In praise of (a little) inflation

At a conference this weekend, talking (once again) about the gold standard, I was struck by the fact that the things economics writers take for granted often sound horrifying to ordinary people.

In this case, the trouble came when I said that it's a good thing that the Federal Reserve errs on the side of having a little bit of inflation, and that in fact inflation in small amounts is probably good for the economy. The reaction of the assorted nice, normal people I said this to was about what you'd expect: they looked as if I'd suggested recreationally vivisecting their cat.

And yet, this isn't really all that controversial. A little bit of inflation lubricates the problem of sticky wages and prices: which is to say, that prices and wages are quicker to adjust upward than downwards.

To liberals, this generally sounds great: once you get a wage gain, it's yours to keep! The problem is, in an economic downturn, or a sectoral slump, the cost of your keeping that wage gain is that oftentimes, someone else gets the lovely parting gift of a layoff. The principle is less controversially illustrated using office leases, which also tend to be sticky. If your industry is in a slump, you might need to renegotiate the lease on your office or factory downwards. That keeps the space occupied, benefitting the owner, and allows you to operate profitably. The problem is, landlords hate negotiating leases downwards; perhaps they can't (because they have a mortgage), or they can't bear the thought of accepting less than they got before, or they may not want to cause trouble with other tenants of similar properties. There's also, as with labor contracts, fear of bad faith; if landlords show a tendency to believe hard-luck stories, everyone will be along with one. Also, even if you are in trouble, they'd rather you wrung a concession from someone else, such as workers or suppliers.

The problem is, if you genuinely are in trouble, inability to renegotiate could result in your going out of business, making everyone worse off. A month or two's vacancy generally costs more than the downward renegotiation would have.

Inflation eases these problems; landlords, suppliers, and workers take real wage cuts without the psychological barriers of lowering their prices. So decreases in demand produce much less severe contractions in output; this is one of the explanations for the relative mildness of postwar, and particularly post-1980, recessions.

Nor is anyone made particularly worse off by this, unless they've a habit of stashing large wads of cash under the mattress; interest rates and cost-of-living adjustments largely mitigate the effect on the most vulnerable groups, pensioners and those on disability. Consistent, but low, inflation is one of the great economic success stories of the 20th century. But somehow, every time you try to explain this, you end up with an audience holding fast to their cats.

Comments

I'm in complete agreement with everything said here, except for this one bit of completely unjustified partisanship: "To liberals, this generally sounds great...." All the liberals I know who have had some basic econ understand the points you're making; and the kind of rabid
ultra-anti-inflationism that keeps some people occasionally lobbying for the gold standard is far more prevalent on the right than on the left.

Posted by philosopher | September 17, 2007 6:35 PM

In a different world, without an expectation of inflation, people might not be so adverse to taking a slight cut. People should be more aware of inflation's static filled bite and then they wouldn't mind the clarity of a more stable money supply.

Prices should be easy to compare, across product, and across time (at least longer than 3 months or a year... without needing to adjust for changes in the velocity of the printing press.)

Think about the way one could compare goods historically and see the value of productivity more clearly, or compare the relative worth of goods over time... whatever.

-MichaelATL

Posted by MichaelATL | September 17, 2007 7:33 PM

Prices are sticky? I guess Apple didn't get the memo.

Posted by RBC | September 17, 2007 8:41 PM

Isn't avoiding deflation one of the best arguments in favor of modest inflation? Deflation and very high inflation are devastating, but it's really hard to nail exactly 0% inflation. If you can manage 2% inflation without risking a slip into deflation, you're close to ideal and have a safety margin.

Posted by Lorenzo | September 17, 2007 9:11 PM

"Deflation and very high inflation are devastating..."

The credit addicted sound so much like those discussing the mythical "perfect dose" at NarcAnon meetings.

Posted by Mark E Hoffer | September 17, 2007 10:27 PM

prices and wages are quicker to adjust upward than downwards.

You confidently claim that prices are "quicker to adjust upward". Hmm... what about computers? They have adjusted in price downwards for all of your life, apparently without trouble. I don't see any conceptual problem with rational businessment gradually lowering prices in a deflationary environment. After all, right now they face the opposite problem, namely, how to gradually raise all of their prices. Somehow they manage it.

Indeed, to the extent that anyone is homo economus, it is and had better be businessmen. If they can't rationally respond to market forces, then I think perhaps you majored in the wrong subject.

Wages? Well, that we don't know about. Certainly people seem to resist pay cuts... perhaps the average man is too stupid and ill informed to know that he's doing better this year than last even getting 1% less money. But perhaps the average man is smarter than you think.

We have all lived our whole lives in a world where every currency is a monopoly fiat currency, and every one of the monopolies has the incentive to inflate. We have lived our entire lives with inflation ranging from moderate to high. In this environment, it is rational to believe that you need to get raises (or at least, not get cuts in pay) if you want to maintain your standard of living. People believe this because it is true in this environment.

But... if everyone had grown up in a world of steady 2% deflation, for their whole lives, do you really think they would resist pay cuts as much? I don't think they would.

Posted by Leonard | September 17, 2007 10:34 PM

philosopher is right. The main problem may be that Megan is attending the wrong conferences.

Also, hostility to inflation tout court on the right may be partly tied to the primacy of "Carter-era stagflation" as the linchpin of the GOP's claim that liberals don't know how to manage the economy. That, and a somewhat reductionist and manichaean cast of mind.

Posted by brooksfoe | September 18, 2007 6:37 AM

Leonard, they empirically did; prices did not fall smoothly during 19h century recessions.

Posted by Megan McArdle | September 18, 2007 7:57 AM

Leonard is right that computer prices have been falling. But on the other hand productivity in the
computer industry is so large that the firms can sustain employment and profits in the face of sharply falling prices.

But there are very few industries with the type of productivity growth that makes this possible.

Falling computer prices are the exception, not the rule.

We have had two periods of deflations in US economic history -- the long depression of the 1880s and the great depression of the 1930s. It appears that deflation and depressions are not just conincidents.

Posted by spencer | September 18, 2007 8:37 AM

brooksfoe and philosopher: Let's compare apples to apples here. Yes, econ-literate liberals do reject rabid anti-inflationism. However, econ-literate liberals also believe:

-All trade barriers should be eliminated.
-Taxes carry deadweight losses and hurt economic growth by more than the amount they divert to the government.
-It's possible, in theory, for a tax hike to reduce tax revenues or a tax cut to increase them.
-A Pigouvian tax is superior to all efficiency standards, including: CAFE, incandescent bans, toilet tank/shower head regulation, solar/hybrid vehicle tax credits, etc.

Yes, those are uncontroversial beliefs among mainstream moderate Democrat economists.

Don't distance yourself from those dumb liberals unless you want to distance yourself all the way.

For my part, I seem to notice that a constant complaint of liberals is that wages don't keep up with inflation.

Posted by Person | September 18, 2007 9:34 AM

You are absolutely right, Person. The CPI is calculated using a skewed system designed to keep real inflation 'hidden', so wages based on, or tied to, CPI some way, always comes up with a smaller number than what is actual.

No matter the pay, I'm sure Megan deserves at least double what Matt's getting.

Posted by falkoyn | September 18, 2007 11:14 AM

falkoyn: I wasn't endorsing that position. FWIW, my position on inflation is:

If you really think it's understated, tell me the basket of commodities and futures I can buy that captures the true, much-higher inflation rate. If you can't, don't complain.

Posted by Person | September 18, 2007 11:32 AM

-Taxes carry deadweight losses and hurt economic growth by more than the amount they divert to the government.

No, liberal economists believe that it depends what you spend the money on. No serious economist believes that a country which cut taxes dramatically and financed those tax cuts by eliminating public schools would see growth rise by more than the amount of the tax cut. Ditto financing tax cuts by eliminating highways and bridges, or shutting down airports, or eliminating the police department. The economy did not benefit in Somalia when the government, by virtue of its disappearance, stopped collecting taxes.

-It's possible, in theory, for a tax hike to reduce tax revenues or a tax cut to increase them.

Liberal economists would probably not think this the case under the existing tax regimes in any country on planet Earth.

-A Pigouvian tax is superior to all efficiency standards

This isn't always the case in every respect. I think most economists agree with regard to gasoline vs. CAFE, but water, for example, is more complicated, as is electricity use. Many of the things you should do to save electricity are complicated, not intuitive, and require redesigning your house to some extent. These kinds of complex shifts are often better addressed by setting standards for new construction, because builders and architects are likely to understand the issues better than consumers.

Posted by brooksfoe | September 18, 2007 12:30 PM

No, liberal economists believe that it depends what you spend the money on.

That would be the *benefit* of the spending, not of the tax. Yes, spending can have benefits. But the point is, the benefit of the spending must exceed the economic cost of the tax PLUS economic cost of the additional DWL. It's not enough simply to show that it exceeds the cost of the tax itself. Informed liberals get this, modal liberals don't.

Liberal economists would probably not think this the case under the existing tax regimes in any country on planet Earth.

Zimbabwe?

Anyway, the point is, it's a theoretical possibility. Modal liberals have a hard time differentiating between whether they are arguing against the theoretical possibility, or whether it holds for a specific circumstance. See: every Laffer Curve debate on the internet, including this blog.

This isn't always the case in every respect. I think most economists agree with regard to gasoline vs. CAFE, but water, for example, is more complicated, as is electricity use. Many of the things you should do to save electricity are complicated, not intuitive, and require redesigning your house to some extent.

That doesn't change the line of reasoning. If the consumer doesn't feel making the change is worth it, *even while paying taxes equal to the negative exteranlity*, a mainstream liberal economist would hold that no further incentive is justified.

***

So anyway, sure, let's purge all those crank liberals and implement the liberal economists' ideas!

Posted by Person | September 18, 2007 12:41 PM

prices did not fall smoothly during 19h century recessions.

Megan, this quote suggests that you believe that sharp monetary contractions are a necessary aspect of a gold standard, inseparable from it. This is not the case. Gold is highly resistant to both contraction and expansion, as I know you will recall from our discussion of Anglo-American. The supply expands very smoothly and predictably, due to mining. A tiny fraction is used in industry.

The paper penumbra of gold can, of course, expand and contract wildly. It depends on the rules around the paper. Sensible goldbugs call for strict limits on the paper. Nobody should be allowed to sell two claims to the same object. This seems fundamental, yet its converse is the basis of fractional reserve.

The boom/bust cycles of the 19th century were not due to a stable monetary supply; rather, to an unstable one. Because of fractional reserve banking, the money supply moved in dramatic and unanticipated ways. A bank could open up anywhere, take some deposits, and start cranking out notes. This created a highly unstable situation: so long as most people were doing OK, no bank run would develop, and things would proceed nicely. The bank would make a tidy profit, being as it was creating money for the price of printing a paper certificate. As soon as times turned a little bad (for whatever reason; perhaps a bad harvest), suddenly a bank run would precipitate. The money supply, at least locally, would suddenly dry up as everyone stopped accepting paper. Synchronize up several such regional crises and you get a recession; a few more, a depression. In all of these cases, a panic would feature prominently: bank runs, and more generally, an attempt to sell paper and back into gold and other hard assets.

In these situations, prices did not move smoothly. No, of course not. But the problem is not the correction; that was the market operating properly to correct a real problem. The problem was the underlying fraud, smoothly spun out, of a large quantity of paper money which were supposedly claims to hard assets, yet in reality not based on anything. So: gradual inflation, say, 5% per year for 10 years, as fractional reserve banks print and circulate money, over 10 years doubling the money supply. Price levels adjust up, maybe 40%. Suddenly a series of banks collapse: 1/2 of all paper money in circulation becomes valueless almost overnight. What does micro say should happen to price levels? Don't you see that the problem here is not the gold basis of the money, but rather the fact that there are multiple legally-valid claims to the same real asset?

Posted by Leonard | September 18, 2007 2:17 PM

The economy did not benefit in Somalia when the government, by virtue of its disappearance, stopped collecting taxes.

Actually, it did.

http://www.peterleeson.com/Better_Off_Stateless.pdf

Could anarchy be good for Somalia’s development? If state predation goes unchecked government may not only fail to add to social welfare, but can actually reduce welfare below its level under statelessness. Such was the case with Somalia’s government, which did more harm to its citizens than good. The government’s collapse and subsequent emergence of statelessness opened the opportunity for Somali progress. This paper uses an “event study” to investigate the impact of anarchy on Somali development. The data suggest that while the state of this development remains low, on nearly all of 18 key indicators that allow pre- and post-stateless welfare comparisons, Somalis are better off under anarchy than they were under government.

Posted by Leonard | September 18, 2007 2:28 PM

the benefit of the spending must exceed the economic cost of the tax PLUS economic cost of the additional DWL.

Person, you wrote that "taxes hurt economic growth by more than the amount that they divert to the government". That's a tendentious way of phrasing things. Obviously one doesn't tax for the pure joy of taxing; one taxes to have money to spend on things, and those things are often investments which make the economy grow faster than it would have if the government had neither gathered that revenue nor spent it.

Modal liberals have a hard time differentiating between whether they are arguing against the theoretical possibility, or whether it holds for a specific circumstance.

I've never seen anyone argue against the "theoretical possibility", because it's a waste of time. Every argument against the Laffer Curve I've ever seen is based on the rather thunderously convincing point that it does not refer to any real existing situation in a human economy observed on Planet Earth; every single actual economy seems to be on the side of the curve where lower taxes means less revenue.

Posted by brooksfoe | September 18, 2007 3:51 PM

Obviously one doesn't tax for the pure joy of taxing; one taxes to have money to spend on things, and those things are often investments which make the economy grow faster than it would have if the government had neither gathered that revenue nor spent it.

Yes, and for the third time, the existence of DWLs shows that that benefit must exceed the economic loss due to the diversion of funds AND the additional DWLs. The crank modal liberal you're distancing yourself from -- the one who endorses rabid anti-inflationism -- refuses to accept this basic principle.

I've never seen anyone argue against the "theoretical possibility", because it's a waste of time.

Then you missed the last four threads about it.

Admitting the theoretical possibility -- which you have to do -- forces you to accept that there may be some downsides to taxes, which is why liberals don't want to admit it. You saw in previous threads how they have to set up convoluted fallback positions so they can "beat" the argument, instead of just saying, "Yeah, higher taxes could reduce revenues, but they won't hurt the economy badly enough in this specific circumstance."

Do you still think Zimbabwe's taxes aren't high enough to pass the Laffer point?

Posted by Person | September 18, 2007 4:43 PM

There's a less psychological reason to prefer a little inflation, or so I was taught long ago. That is, if money is inflating, you have an incentive to spend it now because it will be worth less tomorrow. (This gets way out of hand in hyperinflations.) Another way to say it, is that inflation favors borrowers.

On the other hand, if money is deflating, you are better off hanging on to it and waiting for prices to lower. It is good to be a lender. The economy slows down.

Zero inflation is neutral, but an annual rate of a few percent seems to be the happy point. I suppose there's a big theory about this, of which I am blissfully ignorant.

Posted by Martin E. | September 18, 2007 6:25 PM

The reaction of the assorted nice, normal people I said this to was about what you'd expect: they looked as if I'd suggested recreationally vivisecting their cat

Is it really necessarily to use lurid cat metaphors to make your point? Please be more considerate of cats going forward.

Posted by Kent | September 18, 2007 9:58 PM

if money is inflating, you have an incentive to spend it now because it will be worth less tomorrow. (This gets way out of hand in hyperinflations.)

Yes. But note that this consequence of inflation is a deadweight loss. It encourages people to hold less cash than they otherwise would, that is, to spend more of their valuable time working to maintain a minimal cash balance. This tendancy is obvious in hyperinflation, but present no matter what level of inflation there is. Everyone has the same incentive: nobody wants to hold cash. But everyone cannot spend all their cash: it doesn't vanish with use. Thus, the effect is just a one-time price level change. (In hyperinflation the problem is they keep running the presses.) People more willing to hold cash (or less able to do without it), end up holding a bit more; most people end up holding a bit less.

Another way to say it, is that inflation favors borrowers.

No, this is a different effect of inflation. Unanticipated inflation favors borrowers, redistributing money to them from creditors. Since there are many more borrowers than creditors, this is one reason why inflation is desired in a democracy. But note that only unanticipated inflation can reliably get this effect: a savvy businessman, knowing that inflation is likely, simply raises his interest rate.

Similarly, unexpected deflation would redistribute from borrowers to creditors. However, once again expected deflation would not have this effect. If you expect your money to be worth 2% next year, you should be willing to lend it out at very low nominal interest rates (perhaps 1%), and you'll find borrowers if they find the real rate (3%) acceptable. You only gain a windfall if it turns out that deflation was really going to be 10% or whatever.

The main effect of monetary inflation (expected or not) is redistributory, but not among borrowers and lenders. Rather, everyone with cash is "taxed" for the benefit of the few with the privilege to create the new money. The "tax" falls disproportionately on people to the degree they hold cash.

On the other hand, if money is deflating, you are better off hanging on to it and waiting for prices to lower.

Not unless the effective real interest rate for holding money is above what you can get via a very safe investment. For example, right now you can get 5% on a CD of a few months duration. Inflation is low, about 2.4% at least as measured by the government. Thus, if you put your cash in a CD -- right now -- you can get 2.6% as the real interest rate. So, right now we are subject to the same problem you are trying to say only relates to deflation.

Now, I'll agree with you that sharp deflation -- of above 2%, perhaps -- will have some of the effect of encouraging people to wait to spend, above and beyond what we have right now. However I hardly worry about this as a real problem. The historical rate of deflation on the gold standard was very slight, perhaps 1% a year, and that was with a dramatically increasing population.

[if money is deflating] It is good to be a lender. The economy slows down.

The economy begins to slow down only if the deflation rate is so drastic that some potential borrowers have projects that cannot repay the slightly increased real interest rate. But this is a hazy margin, because different projects will have different expected values. For example, with 10% deflation, you can still lend at 1%, if you can find people willing to borrow at 11% real rates, which is hard to do, but still possible. (People will borrow at 18-20% real rates! It's hard to believe, but true.) So, yes, in this sort of situation, the economy would cease to initiate very many new capital projects, and even defer a lot of day-to-day spending.

What does that tell us? That it is not a sustainable pattern. Rather, once the price level adjusts for a while, a new equilibrium will be reached where people again begin to spend down their cash balance. The only cases I can think of historically where we have seen such deflation are depressions which follow inflationary money policies. A 10% deflation is just not going to happen year in and out.

Posted by Leonard | September 18, 2007 10:26 PM

Leonard -
The original question was about the "optimal" inflation level, and the argument was that it lets relative prices adjust over time, overcoming ratchet effects. (Such was my naive reading.)

My question is whether a small inflation acts as an economic stimulus in a more direct way - such as increasing velocity of money.

If a constant rate of inflation gets priced into transactions, then inflation doesn't matter? (perhaps the Cheney position)

Posted by Martin E. | September 19, 2007 1:11 AM

'The Two Income Trap' hypothesis which is sort of 'a 2 can't provide as well as one used to' discussion really wouldn't be a subtle one except for the fact of inflation. Zywicki, for instance, comes to the conclusion that the reason is a bigger percentage tax burden. If we hadn't over the last generation substituted, is it Italian money, for what we used to have, we wouldn't be so confused. I don't know that 'confused' is where we should be about our assets and spending etc. which is why people react as if it were dissecting the cat. I wonder if the target band for inflation was the 2% between -1% and +1% rather than 1-3% if we would have the same need for liquidity that the Fed saw today. There might be more respect for the idea that real assets could decline in value.

Posted by michael | September 19, 2007 1:49 AM

I'm basically in agreement with brooksfoe's responses to person: no liberal has any trouble acknowledging the _theoretical_ points there, if they are treated as general-but-far-from-exceptionless; but we think there's no reason to think that they have much practical application to the actual policy decisions that are being faced in this country.

And one has to consider the requisite second half of the argument, which are the various ways in which modal conservatives get this sort of stuff wrong. E.g., by completely neglecting the possible net economic benefits of a tax, and focusing only on distortions and DWL. Also it matters how you measure DWL; if there is returning marginal utility to income (and, of course, there is), then DWL measured in dollars might not be an overall DWL of utility. That's part of what welfare economics is about, I believe.

And this is a case of just plain missing the point: "For my part, I seem to notice that a constant complaint of liberals is that wages don't keep up with inflation." It is important for competition that _an individual's_ wages, or even _a sector's_ wages may _sometimes_ fail to keep up with inflation. But if _everyone's_ wages are _persistently_ failing to keep up with inflation, then something's gone awfully screwy.

Posted by philosopher | September 19, 2007 10:12 AM

I'm basically in agreement with brooksfoe's responses to person

So what's your stand on whether Zimbabwe is taxing above the Laffer point?

no liberal has any trouble acknowledging the _theoretical_ points there

No, as I explained before, many liberals do have a hard time grasping even the theoretical concept of DWLs and the Laffer curve, ("taxes can hurt productivity??? NO WAY!")

if they are treated as general-but-far-from-exceptionless;

Exceptions?

"Cutting taxes can increase revenue."

"Taxes carry DWLs."

The excecptions to those are pretty ill-defined. How do you find an exception to "Tax cuts *can* increase revenue." ?

but we think there's no reason to think that they have much practical application to the actual policy decisions that are being faced in this country.

??? Are you high? Incorporating tax DWLs as a cost that a program must overcome doesn't have much practical application? The efficiency superiority of Pigouvian taxes compared to physical efficiency standards "doesn't have much practical application"? Trade with foreigners as equivalent to technological improvement "doesn't have much practical application?

And one has to consider the requisite second half of the argument, which are the various ways in which modal conservatives get this sort of stuff wrong ...

No. You're creating self-serving caricatures on both sides. To make this brief.

-Yes, there are econ-literate liberals. No, they are not mainstream.

-Yes, econ-literate liberals reject rabid-anti-inflationism.

-Yes, econ-illiterate liberals make up the majority of liberals, or we would not have such ridiculous policies for achieving liberal goals.

-Yes, the same holds for conservatives.

Hey, I wish liberals knew more economics. But sadly, this isn't the case, so quit trying to marginalize the importance of these unwashed masses.

It is important for competition that _an individual's_ wages, or even _a sector's_ wages may _sometimes_ fail to keep up with inflation. But if _everyone's_ wages are _persistently_ failing to keep up with inflation, then something's gone awfully screwy.

No, I think you missed my point there. If a consistent liberal grievance is "wages don't keep up with inflation", then those very same liberals probably *are* rabidly anti-inflationist because such a policy, under that assumption, would keep wages higher relative to prevailing prices.

Posted by Person | September 19, 2007 11:29 AM

The original question was about the "optimal" inflation level, and the argument was that it lets relative prices adjust over time, overcoming ratchet effects.

Yes, and I don't buy the argument that small amounts of inflation or deflation are harmful.

First, consider goods. Prices for various things are always going to be moving around anyway; so long as the effect of inflation/deflation is small relative to non-monetary effects on price levels, then it seems to me that the inflationary effects will be trivial.

Consider as an example gasoline. Its prices move all the time, weekly or more often. If its price "ought to" trend up 5% this year due to inflation, how is that a problem? It goes up and down more than that several times a year, anyway. Almost all prices for goods are like that. People adjust them at least yearly anyway.

Wages, and services whose costs are largely wage (that is, not based on costs of material inputs), are the things that Keynesians are worried about. But even here, I am just not worried. For every irrational employer, afraid to demand a wage cut that he knew would be necessary, there's mass of irrational employees afraid to ask for a wage increase which they merit. Irrational "stickiness" applies all over the place, if it applies anywhere. Mild inflation helps the market for wages clear if it contains numerous unproductive but irrationally demanding people. Mild deflation helps the market for wages clear if it contains numerous productive but irrationally meek people. I think both things are probably true. Seems like a wash to me, if it is anything. It also seems like a really weak reason to make a decision either way about fiat currency vs anything else.

My gut feeling, without really thinking hard on it, is that zero is the ideal 'inflation' level; however, saying that smuggles in several assumptions at least one of which is surely untrue. "Inflation" is not some objective thing, unless you want to talk only about the quantity of money (and even there the money-ness of various instruments varies, making it very hard to measure how much money there is). What most people think of when they say "inflation" is something like "average price level", i.e., something like the CPI. But this is nowhere near objective; it's a weighted average of various price level changes, computed using lots of handwaving.

So, when I say "zero" inflation is best, I am thinking basically of a (very hypothetical) money which inflates/deflates only to track population increase/decrease, and which creates new money in some egalitarian way which spreads the benefit to everyone. This is not possible for any real money. However, using gold for money would approximate this far better than a state-monopoly fractional reserve system.

My question is whether a small inflation acts as an economic stimulus in a more direct way - such as increasing velocity of money.

Ah, "economic stimulus". Well, yes. Inflation acts as a stimulus, by falsifying the signalling that a free-market economy does via price levels (in this case, the price being interest rates, which mediate currect vs future consumption). This is a bad thing, not a good thing. This is another reason to be against fiat money, but it takes us rather farther afield. I refer you to the wiki page on the business cycle, particularly the Austrian ideas on it, which I believe are correct. http://en.wikipedia.org/wiki/Business_cycle#Austrian_School

If a constant rate of inflation gets priced into transactions, then inflation doesn't matter?

No, it still matters; only some of the effects of predictable inflation can be worked around by the market. (The effects are somewhat less bad than the effects of unexpected inflation.)

When inflation is predictable, the gains always go to those who get to create money, and those who are getting the new money earliest (while prices adjust upwards). The losers are people who get the new money later, after prices have adjusted, and people who are fooled by the falsified interest rates, chiefly entreprenuers whose businesses fail. More widely, everyone is affected negatively when the economy does not grow as fast as it could grow, which is the consequence of the continual downward falsification of the interest rate.

With unexpectedly high inflation, these same groups benefit or lose, but in addition there is also a redistributionary effect. The possibility of unexpected inflation induces various responses in the market which are not otherwise necessary, most particularly an increase in real interest rates that must be charged.

Posted by Leonard | September 19, 2007 11:49 AM

philosopher - Re: "But if _everyone's_ wages are _persistently_ failing to keep up with inflation, then something's gone awfully screwy."

Not necessarily. Wages are not total compensation. If non-wage compensation is increasing total compensation can increase even as wages decrease in real terms.

Also for the short run total compensation not keeping up with inflation isn't necessarily all that screwy.

Posted by Tim Fowler | September 19, 2007 12:10 PM

Leonard - Re: "So, when I say "zero" inflation is best, I am thinking basically of a (very hypothetical) money which inflates/deflates only to track population increase/decrease"

Just population increase? Or increase in gross production. If the money supply only goes up with population increase, and gross production goes up faster than the population (which has typically been the case at least in modern times, say since the industrial revolution), than a money supply that only goes up with the population would lead to a general decrease in prices, which many would consider deflation.

Posted by Tim Fowler | September 19, 2007 12:17 PM

"For every irrational employer, afraid to demand a wage cut that he knew would be necessary, there's mass of irrational employees afraid to ask for a wage increase which they merit."

Really? In 35 years in the labor force, nearly every annual performance review came with a raise - except for a few years when the company was losing money fast. (That's a good time to look for another job, regardless of whether you're getting paid what you're worth.) I have asked for a larger raise - by sending my boss a survey of salaries in similar jobs, for him to use in arguing with upper management to try to bring , well before the annual performance review.

If someone is afraid to ask for a raise, I'd conclude that either they know they are already getting paid at their market value or higher, or they're working for a vicious nutcase. And the second situation is a really good reason to go looking for another job, unless you're willing to tolerate it because no sane employer would pay you anywhere near as much...

Prices of goods and services that are sold in separate discrete transactions (like hamburgers, gasoline, or anything from Walmart) are usually free to change in response to market changes. Wages and prices of goods and services that are sold on a continuing basis (like utilities, rent, or a daycare fee) tend to be quite sticky. Cut pay and employees feel insulted and look for other jobs - I suspect that in that situation many will change jobs if possible even when the pay and benefits are exactly the same as the pay after the cut at their previous employer. And if I feel that I'm getting old and my expenses are down since my kids are out on their own, it's still pretty risky for me to suggest that I work not quite as hard for not quite as much pay - finding a new job that fits me better is far more likely to work out...

Likewise, a rent increase has to be justified to the renter, but you aren't going to get a rent decrease just because the place is getting older and less valuable. The only way that rent is going down is to a new renter - after you've moved out and the dump has been vacant for long enough to bring the owner to reality.

Mild inflation can ease these situations. The employer who thinks some employees are a little overpaid doesn't cut back on pay, instead they freeze raises or keep them below inflation. A landlord that is keeping the building in good condition can tell the tenants that the rent is going up because of inflation. Or if the place is paid off long ago and now showing it's age, leave the rents where they are and the gradually falling real value of the rent will more or less reflect the gradually falling value of the apartments, without anyone admitting anything...

Posted by markm | September 19, 2007 4:54 PM