you thought they were, you must have been confused.

When economists say “inflation” they are talking about changes in the price level.  All economists realize that this is what other economists mean.  If you want to use it in a different way, and be clear that this is what you are doing, because it allows you to say something interesting, I have no problem with that but if you are just ridiculing someone for using a word to mean what most people commonly accept it to mean, you are wasting everyone’s time.

  1. John S
    April 12, 2014 at 4:07 am

    It really is sad that Schiff continues to get so much mainstream media time on CNBC and the like. He’s done far more damage to the cause of free market economics than any “Keynesian.”

    Just curious: perhaps you’ve posted on this before, but do you see any risks in a permanently expanded Fed balance sheet?

    • Free Radical
      April 12, 2014 at 7:16 am

      Yeah that whole “people actually buy more when the price is lower” thing really made my head hurt. There’s really a lot of work to do to rescue libertarian economics I’m afraid.

      The short answer to your question is yes. Lately I have been putting more effort into criticizing critics of the Fed than to criticizing the Fed itself but I actually think central banking is a really bad idea and has the potential to be amazingly destructive. But I think the danger is not well understood by Fed detractors. They are mostly just determined to fight anything they consider “monetary easing” or “money printing.” They don’t really understand how the mechanism works. The real problem, in my opinion (and this id debatable), is that monetary policy HAS to get progressively “looser” (by conventional, non-MM standards) in order to keep the economy from collapsing. This means that the CB and the central gov. have to take over more and more of the economy.

      But when people on the right just say “stop printing money” (by which they mean tighten monetary policy) they are calling for something that will wreck the economy (and their reputations along with it). That won’t work. I always has the opposite effect which is to cause a recession which then becomes the justification for gov and CB getting more interventionist. Will try to do a post on this. I think I have some old stuff about it but my views have evolved enough since I started this blog that, depending on how old it is, it might not be entirely accurate anyway.

  2. Gus
    September 8, 2015 at 1:08 pm

    However, interests on long term loans are generally computed
    based on the outstanding balance. Is there any reason why any company should offer a car
    loan to anybody with a poor credit record. When someone agrees
    to co-sign a loan contract he is legally guaranteeing that
    the signer will pay.

  1. No trackbacks yet.

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s

%d bloggers like this: