The Fed doesn’t have a crystal ball
The following quotation from Nouriel Roubini seems to express a commonly held view:
Asked to grade the central bank’s job, Roubini gave the Fed a “D” for missing the crisis altogether and downplaying its possible impact, but a “B-plus” after the credit debacle had unfolded.
“I give them credit for being very creative and very aggressive,” he said.
I just don’t get this view. All the major investment banks with their million dollar Ivy League employees missed this crisis (and its eventual impact), and yet the Fed was supposed to have predicted it? The Fed pays much lower salaries than Wall Street.
In modern teaching an “A” is the grade you get when you have done nothing wrong. So I give the Fed an A for predicting the crisis, because they did nothing wrong there, and a D for adopting a monetary policy far too contractionary for the needs of the economy in September and October 2008, when things obviously starting getting much worse. I’d be happy if they simply did their job by keeping expected NGDP growing on a 5% track, I don’t expect them to be some sort of Nostradamus.
Tags:
17. June 2009 at 06:43
I think you sometimes forget that not everyone agrees with your definition of the Fed’s “job.” I want the Fed to “just do their job” too, but I have in mind stable prices or, even better, no injections of more fiat money.
17. June 2009 at 06:46
is it the firefighter’s job to water your lawn? 😉
17. June 2009 at 08:09
“The time to worry about depressions is, unfortunately, when they are furthest from the minds of most people.”
— F. A. Hayek
Dozens of Hayekian economists and market watchers _did_ call the boom and bust. The didn’t identify every detail and didn’t predict exact dates and times — but they focus a spotlight on many of the key elements and explained what was going to happen.
See for example Peter Schiff’s stuff …
17. June 2009 at 10:52
Scott, for once I’m totally flummoxed by one of your posts. You say you can’t expect the Fed to have predicted the crisis because their economists are not paid enough, but then you say they did predict the crisis. Actually the Fed did underestimate the severity of the crisis for a long time. I can remember when Bernanke testified that they expected total mortgage losses of $100 billion.
I don’t buy your simplistic view of incentives either. There could be economists that find working for the Fed more challenging and fulfilling than working on Wall Street.
I give your blog an “A” but this post a “D”. I hope you’re just having an off day.
17. June 2009 at 11:30
Do you think they have the information and tools necessary, over the short as well as the long term, in the fine as well as coarse control? Gdp futures would help. I don’t think the Fed itself thinks it is as knowing and powerful as it would have to be. I am not sure that is modesty or experience.
17. June 2009 at 15:39
Scott,
Saying that the Fed didn’t predict the crisis is like saying that the bank robbers didn’t predict that an alarm would go off and the cops would show up. It’s not the Fed’s job to predict crises, and it’s not a bank robber’s job to predict that he’ll get caught.
The Fed is in the business of helping its paymaster Our Enemy, the State control the economy as best it can, monopolize the money supply, influence interest rates and other data, er I mean stabilize the economy and the price level, and maintain full employment in accordance with the ’46 diktat-cum-law.
Bank robbers are in the business of doing what the IRS does, only they don’t legitimize their actions by invoking the flag, national security, public goods, and the Keynesian multiplier, among other statolatries and absurdities.
You want the Fed to do its job the way you would want a bank robber to do his job. I’d rather see both retire from their respective criminal activities.
Lysander Spooner pointed out that all legislation is “an absurdity, a crime, and a usurpation.” It was an act of legislation that created the Fed, so therefore the Fed is a criminal organization.
17. June 2009 at 17:20
[…] — Scott Sumner […]
17. June 2009 at 18:12
Some of your libertarian readers object to all government management of the economy, in part as violating human rights (in their fundamentalist view of the latter), but also because they believe, Austrian-style, that central authorities cannot acquire the widely dispersed information they would need to do a good job. Therefore they are unhappy, in particular, about your relatively accommodating (rather than hostile) attitude toward the Fed. But I take it you think that the information the Fed needs *is* readily available, from the markets. Whatever your view of other forms of government management of the economy, you consider *government management of the money supply* to be quite feasible.
Two questions: (1) Do you also think it *preferable* to a laissez-faire monetary system? (2) What other sorts of government economic management do you consider feasible, by relying on the information available in markets?
18. June 2009 at 00:45
I think Bill Stepp’s tongue is a little inside his cheek. But, Philo puts the problem quite clearly though.
If the Fed can’t predict crises then what’s the point of it. Banks in a free-market monetary system could reinflate themselves – if their customers and shareholders wanted that.
18. June 2009 at 03:59
Bob, I would have been happy if the Fed had followed its own definition. They clearly have some sort of implicit target or goal for the economy, and they clearly set their fed funds target at a level where even they thought the economy would under-perform their goal in October 2008. They failed by their own criteria.
Alex, No.
Greg, Your comment doesn’t directly undercut my view that the Fed should not be able to outsmart the market consensus. There are 6.6 billion people in the world. In any stock upswing there will be people crying bubble. Sometimes they will be wrong (1980s), and sometimes it will look like they were right, ex post (late 1990s.) That’s the nature of markets. But the Fed has no way of knowing when the people crying bubble are going to be eventually shown to be right. How could they know if the market consensus can’t figure it out. And your comment that they didn’t provide “exact dates and times” is very important. Galbraith predicted the 1987 stock market crash, but he did so in January 1987, when the information had no value because stocks hadn’t yet reached the value they held AFTER the crash. So it would have been stupid to sell on his advice. I am not at all impressed by doomsters who are always crying doom, and then say “I told you so” when some highflying market crashes. The truth is that Nouriel Roubini is famous for a reason, much as others might resent that fact. He is just about the only person who predicted the full crisis, and explained exactly why it was going to happen. Most others just predicted fragments.
PhilP, It wasn’t quite as stupid as you think, although certainly not one of my better posts. What I meant was that the Fed should not have been expected to predict the original subprime crisis. When markets turned very bearish during late 2008 the Fed should have observed (not predicted) those markets, and acted accordingly. They needed a monetary policy expansionary enough to keep 5 year TIPS spreads over 1.5%, at a minimum. (Of course I prefer NGDP targeting, but we lack precise data.)
Yes, those who don’t care about money may work for the Fed for idealistic reasons. But I was once offered a job by the NY Fed. That shows they are willing to hire people who cannot outsmart the markets (as I certainly cannot outsmart the markets.) I have also talked to lots of other economists in my career, and I haven’t found many Nostradamuses. Indeed I don’t recall meeting any. (But I’ve never met Roubini.)
Lord, We desperately need an NGDP futures market. But even so, they had enough info by September 16th to act much more boldly. I have a new post showing this, which will be up soon.
Bill, Until we reach anarchy, I’ll keep pushing the Fed to do as little damage as possible. Keep your power dry!
Philo, I’m not sure about laissez-faire money, as I don’t know what it would look like. My hunch is that if just one country did so, the banks of that country would produce banknotes backed by dollars or euros. If all governments exited money, then perhaps gold would become the medium of account. It’s hard to know how well a laissez-faire gold standard would do. In the real world the gold standard was often abandoned in war, and once you do stuff like that, it doesn’t work as well as fiat money.
Overall I am a pragmatic libertarian. It seems to me that the government’s major role in the economy should be dealing with complex property rights problems, aka externalities like global warming. If global warming is real (and most scientists think it is), you might want a carbon tax. The government can also provide a safety net. I actually have an open mind on most issues regarding the role of government. I think it is quite plausible that private charities can handle the safety net better than government, but also quite plausible that governments can do better. Thus unlike dogmatic libertarians, I don’t view taxation for redistribution as “theft.” Rather I’d want to look at evidence as to how well these programs work, and then try to improve them.
18. June 2009 at 04:23
Scott, you should read George Selgin and Laurence White about Scottish Free-Banking sometime.
18. June 2009 at 06:56
“It seems to me that the government’s major role in the economy should be dealing with complex property rights problems, aka externalities like global warming.” These problems are very complex and difficult, so we mustn’t expect governments to “solve” them; in evaluating government performance we must avoid setting the bar unrealistically high. But my question was whether information available in markets–perhaps specific “prediction markets”–could help.
“The government can also provide a safety net.” Again, is there any way of harnessing the marketplace’s function of aggregating information so as to enable the government to do this job better?
18. June 2009 at 12:16
Imagine a world in which economic events result from a combination of causes, some predictable and manageable, and others random (both truly random, as per random chance, and effectively random in that there is at yet insufficient knowledge and understanding of them to foresee their consequences).
In that world, if the Fed and other central banks performed their job perfectly all predictable recessions would be averted, making the world a better place. Yet recessions would still occur, although less frequently than in earlier eras.
And when they did, pundits everywhere would say: “Once again, the Fed didn’t see it coming. The Fed never sees any recession coming. It’s zero for the last ten! How incompetent can these central bankers be? What good are they?”
One can’t prevent the results of the unknown from becoming true.
But once the results of the unknown start being realized in a recognizable, familiar form, one ought to be able to do something about it. The quicker the better.
18. June 2009 at 15:10
It looks like Hayek is on your side, Scott:
Hayek in P&P (108-109):
“It does not seem open to doubt that the amount of money necessary to carry on the trade of a country fluctuates regularly with the seasons, and that central banks should respond to these changes in the ” demand for money “, that not only can they do this without doing harm, but that they must do so if they are not to cause serious disturbances. It is also a fact which has been established by long experience, that in times of crisis central banks should give increased accommodation and extend thereby their circulation in order to prevent panics, and that they can do it to a great extent without effects which are injurious.”
and better yet on 121:
“But I think that what I have already said on this point will be sufficient to justify the conclusion that changes in the demand for money caused by changes in the proportion between the total flow of goods to that part of it which is effected by money, or, as we may tentatively call that proportion, of the co-efficient of money transactions, should be justified by changes in the volume of money if money is to remain neutral towards the price system and the structure of production.”
18. June 2009 at 15:16
This, of course, would not be an accurate characterization of Austrian business cycle folks:
“doomsters who are always crying doom, and then say “I told you so” when some highflying market crashes.”
Some of them were fairly specific in both details and in saying how unusual the current situation was, as compared with other times. Was Roubini more specific? I’m not sure he was.
19. June 2009 at 01:32
Just to backup what Greg has said. In Price & Production Hayek recommends to central banks that the product MV (or PQ) should remain fixed. He says on page 124 “any change in the velocity of circulation would have to be compensated by a reciprocal change in the amount of money in circulation if
money is to remain neutral towards prices.”
See the comment Steve Horowitz wrote on the Austrian Economists blog recently:
http://austrianeconomists.typepad.com/weblog/2009/06/keynes-versus-hayek-round-78.html?cid=6a00d83451eb0069e201157129eb66970b#comment-6a00d83451eb0069e201157129eb66970b
However, none of this justifies having a central bank. The commercial banks would do something very similar without it and probably do a better job.
Anyway, I’m going to shut up about what so-and-so said, it’s a boring subject.
19. June 2009 at 04:01
Current, I read that a long time ago, and will reread it sometime this year (but right now am overwhelmed with work.)
Philo, Aaron Jackson and I wrote a paper discussing how prediction markets could be used to predict CO2 concentrations and global average temps.
I don’t know how prediction markets could be used to figure out the best safety net, but I would guess there are ways. They can be used almost any time there are uncertainties about the effects of policy changes (such as welfare reform.) I also think we should look at successful natural experiments, such as Singapore’s safety net, which focuses on self-insurance, and hence requires very little government spending.
Jim Glass, I presented a paper last year in Washington on exactly that idea. If the Fed is doing its job properly, any recession should be unpredictable. But note that the converse may not be true, an unpredicted recession doesn’t imply they are doing their job, especially if the intensification of the recession is predictable—see my newest post.
Greg, Thanks, those are exactly my views.
Greg, I don’t know enough to contest your observation, but Roubini had very specific knowledge of the problems in the banking system. I’d wager that side by side his predictions would look more detailed and impressive. But I don’t doubt that others expressed a general concern about housing and banking.
There is a bigger issue, however. The Fed cannot rely on the views of individuals, BECAUSE EX ANTE THEY DON’T KNOW WHICH INDIVIDUALS WILL BE THIS DECADE’S NOSTRADAMUS. They MUST rely on consensus forecasts, that is an institutional necessity. In my newest post I show that things got really bad precisely when they ignored the consensus forecast.
Current, Yes, I knew that Hayek had proposed this NGDP targeting idea. It is similar to mine, except the average inflation rate would be lower (negative 3% vs Bill Woolsey’s 0% and my 2%.)
It is not obvious that commercial banks in a competitive industry can do as well as a central bank. Why? because macro stability is a classic public good. It’s like saying individual companies can deal with global warming better than the government. I certainly think it is possible, maybe even likely, that the government will botch its global warming policy. And we know they have botched monetary policy. But no individual bank would have the incentive to issue the exact number of banknotes that would stabilize NGDP. Indeed, we have no idea what the price level would look like under laissez-faire. If there were competing media of accounts (as Hayek proposed if I am not mistaken) then you have a problem with mass confusion in the pricing system. If private money adopted a single unit of account like gold, there is no guarantee that we’d get macro stability. It might actually be better in that case to let big banks collude, but then they’d be acting like a quasi-central bank, and the public might just demand a real central bank.
19. June 2009 at 05:42
Scott: “I read that a long time ago, and will reread it sometime this year (but right now am overwhelmed with work.)”
Fair enough.
Scott: “Yes, I knew that Hayek had proposed this NGDP targeting idea. It is similar to mine, except the average inflation rate would be lower (negative 3% vs Bill Woolsey’s 0% and my 2%.)”
Yes. That difference is important if there is a time structure of production.
Scott: “But no individual bank would have the incentive to issue the exact number of banknotes that would stabilize NGDP.”
NGDP is an aggregate though. Here is the problem, mainstream economists have aggregated the problem. Then mainstream economists have accidentally come to the conclusion that doing that shows that the problem is one of aggregates.
A bank doesn’t need to know how many banknotes to issue to satisfy an overall aggregate. As we’ve discussed earlier you don’t see Central banks as predictors, and I take it you don’t see Commercial banks that way either. Both need to respond.
Now, what if there is increase uncertainty amongst the people of County Limerick in Ireland? What will happen is that people will ask for loans rather than cut back expenditure. Under free banking what would happen is that the local banks of issue accommodate that increase demand by issuing money substitutes. They do so while ensuring that their own books are sound – otherwise their shareholder and depositors would leave them.
So, the money supply becomes something dealt with locally.
Free banks must be backed by real savings. So the interest rate comes about by time preference. That means there is no overall distortion of the structure of production.
I agree that this is not obvious. But I think that it is what would happen. Hayek, Von Mises, and recently White and Selgin give good explanations why.
Read the posts Horowitz & Selgin gave today on the Austrian Economists blog:
http://austrianeconomists.typepad.com/weblog/2009/06/which-monetary-policy-rule-suffers-from-the-fatal-conceit.html
I agree that the question of how all this would apply today is a difficult one.
19. June 2009 at 05:51
Scott: “If there were competing media of accounts (as Hayek proposed if I am not mistaken) then you have a problem with mass confusion in the pricing system.”
Also, Austrians these days think that Hayek was wrong about some things he said in the 70s about competition in money. Competing fiat money is unlikely to be viable. Competing fractional-reserve commodity money however has worked in the past.
19. June 2009 at 06:27
Pardon my ignorance. It just struck me that governments that relied on markets to set policy might escape most of the force of the Austrian objection that markets but not central authorities are able to deal appropriately with widely dispersed information. But obviously I am way behind the curve, and ought to read some Aaron Jackson. (Might you provide a reference to your article with Jackson, or is it unpublished?)
19. June 2009 at 06:44
Philo: “It just struck me that governments that relied on markets to set policy might escape most of the force of the Austrian objection that markets but not central authorities are able to deal appropriately with widely dispersed information.”
I think it helps. The problem is that forcing some market into a certain shape then expecting it to work well is not wise.
For example, consider the market for Stout. We could legislate so that Stout is sold on a centralized market from 9am to 5pm every weekday in Dublin.
To obtain Stout a tippler phones the Stout market and puts in a bid for a quantity specifying Guinness or Murphys, etc.
The Stout is then couriered to the customers the next working day when they are invoiced.
Such a market allows market forces to play a role. However, it may well be better to allow the shape of the market to be determined by market forces too.
20. June 2009 at 08:03
Current, No, there really isn’t any significant difference between -2%, 0% and 2% inflation, and least not in the sense that most people assume. (The tax consequences are a bit different, and there is less chance of liquidity traps at higher rates.) Whenever I read Austrian and Post Keynesian econ I get the sense that they don’t understand, or don’t believe in the superneutrality of money. This is the idea that small, once and for all, changes in the trend rate of inflation don’t have important real effects. In a world of 0% inflation, workers get used to 2% pay increases, in a world of 2% inflation workers get used to 4% wage increases. The real wage in unaffected. The real interest rate is unaffected. Austrians seem to have this idea that a higher trend rate of inflation creates bubbles. Not so, as nominal interest rates rise one for one with trend inflation.
It is also not true that free banking will produce macroeconomic stability. Assume the most common type of free banking, the gold standard. Now assume a major gold discovery. How does free banking stabilize the price level in the short run, and the long run? You will find it won’t. The expectation that gold will lose value will reduce the real demand for gold, making prices rise by even more than the QTM would predict.
BTW, I completely agree with the Horowitz piece, except for the last part where he claims that free banking will supply the right amount. I don’t see what incentive banks would have to do that. Stable prices or stable NGDP doesn’t appear in their profit function. Having said that, i am not opposed to free banking, but would prefer a central authority (Fed or private bank cartel) control the supply if interbank reserves. So I’m not anti-Austrian/libertarian. But I think there are network effects.
Current#2, I don’t agree that free banking “has worked” in the past, if you mean resulted in less volatility in the inflation rate, or NGDP growth rate, than we had between 1982 and 2007. If you mean it has produced rough stability in the inflation rate, fine, but so has fiat money since 1982
Philo, You can find it on Aaron’s website:
http://aaronljackson.net/
Current3#, I agree that artificial markets don’t always work well. An example is the California electricity “market”, and I fear the carbon permit market may also have problems. But some have done pretty well. The market for coal pollution (SO2?) did a pretty good job of cleaning the air at reasonable cost.
23. June 2009 at 06:22
Scott: “No, there really isn’t any significant difference between -2%, 0% and 2% inflation, and least not in the sense that most people assume. (The tax consequences are a bit different, and there is less chance of liquidity traps at higher rates.) Whenever I read Austrian and Post Keynesian econ I get the sense that they don’t understand, or don’t believe in the superneutrality of money. This is the idea that small, once and for all, changes in the trend rate of inflation don’t have important real effects. In a world of 0% inflation, workers get used to 2% pay increases, in a world of 2% inflation workers get used to 4% wage increases. The real wage in unaffected. The real interest rate is unaffected. Austrians seem to have this idea that a higher trend rate of inflation creates bubbles. Not so, as nominal interest rates rise one for one with trend inflation.”
That’s right, we don’t believe in the things that you have mentioned above. The example you give of a worker getting 4% wage increase doesn’t work in practice. My own wage, for example, didn’t increase this year. However, I can’t use the CPI to tell what this means in real terms. I don’t buy the items on the CPI in the same proportion as the bundle, not even close. What affects my actions is how my bundle of goods move in relation to money.
What mainstream economists such as yourself fail to recognize is that an increase in the money supply must take place at some distinct place. There is a place where the money first comes in, an injection point normally many of them.
These injection points occur in finance and business. Agents in these areas are even less able to recognize real effects than I am as a worker. Consider a unique business such as Apple for example. Now, the Federal reserve inject money into the financial industry. Those who are the first receivers can buy existing assets at a lower price. It may be that the owners of those existing assets can accurately expect this and mark those assets up to compensate. In my view though that is unrealistic, the information for doing so doesn’t exist in one place. It would take huge efforts and cost to bring collect it.
See Horowitz’s article “The Costs of Inflation Revisited”:
http://www.gmu.edu/rae/archives/VOL16_1_2003/5_Horwitz.pdf
Scott: “It is also not true that free banking will produce macroeconomic stability. Assume the most common type of free banking, the gold standard. Now assume a major gold discovery. How does free banking stabilize the price level in the short run, and the long run? You will find it won’t. The expectation that gold will lose value will reduce the real demand for gold, making prices rise by even more than the QTM would predict.”
The question about gold discoveries is a complicated one. I agree that they may cause changes to the price level from the money side. I would argue though that, firstly, a gold discovery would not be large compared to current stocks. Secondly, we have persistent shocks like this from the central banking system. Why not get rid of those and have only one every few decades.
Scott: “BTW, I completely agree with the Horowitz piece, except for the last part where he claims that free banking will supply the right amount. I don’t see what incentive banks would have to do that. Stable prices or stable NGDP doesn’t appear in their profit function. Having said that, i am not opposed to free banking, but would prefer a central authority (Fed or private bank cartel) control the supply if interbank reserves.”
To summarize the free-banking proponents point of view. Each bank issues their own money. Individuals and other banks ask for the money to be redeemed in gold from time-to-time. Mostly this will be other banks requesting this through clearing banks. A bank issues banknotes backed by gold and assets, these circulate. The bank has an interest in maintaining a sensible supply of money. If uncertainty increases and the amount of money held increases then there will be less calls for the bank to redeem. It may therefore reduce the fraction of gold reserves it holds by issuing more notes. The opposite happens when there is a decrease in money holding.
Scott: “I don’t agree that free banking “has worked” in the past, if you mean resulted in less volatility in the inflation rate, or NGDP growth rate, than we had between 1982 and 2007. If you mean it has produced rough stability in the inflation rate, fine, but so has fiat money since 1982”
It’s a question open to interpretation I agree. However, I find your dates interesting. Why start at 1982? Why not 1912? I don’t think the overall performance of central banking has been particularly good.
Scott: “I agree that artificial markets don’t always work well. An example is the California electricity “market”,”
Yes. The UK electricity market works well too.
However, just because a market can be centralised this way doesn’t mean that it’s best to do that. Electricity isn’t like money. Once electricity goes into the grid it is indistinguishable. Money need not be that way.
24. June 2009 at 06:01
Current, The Horowitz article doesn’t relate to my point about superneutrality, because he is talking about changes in inflation uncertainty, which I am holding constant.
The fact that you didn’t get a 4% pay increase doesn’t undercut my argument, as we didn’t have 2% inflation this year, we had negative 1%. And even if we had had 2% inflation, my argument still would not be affected. You will get the same real pay increases at 2% inflation as at 0% inflation, regardless of your individual consumption bundle. Your bundle may show more or less inflation than the official rate, but that would be equally true at 0% or 2% inflation.
You are wrong that we have had big money supply shocks from central banks in recent history. Most of the instability occurs from money demand shocks, which are just as likely to occur under a gold standard. Indeed the rate of inflation was more unstable under the gold standard than in recent decades. The average rate was lower, I admit, but superneutrality shows us that the average doesn’t matter. And again, since the Horowitz paper refers to inflation uncertainty, it certainly doesn’t provide any support for the gold standard.
And I don’t agree with the following assertion:
“The bank has an interest in maintaining a sensible supply of money.”
The bank has an interest in maintaining the profit maximizing amount of money. That’s why you guys have to pray that there is no huge gold discovery, as there is no self correctly mechanism. But you also underestimate gold demand shocks.
I used 1982 because that’s when the world’s central banks figured out how to do fiat money. I am not a fan of central banking, I just want to make our system work better. We have a Fed, and will continue to have one. And it can be improved, just as it has already been improved. I wish the Fed had not been created in 1913, it certainly made things much worse in the short run.
You misread my last point, I was agreeing with you that artificial markets often don’t work well (as in California, which worked poorly.)
24. June 2009 at 09:59
Scott: “The Horowitz article doesn’t relate to my point about superneutrality, because he is talking about changes in inflation uncertainty, which I am holding constant.
The fact that you didn’t get a 4% pay increase doesn’t undercut my argument, as we didn’t have 2% inflation this year, we had negative 1%. And even if we had had 2% inflation, my argument still would not be affected. You will get the same real pay increases at 2% inflation as at 0% inflation, regardless of your individual consumption bundle. Your bundle may show more or less inflation than the official rate, but that would be equally true at 0% or 2% inflation.”
As I understand it what you are trying to say is that steady and expected inflation can’t affect real variables. I disagree, so does Horowitz. I’ll quote what Horowitz says on page 81.
Horowitz: “This understanding of the relative price effects of inflation raises suggests that the standard distinction between anticipated and unanticipated inflation is problematic (Laidler 1990:51). That distinction assumes that if one knows that the monetary authority intends a five percent increase in the money supply, then the price level and the price of one’s own goods or services will also rise by five percent. Note the shift from the generally correct proposition that increases in the money supply will proportionately affect the overall level of prices, to the claim that each and every price will be affected by that amount, if the money supply increase is expected. When one drops the assumption that money enters the economy in proportion to the existing holdings of money users, then the leap from what is true of theaverage to what is true of individual prices collapses. Even if the amount of the money supply increase is known, one would have to know also the precise path the additional money would be taking to know how the price of one’s own good or service will be affected by inflation.10 For example, it is conceivable that if a particular good is highly demanded by those who are disproportionately favored by the inflation, then wewould expect a rather significant rise in that good’s price, particularly relative to those goods not demanded by the same people. Even after a one-shot inflation worked its way through, there would be no assurance that any given price (and surely not every price) would have increased by precisely the percentage change in the money supply.11 Although knowing the money supply increase would provide the best guess at one’s own price change (since it equals the average price level effect), there is no reason to believe that it would be correct in any specific number of particular cases. To argue that anticipated inflations would be neutral because individuals would know the effect on the general price level is only valid if one assumes that the effect on each and every price is the same, and is equal to the average effect. Given how unlikely that outcome is, even anticipated inflation (in the sense of knowing the behavior of the monetary authority) will still generate relative price effects. In the discussion to follow, we will not distinguish between anticipated and unanticipated inflations, other than to note that the costs of inflation will be somewhat greater in unanticipated inflations, but not by enough to justify distinct treatment.”
and on p.80 “In fact, excess supplies of money enter the market at specific times and in specific places depending on the particular actions taken by the monetary and fiscal authorities. Empirically, if additions to the money supply are made through open market operations, new reserves arrive at those banks who either sell securities directly to the Fed, or to those banks who have the accounts of security dealers who participate in FOMC transactions. Specific banks receive the new money first, and their decisions about what loans they will then make, and the spending decisions of the recipients of those loans, will be the proximate causes of a first round of relative price effects”
As you point out inflation has been higher. This is what causes the artificial lengthening of the structure of production in Austrian business cycle theory and the consequent bust. Horowitz’s paper goes on to mention that.
Scott: “The bank has an interest in maintaining the profit maximizing amount of money. That’s why you guys have to pray that there is no huge gold discovery, as there is no self correctly mechanism. But you also underestimate gold demand shocks.”
I accept that gold shocks could cause changes in the amount of money. I don’t think they need be as bad as they were in the 19th century when various and central bank actions made them much worse. I don’t think they would be as bad either because so many of the rich deposits of gold have now been mined out.
I don’t think that the problem of gold shocks is as bad though as the problem of persistent inflation. As the Austrian economists explain that will cause booms and subsequent busts. I would agree with you if I thought that you were right about persistent money supply increase and inflation but I don’t.
Scott: “I used 1982 because that’s when the world’s central banks figured out how to do fiat money.”
I see.
Scott: “I am not a fan of central banking, I just want to make our system work better. We have a Fed, and will continue to have one. And it can be improved, just as it has already been improved. I wish the Fed had not been created in 1913, it certainly made things much worse in the short run.”
I think we agree that things need improving. The question is how? I would be open to a method of improvement that involved retaining the Fed.
Scott: “You misread my last point, I was agreeing with you that artificial markets often don’t work well (as in California, which worked poorly.)”
OK, I see.
25. June 2009 at 05:24
Current, It would take me too long to explain all the flaws with the Horowitz quote you cite. But classical theory certainly does not predict that if prices are expected to rise by 5%, then all individual nominal prices will also be expected to rise by 5%. Classical theory does allow for expected changes in relative prices. I also don’t understand his point about the transmission mechanism. For anticipated money, it makes no difference where it is inserted into the economy.
Regarding gold, I agree the supply increases are likely to be pretty gradual, but that is an empirical judgment. Once you concede that free banking could not prevent a massive discovery from being inflationary, you have also conceded the same about a massive drop in bank gold demand. And instability in gold demand is far more likely than instability in gold supply, and in fact it has always been that way. So that doesn’t solve the problem. The Austrians sort of wave their hands and assume that shifts in gold demand won’t be a problem, but they have no model to show that.
25. June 2009 at 08:47
Scott: “It would take me too long to explain all the flaws with the Horowitz quote you cite.”
Fair enough. We have been talking a lot recently. Thank you for all your discussions by the way, they have been very interesting.
Scott: “But classical theory certainly does not predict that if prices are expected to rise by 5%, then all individual nominal prices will also be expected to rise by 5%.”
How is the actor supposed to change his behaviour in accordance with an expected rise in the general price level if that isn’t relevant to him? That is the problem.
Steve Horowitz may be wrong in what he says at the beginning of be piece I quote. I don’t see though how it affects the validity of the rest.
Scott: “I also don’t understand his point about the transmission mechanism. For anticipated money, it makes no difference where it is inserted into the economy.”
I think it does. Let’s say I know the Fed are going to increase the money supply by 5%. They will do that by buying bonds. How can that be equivalent to dropping money into peoples bank accounts in proportion to the amount that they are holding?
Scott: “And instability in gold demand is far more likely than instability in gold supply, and in fact it has always been that way. So that doesn’t solve the problem. The Austrians sort of wave their hands and assume that shifts in gold demand won’t be a problem, but they have no model to show that.”
Well I’m going to take an out here. It would take me too long to discuss this in detail. I’ll do that some other time. I’ve got a difficult argument in capital theory above to think about.
26. June 2009 at 06:24
Current, Economic actors make the same calculation under any trend rate of inflation. They take their own nominal pay raise, subtract out the trend rate of inflation, and get the real pay raise. The only advantage of zero inflation is that it makes the math easier. But of course Austrians don’t favor zero inflation. Positive 2% and negative 2% cause equal computational problems.
On the money injection, try this analogy. What difference would it make whether the government pumped water into Lake Erie, or Lake Ontario? If it was pumped into Lake Erie, it would simply flow over Niagara Falls and end up in Lake Ontario. Where the cash ends up a few days later has nothing to do with where it is first injected by the Fed.
30. June 2009 at 03:20
Scott: “Economic actors make the same calculation under any trend rate of inflation. They take their own nominal pay raise, subtract out the trend rate of inflation, and get the real pay raise. The only advantage of zero inflation is that it makes the math easier. But of course Austrians don’t favor zero inflation. Positive 2% and negative 2% cause equal computational problems.”
I agree with you that the problem is similar under different trend rates of inflation. I disagree though with your description of the problem.
If I take my nominal pay and subtract the CPI then I get a figure. However, that figure doesn’t directly relate to my behaviour. It doesn’t tell me how to act and is not really that useful. What is interesting to me is how my own income relate to my own expenditures. No one person buys the CPI basket of goods. For example, for myself Guinness is under-represented in the CPI basket.
This is even more true for businesses. HP for example can’t look at it’s revenue, profits and look at the producer price index and work out how well it’s doing. It can’t because like most businesses it’s costs are not firmly related to the PPI.
So, in the both cases individual actors need to estimate how price changes will affect them.
This is a problem because of how it interacts with the interest rate and production structure. But that’s a separate topic.
Scott: “On the money injection, try this analogy. What difference would it make whether the government pumped water into Lake Erie, or Lake Ontario? If it was pumped into Lake Erie, it would simply flow over Niagara Falls and end up in Lake Ontario. Where the cash ends up a few days later has nothing to do with where it is first injected by the Fed.”
I see what you mean but I don’t really agree.
When the Fed reduce rates they often do so by buying bonds. In that case they bid up the price of bonds. An increase in saving will do the same. However, there is a difference. An increase in saving will come with a corresponding reduction in consumption by the savers.
The price of bonds are bid upwards then those who sell bonds spend their money. They bid up the price of things that they buy. There are multiple differences to relative prices. Some expected, some not.
(Like the discussion about Keynes in the other threads this is about the difference between the market rate of interest and the natural rate. A central bank can force the market rate of interest below (or above) the natural rate.)
1. July 2009 at 04:26
Current, Fine, you can say the CPI is not accurate for individuals, but that is true of any index, including the alleged stability that Austrians think they can achieve. Tell me what index you’d like to see stabilized, and I’ll tell you that it wouldn’t apply to many individuals. So your comment actually supports my point.
You said,
“When the Fed reduce rates they often do so by buying bonds. In that case they bid up the price of bonds.”
There are so many problems here I hardly know were to start. First of all, it would be news to bondholders during the Jimmy Carter adminstration that easy money raises bond prices. We had the easiest money of my lifetime, and yet bond prices fell in half during the Carter years. Even Austrians believe that Volcker instituted a tighter money policy (or at least Austrians that post here) and yet bond prices soared. Second, there are cases where easy money can raise bond prices, but it has little to do with the fact that the Fed buys bonds. The same would happen if they bought stocks or land. Bond prices don’t rise (when they do) because bonds are bought, they rise because of the liquidity effect. The rise (if they do) only because prices are sticky in the short run and thus more nominal money means more real money until prices adjust. At that point interest rates are unchanged.
And there is no single natural rate of interest. The natural rate of interest is defined as the interest rate that provides macro stability. Thus there are as many natural rates as their are monetary targets.
1. July 2009 at 09:46
Scott: “Fine, you can say the CPI is not accurate for individuals, but that is true of any index, including the alleged stability that Austrians think they can achieve. Tell me what index you’d like to see stabilized, and I’ll tell you that it wouldn’t apply to many individuals. So your comment actually supports my point.”
Yes. But, the question we need to ask ourselves is whether the actions of individuals will correlate or not. Whether when taken together individual decisions will follow a pattern.
In this case I think the answer is yes. From the initial injection points the money enters markets and causes prices to change. It then spreads through the economy in a definite direction – from investment goods down.
A target such as stabilizing MV=PT doesn’t do that in my opinion. Or, at least, when it does it it does it to the least possible degree.
This change is only one of the things it does, the other is to change the relative price between present and future goods. Which is where debates over the natural rate of interest come in.
Current: “When the Fed reduce rates they often do so by buying bonds. In that case they bid up the price of bonds.”
Scott: “There are so many problems here I hardly know were to start.”
I should have said ceteris paribus in there somewhere. I agree that those other things happen that you mention. As you say later “The rise (if they do) only because prices are sticky in the short run and thus more nominal money means more real money until prices adjust.” That’s exactly the point.
Scott: “And there is no single natural rate of interest. The natural rate of interest is defined as the interest rate that provides macro stability. Thus there are as many natural rates as their are monetary targets.”
Well, what “natural rate means” is a matter of definitions. Your definition is one of a number of sensible definitions.
What I meant was that an injection of money can force the market rate of interest below the rate which produces equilibrium in the savings/investments market. (That saving/investment equilibrium can never be fully achieved, my point is that doing this forces things further away from what could be).
Anyway, I’m getting into Austrian theory here and I’d have to explain lots of it to make sense.
As I said in the previous post, I expect you are rather busy with this blog at present. I don’t mind at all if you don’t reply to this post. I’m sure the subject will come up again in the future. Besides I still have to reply to David Glasner and Bill Woolsey about interest theory.
1. July 2009 at 10:28
I think the Austrians have good claim on the natural rate of interest as having a single definition, and it is not the same as the ‘neutral rate’ of interest which is how Scott appears to be using the term.
2. July 2009 at 02:00
Jon,
Scott does seem to be talking about the neutral rate. However the question is complicated. To begin with there are a constellation of interest rates, short term ones and longer term ones. There are various different definitions of the natural rate depending on whether the pure time-preference theory of interest or a theory that includes productivity is being used.
Really this term should be defined whenever it is used. Anything else is too confusing.
2. July 2009 at 06:24
Current and Jon, As far as I know Wicksell came up with the “natural rate” concept, and he applied it to a situation where prices were stable. The natural rate was the rate that stabilized the price level. Equating savings and investment doesn’t help at all. They are always equal, the question is at what price level? It’s like the supply and demand for oil were equal in 2008, but at rapidly changing price levels.
If Austrians want their own definition of “natural rate” that’s fine, but don’t pretend it is the only definition that makes sense, as it is just as arbitrary as any other.
2. July 2009 at 07:06
This is the problem, the Austrian definition is different from Wicksell’s. Wicksell himself gave other expressions for what the natural rate represents.
Scott: “Equating savings and investment doesn’t help at all. They are always equal”
I wasn’t very clear there. There are two things that pay for investment. Firstly, there is saving where people put aside money and therefore restrict their consumption. Secondly, investment may be funded by new money created by a bank. Austrians don’t class this later instance as proper “saving”.
3. July 2009 at 12:22
Current, If the terms are defined correctly, savings always must equal investment. Money creation is a tax on existing money balances. I am pretty sure that means it should actually count as government revenue, and hence “saving” rather than a “deficit.” But perhaps someone will correct me. I’m not saying that money creation doesn’t have some interesting and distinctive features, but I happen to think they work through wage and price stickiness, not investment.
6. July 2009 at 01:43
Scott,
We should distinguish here between the short run and the long run.
Someone could say this about your view on current events. “Monetary expansion must lead to inflation through the money quantity equation MV=PT. V must go back to it’s long-run trend.” I’m sure you would reply to that criticism that it refers to a long-run trend. It is what Steve Horowitz calls a “Humean” time period. The time after which all the dust has settled. And, as we all know you can’t use two points each taken when the dust has settled to prove that the dust is not a problem.
Exactly the same applies to savings and investments. In the Humean long-run the money creation becomes a tax on money holdings. However, in a shorter run money issue has a similar effect on prices as an increase in saving does. It does not come along with any corresponding decrease consumption by any community of savers. (In the long run that decrease is forced upon money holders by the effect of the taxation on those money holdings.)
See slides 35-40 of the deck I mentioned earlier:
http://www.auburn.edu/~garriro/cbm.ppt
10. August 2009 at 15:04
[…] je ľahké povedaÅ¥, že kuvici mali pravdu a svet sa roky rútil do skazy. KumÅ¡t je v tom, ako dopredu zistiÅ¥, ktorý Nostradamus je ten pravý. Nie po boji, ale poÄas neho. Nehovoriac o tom, že krÃzu – […]