We don’t expect inflation, but we expect to expect it soon
Is the title sentence a logically monstrosity? I think so. How can one not expect something to happen, but nonetheless expect to expect it in the near future? I’ll leave that to the epistemologists but in some weird way I think this post’s title reveals how monetary policy went off course.
Any doubts you might have had that it is in fact seriously off course should be erased if you read this excellent expose of Fed policy from Tim Duy. (HT: StatsGuy) Here’s one passage:
As I await the employment report, I am reflecting on the flow of information over the past week and find myself somewhat dismayed by the apparent policy implications. The spate of FedSpeak in recent days leaves one with the uneasy feeling that monetary policymakers are more willing to use unconventional monetary policy to support Wall Street than Main Street. The most hawkish appear eager to normalize policy at the earliest opportunity possible, and even the dovish, grasping onto green shoots, appear to think they have done enough to support recovery. It is as if the FOMC has concluded that the risks are now entirely one-sided toward inflation.
. . .
With the primary build out of the internet backbone complete, the US appeared to experience a dearth of traditional investment opportunities (I suspect that the need to expand production domestically was made moot by an international financial arrangement that favored the establishment of productive capacity overseas), and, like water flowing downhill, capital was thus allocated this decade to residential investment, which, we now know was more about consumption than investment, and the resulting economic activity was anemic by historical standards.
This line of argument leads one to believe that withdrawing monetary stimulus would be a significant policy error, especially if investment growth remains constrained as we saw this decade. In fact, it would lend additional credence to reports that the Fed needs to do much, much more – a massive, unsterilized expansion of the balance sheet – should they even hope to stimulate sufficient investment demand to absorb underutilized labor. Instead, FOMC members appear to be concerned that stimulative policy will be the root cause of the next financial crisis. That, however, appears to me to confuse monetary with regulatory policy. The former should speak to inducement to invest, while the latter speaks to protecting against significant misallocations of capital.
Then Duy quotes Richmond Fed President Lacker from a Bloomberg interview:
The Federal Reserve will need to raise interest rates when the economic recovery is “firmly” in place, even if unemployment lingers near 10 percent, Federal Reserve Bank of Richmond President Jeffrey Lacker said.
Duy’s reaction is exactly how I would have reacted:
Seriously, raising rates even if unemployment is 10%? LACKER SAYS THIS ON THE DAY WE GET CORE PCE INFLATION SLIDING TO 1.3% Y-O-Y! This redefines the term “hawk.”
From where does this fear of inflation emanate? That brings us to perma-hawk Philadelphia Fed President Charles Plosser:
“Unfortunately, slack was poorly measured and turned out to be not as significant as first estimated. Thus, the Fed’s monetary expansion led to rising inflation for the balance of the 1970s. One lesson learned during this episode is that inflation expectations can matter a great deal, and if they become unanchored “” that is, if the public comes to believe that the Fed will not do what is necessary to preserve price stability “” then inflation can rise quickly regardless of the amount of so-called slack in the economy. The price we paid to regain control of inflation and the Fed’s credibility to do so came in the form of the 1981-82 recession and was a steep one.”
Again, Duy’s reaction to Plosser hits the nail right on the head:
The experience of the 1970s is such a tired and faulted analogy. To generate a wage-price inflation spiral, you need to explain the mechanism by which rising inflation expectations (which don’t exist anyway) get translated into high wages. I do not see that current institutional arrangements in the US allow this; nor do we see it in the data:
Reading this is both exhilarating and depressing. Exhilarating because it’s good to see other economists talking about the need for more monetary stimulus, and depressing because it is one more blog that I need to read, and I am already overwhelmed. Read the whole post.
Now let’s take a recent look at inflation expectations:
Period TIPS Spreads CPI futures Average
2 year 0.50% 0.94% 0.72%
5 year 1.41% 1.71% 1.56%
On theoretical grounds you’d expect the CPI futures to overstate inflation (as risk averse people hedge against inflation risk in this market.) And the TIPS spreads should understate inflation expectations (because non-indexed bonds are slightly more liquid, and hence offer a bit lower expected return.) So I think the average is a reasonable ballpark estimate of inflation expectations. And I’d add that recent actual inflation is also low, as is the consensus forecast of economists. That doesn’t mean the forecasts won’t be wrong, but it does suggest that the consensus forecast, however measured, is for low inflation.
So given that the Fed is widely viewed as shooting for about 2% actual inflation, shouldn’t they be trying to raise inflation expectations, rather than lower them? You might argue: “It’s more complicated, the Fed isn’t a strict inflation targeter.” That’s right, as in the Taylor Rule, they also look at real output relative to capacity. And in every single new Keynesian model you can name, the Fed is supposed to err on the side of higher inflation when output is below trend, and aim for lower than normal inflation when output is above trend. So under strict inflation targeting the Fed is making a tragic mistake, they should be trying to encourage higher inflation expectations, and yet for some reasons they are encouraging lower inflation expectations. And under a more realistic “flexible inflation targeting” regime? Well then they are making an even more tragic error, they are even further off course. Tim Duy’s frustration is exactly appropriate. Fed policy right now can only be described as bizarre. It is not clear why they are doing what they are doing.
So let’s speculate a bit. It seems to me that the Fed is for some strange reason assuming that there is a logical distinction between inflation expectations and expectations of inflation expectations. In other words, they see that inflation expectations aren’t a problem right now, but they expect them to be a problem in the near future unless they tamp them down with some hawkish talk. But that commits a fundamental logical error; the Fed is ignoring the fact that those inflation expectations (both market and private forecasters) are formed with knowledge of current and expected future Fed policy, including the 0.25% target rate and the massively bloated monetary base that everyone seems worried about.
[BTW, I wish the Fed would stop calling it “base” money; bank reserves are now essentially T-bills. Only currency is still interest-free. And monetary theories of inflation are based on explaining the supply and demand for non-interest–bearing money.]
Here’s what the Fed doesn’t seem to realize. If the market thinks inflation will be too low under current and expected future Fed policy, then the Fed needs to send out more dovish signals not hawkish signals. They need to get 2-year inflation expectations up to around 2% or 3%. And that would require a much more expansionary monetary policy. Yes that’s right, I’m suggesting we go back to how things used to be in 1958 and 1983, when very steep recessions were followed by very rapid recoveries. Why not? Instead, everyone is estimating really high unemployment for years to come. Correct me if I am wrong, but I don’t recall the 1958 and 1983 recoveries leading to double digit inflation. Sure we’ve got some structural problems, but in 1983 we were in the midst of a painful downsizing of the so-called rustbelt’s manufacturing capacity as a result of technology and foreign competition. That was a huge structural problem, similar to our overbuilt housing stock. Indeed, because of population growth the housing overhang may be solved more quickly than that earlier structural adjustment. And yet RGDP grew very fast in 1983-84.
So if the markets don’t expect high inflation that means the markets don’t expect to expect high inflation. And the Fed shouldn’t either. For the umpteenth time, there are no lags between monetary policy and changes in inflation expectations.
Tags:
5. October 2009 at 17:26
As long as expectations are anchored, why does it matter if they are anchored at 2% or 0.72%?
5. October 2009 at 19:05
Epistemologists call the putatively violated norm of rationality “reflection”. Van Fraasen gave a diachronic Dutch Book argument in 1984 that gets cited a lot, but I doubt he was the first. Though put in terms of degree of belief modeled by a probability measure, reflection says pretty much precisely that you should currently expect what you expect to expect. In response to counterexamples (e.g., you know you’ll believe something stupid after a few rounds), reflection usually gets amended to include that you know that your expectation will only change by rational updating on evidence (viz., Bayesian or Jeffrey conditionalization).
Anyhow, if I’m following you correctly I don’t see how the Fed violates reflection with some hawkish talk… Maybe the Fed thinks the market’s irrational. That’s what it would take for market expectation of inflation to change despite full information (“knowledge of current and expected future Fed policy”): i.e., the Fed expects the market to violate conditionalization and thereby violate reflection. In any case, it doesn’t seem to me that the Fed is at all implicated in violating reflection by behaving as though they think the market will if they don’t hear some hawkish talk. They might be wrong, and the economists are far better qualified than the epistemologists to figure that out, but I don’t get how they’re violating reflection.
5. October 2009 at 19:06
Scott,
So, if you and Tim Duy are correct, then each sign of the economy stalling should be seen as a prelude to the Fed’s determination to further* raise inflation expectations.
If the above is correct, and the October employment report printed a 500k loss, then should a market actor with the following assets (i.e. buy or sell):
1) Treasury bonds
2) Gold (please assume jewelry demand is marginal — in terms of magnitude of change — compared to investment demand)
3) Equities?
*By “further”, I imply that for every increase in deflationary expectations, the Fed must attempt to raise inflationary expectations MORE. For instance, to counteract cash hoarding at a TIPS spread of 0%, the Fed must promise an inflation “tax” on cash of 3%. But to prevent cash hoarding at a TIPS of -3%, the Fed must impose a higher cost on cash holdings — say, inflation of 5%.
5. October 2009 at 20:16
Scott: The basic problem is that Fed’s position vis-a-vis the base is unstable as I discussed back in this context,
— The Cochrane Debate’
Once an inflation picks up, banks will be driven by nominal yields to convert reserves to loans. As they do so, inflation will accelerate. It is this hypothetical that is causing quite a bit of consternation. If the process gets started, it may run away.
But what to do about this? The Fed is jammed in a corner. I believe they feel (quite plausibly) that they cannot unwind their long-dated balance sheet today without lowering inflation expectations (but as you say that is not consistent with their target).
IMO, they are trying to hold steady until the economy is strong enough to endure reducing the base.
6. October 2009 at 04:42
How would, say Saudi Arabia for instance, agreeing to accept other currencies besides the dollar in exchange for oil affect the whole inflation expectations scenario? I’ve heard news reports that exactly such a thing is being discussed with OPEC between China, Japan, and a couple of other countries. I’ve read that Iraq did this in 2000, and Iran switched to euros in 2006. Could this have had anything to do with what you observe as contractonary monetary policy during the run up to the financial crisis last year in attempt to soak up the excess dollars floating around over seas?
I’m sorry if I have a lot of questions. I’m just finding out that our fiscal and monetary policy over the last few decades has had much to do with oil being priced in dollars only, and I’m concerned that it might be coming back to haunt us like a tempest in a tea pot.
6. October 2009 at 04:43
Regarding the title…
Imagine that you live on a hill, on a fairly large island surrounded by sea monsters and fog.
The government has made a commitment to deliver the mail to this island, but often, the ferry carrying the postal carrier and his jeep are eaten by Charybdis, or the Kraken or what-have-you.
But the government always ponies up for another ferry and postal worker and jeep. They keep copies of all mail, and send a copy with each delivery, so you will never lose mail, your stack of deliveries just gets larger with each transit.
And you sit on your hill, and you watch for the ferry to make it to the landing, and for the jeep to roll off the ferry.
But that jeep has a _lot_ of mail to deliver, and it is impossible to determine how long it will take before the jeep gets to your house.
Once you see the jeep, you can safely *expect* to get mail sometime in the reasonably near future. There will be some urgency then to write your letters and such, to make sure that they get delivered back to the mainland.
But, until you see the jeep, you can only expect to expect to get mail. There’s really no point in writing letters, since they may be hopelessly out of date by the time they are delivered.
In other words, shifting my investments around right now to avoid inflation may not be prudent, given that I don’t know how long it will be until the inflation arrives. I’m looking for a more definitive sign before I start to plan for inflation.
6. October 2009 at 05:11
Scott
When the Fed is assuming that people will expect that in the future expectations of inflation will rise it is in effect saying that the “law of iterated expectations” doesn´t hold. The “law” reflects the commonsense idea that the expectation of an expectation that will be based on more information than is currently available will simply be the expectation given the lesser available information.
Incredibly, then, by talking about tightening, the Fed is actually trying very hard to have people REDUCE their inflation expectations (from an already low level)today!
6. October 2009 at 05:15
Scott,
I suggest we set up an experiment to see whether the Fed can really control inflation expectations (not influence — control). Assume the existence of a costless, perfect hedge for ALL FUTURE increases in the price level. If you are right, and the Fed can fine-tune price level expectations, then there is little reason for this “hedge” instrument to outperform equities. That is, for every increase in NGDP expectations, there should be a similar increase in RGDP expectations.
Wait. Such an instrument does exist. Almost. Gold is not a perfect hedge, and there is some influence from consumption (jewelry demand), although the volatility in that demand is small compared to the volatility in “hedging” demand.
So watch gold. If gold outperforms equities, as it has for the past eight years, and if it does so strongly, then I suggest the market does not believe that, in the long term, NGDP and RGDP will travel in lockstep. Therefore, it would also be true that the market does not believe the Fed can fine tune inflation expectations.
BTW, you seem to make a number of assumptions about the impact of Fed actions on expectations. Specifically, you imply that they have a linear relationship: raise interest on reserves by x, and expectations will be a stable, linear function of that increase in rates. To my knowledge, there is little or no empirical or theoretical support for this view, as it ignores the obvious existence of feedback loops.
6. October 2009 at 06:06
Scott:
I don’t agree that interest bearing reserves are the same as T-bills. I don’t agree that money cannot bear interest.
Money is the medium of exchange. It is still the medium of exchange if it bears interest. T-bills aren’t the medium of exchange.
Money is currently the medium of account. It is still the medium of account if it bears interest. T-bills aren’t the medium of account.
If money earns interest, then the demand for money depends on the interest rate paid on money, as well as many other things. It is still the medium of account and the medium of exchange.
Because money is the medium of exchange (interest bearing or not,) it can be spent into existence and people can demand more by spending less on other things.
Because money is the medium of account, money denominated as a dollar cannot have a price of 99 cents or $1.05 to clear its market. Instead, the dollar prices of other things must change.
Again, what changes is that the interest rate on money isn’t always fixed at zero, and changes in that interest rate can impact the demand to hold money.
6. October 2009 at 07:56
Scott,
It seems to me that velocity “fell off a cliff” (i.e. money demand rose rapidly), and the Fed offset this by substantially expanding money supply. My impression is that the increase in money demand was a spike, and not a reflection of a long-run change in money demand. Are you saying that the market expects the increase in money demand to hold for a long time? That seems to me the only way to avoid inflation, but why should be expect a panic-induced spike in money demand to last for such a long time, especially once a recovery is underway and confidence returns?
6. October 2009 at 08:53
A couple of points from me:
(1) The cpi can be manipulated to some degree, so expected cpi probably understates meaningful inflation expectations. I am not saying that the official statisticians would lie, but where methodological ambiguities exist, the official statisticians can be leant on to come down on the convenient side (eg Boskin).
(2) As Bill says, some interest-bearing accounts would generally be considered to have some moneyness. The Belongia that you referred to a few posts ago literally wrote the book on how to judge the degree of moneyness.
6. October 2009 at 09:17
Scott,
I see one flaw in your argument. Markets might not be expecting inflation because they are anticipating that the Fed will not let inflation kick in. It is like if I´m in a plane and you ask me if I think that the plane will land safely. My answer would be yes, but that doesn´t mean that the pilot shouldn´t do their job to prevent the plane from crashing into the ground.
Alex.
6. October 2009 at 10:07
There are several actors in and around the Fed that are concerned about commodity prices and the value of the dollar. With the dollar falling and precious commodities on the rise, which some argue are better predictors/measures of inflation than TIPS, bonds, or the CPI, there is a school of thought that the current monetary expansion will result in inflation 2 to 5 years hence. While I would suggest that the empirical correlations that these observations are based on include only limited time-frames (and actually went negative last year), I don’t influence policy. Gold is a purely speculative asset, and speculators (including sovereign wealth funds) are hedging against future monetary uncertainty.
The other reason I suspect that the Fed is taking a hawkish stance in public is that it is attempting to educate and inform the public about how it can control inflation should should inflation pressure emerge, and it is under political pressure to reign in its balance sheet.
6. October 2009 at 11:00
pushmedia1, You said;
“As long as expectations are anchored, why does it matter if they are anchored at 2% or 0.72%?”
People need to step back from inflation and ask what sort of problem does American face? Because the Fed can only control inflation by controlling AD, it stands to reason that if low inflation is not a problem, then a lack of AD is also not a problem. That is a perfectly defensible opinion, but I just want people to be aware of the implications of their view that low inflation expectations aren’t a problem. It would be completely illogical to argue that we need more AD, but we (demand-side policymakers) don’t want higher inflation. So that’s my first point.
My second point is that the Fed called for fiscal stimulus. The purpose of fiscal stimulus is to raise AD. In other words it has the same purpose as monetary stimulus. The worst conceivable policy would be tight money and an easy fiscal policy. That would balloon the budget deficit, and the expansionary impact of the fiscal stimulus would be offset by the tight money. And that’s what we have.
So while I think it is perfectly defensible to advocate 0.7% inflation, it is not reasonable to defend the Fed, as by its call for fiscal stimulus it signalled a preference for more AD. Indeed the Fed has yet to call for an end to the fiscal stimulus program. I wish they would.
As to which inflation rate is best, I think it quite possible that 0.7% is better than 2% in the long run. But the time to suddenly reduce the rate of inflatiion is not in the midst of an economic crisis when loans were taken out in anticipation of higher inflation, and where wage contracts were signed with the same understanding. Indeed this is the worst possible time to undertake a policy of disinflation. If we were going to do so, it would have been much better to do so in the midst of the housing or tech booms, not in the middle of a recession.
jrshipley, Thanks for the info on epistemology. I anticipated amny people would raise the issue you mentioned, and I should have said more about it in the post. First I’d like to distinguish between actual inflation and expected inflation. Surprisingly, the Fed is much more worried about expected inflation than actual inflation. As long as an inflation blip doesn’t affect expectations, it is transitory. It wouldn’t affect wage demands, and hence it would have very little effect on the core rate (which is what the Fed really cares about.) Indeed, Fed officials often say the same thing, that what they are really scared about is inflationary expectations, not inflation itself. So it’s not a problem if actual inflation is higher than the market expects, as long as inflation expectations stay well behaved.
My second point is that inflation expectations are observable in real time. So if they start to creep up, then the Fed can tighten to keep them in line.
I certainly knew that my post did not represent an airtight logical proof that the Fed was wrong. I understand that they may see think differently, and that they might think inflation expectations could get out of control before they had a chance to respond. I don’t see that fear as at all plausible. So while I think the Fed is thinking about the entire issue in the wrong way, I understand that it is a difference of interpretation, not a simple case of bad logic. But I still think the exercise was useful, it forces the Fed (if they read it) to ask themselves: “Are we smarter than the market?” Perhaps they are, but over the past two years the market has been consistently ahead of the Fed, the market has made the better forecasts.
David, You said;
“So, if you and Tim Duy are correct, then each sign of the economy stalling should be seen as a prelude to the Fed’s determination to further* raise inflation expectations.”
I can’t speak for Tim, but I favor a NGDP target. So it depends what you mean by “stalling.” The real economy was stalling in 1974, but NGDP grwoth was high, and so we didn’t need easy money.
My views don’t have any investment implications, as I believe markets are pretty efficient, and will price in the Fed’s expected response.
I don’t follow your statement about creating 5% inflation. As you know I favor NGDP targets. But let’s say the Fed doesn’t agree with me, and targets inflation. I would favor a much lower rate, something like 2% long term. If expected inflation was negative 3%, I’d try to raise it to 2%. In the post I only mentioned 2-3% because under the much more popular “flexible infaltion target” approach you have slightly above 2% inflation when in a recession, and slightly below 2% inflation when in a boom. In other words a countercylical inflation rate.
Jon, I’m not sure I follow. They could easily reduce the base right now, they don’t have to wait until the economy is strong. If you had no interest on reserves, you’d have to sharply reduce the base, even more so with a slight negative penalty on excess reserves. If you are worried about bank profits you could keep the rate positive on required reserves. I see the Fed having all sorts of options that they aren’t using. If the Fed loses money on its balance sheet, because it bought some long term bonds at low yields, that reflects its own stupidity. it has INSIDE INFORMATION, it ought to know its own plans for future inflation. I’m being sarcastic of course, I do believe they are that incompetent.
Bonnie, I’m not sure it would have much impact. We don’t pay cash for oil, so it doesn’t have much impact on the demand for base money, which is the sort of money directly controlled by the Fed. Base money is now mostly cash in people’s pockets, or excess bank reserves. I don’t see any impact on the demand for either asset being affected much by an oil pricing decision. Any effect might be more psychological, a confidence issue.
jb, OK, but if I wanted to be reductive I could just say “All you are describing is changing subjective probabilities.” So if we applied it to the current issue, and I wanted to make your best case, I’d say “The Fed is trying to reduce the variance of inflation expectations, make it more certain, even at the cost of leaving the AVERAGE expectation lower than desired.” If that’s the implication, it is certainly defensible. In the end I still favor higher inflation expectations, as I don’t think the uncertainty argument is strong enough given the severity of this recession. But I think I see your point.
6. October 2009 at 11:28
Marcus, Yes, they are clearly trying to reduce inflation expectations. The question is why? Do they worry that inflation expectations might rise faster than they can control them? That would be bad reasoning in my view. (see my earlier responses.) Or do they see something I am missing?
David#2, I do not think RGDP and NGDP move in lockstep. All sorts or real factors influence the gap. Indeed even nominal shocks influence the gap, as David Hume showed. I think the two indicators I cited in the post are a much better way of looking at inflation expectations. Real gold prices are affected the the boom in India and China.
Bill, You said;
“I don’t agree that interest bearing reserves are the same as T-bills. I don’t agree that money cannot bear interest.
Money is the medium of exchange. It is still the medium of exchange if it bears interest. T-bills aren’t the medium of exchange.”
I have two points. First, bank reserves, especailly excess reserves, aren’t really a medium of exchange. I will admit that there is a derived demand for RR as a result of the demand for checking balances. But the ER number is meaningless. The are just held as assets, like T-bills. the ERs aren’t circulating, and they aren’t even backing DDs.
Second, if you pay interest on money, the QTM breaks down. You can no longer assume if M doubles, P doubles in the long run. Suppose the Fed doubled M, but paid a higher interest rate on M than alternative assets, that would actually be contractionary.
Yes, in a sense you are right, the ERs can be costly converted into a medium of exchange (currency or DDs). But my point wasn’t technical, it was that it’s as if they were like T-bills. If you pay interest on them at a rate higher than on 3 month T-bills, then they don’t have the normal inflationary impact predicted in models. And that’s even true in the long run. Suppose in five years the T-bill yield is 4.5%, but reserves earn 5%; banks will still be hoarding ERs. There are other countries that run monetary policy this way, and they don’t have high inflation.
Lee, It depends, the market expects the Fed to do what is necessary to keep inflation low. If demand drops off the market expects the Fed to reduce M. But it also knows the Fed can pay higher interest rates on reserves (which is why velocity fell in the first place.) Other countries have done this.
Rebeleconomist;
1. Yes, you are right. But I don’t think the CPI biases are a big issue now. As you know I look at NGDP growth, and it shows an even bigger deflationary shock than the CPI.
2. It is possible to believe in the importance of M, and in the monetarist transmission mechanism, and also believe that the Fed is acting in such a way that M is endogenous. They adjust M to hit their policy goals. But only to some extent, and not recently. This is where Michael’s criticism has such force, it would be OK if M bounced around if expected NGDP was stable. But both are highly unstable right now, so we are flying blind.
Alex, Bob Murphy had the same argument. But the analogy doesn’t hold, as pilots know more about how steering affects the plane than passengers. But the market knows much more about how monetary policy affects inflation than the Fed does (as has been repeatedly shown recently.) They should look to the market for guidance, but pilots should not look to passengers for guidance.
Dan, You said;
“While I would suggest that the empirical correlations that these observations are based on include only limited time-frames (and actually went negative last year),”
You may be right about their motivation, but I’ll take the TIPS anyday. And the negative correlation that you allude to from last year (when commodities screamed that tight money was needed) turned out to be pretty consequential.
6. October 2009 at 12:34
Scott,
My point is that right now we can only say:
“Markets do not expect low inflation”
But why do markets expect low inflation? Is it because the current policy is not inflationary or is it because they expect the Hawks to kill the Doves? How can you tell?
Alex.
PS: What do you think about Brazil and China buying SDR´s?
6. October 2009 at 12:44
Sorry the previous post should have read
“Markets expected low inflation”
6. October 2009 at 16:04
Scott,
With regard to inflation targeting, you contradict yourself.
Above, you state that a .7% target might be desirable, but that you would prefer a higher one due to the fact that we are in a recession. Later, you state that the choice of target should be independent of the depth of deflationary expectations.
You might argue that you were referring to reducing the inflation target, rather than setting one. Perhaps, but I think the concept is still valid: the more the public desires to hoard cash in expectation of falling prices, the more disincentive they need to reverse that preference.
Take this example: say investors fear a 5% p.a. drop in the price level, and choose to hoard cash. The Fed promises to create 2% inflation. Investors will assign some probability to that promise, and determine an expected value for future inflation. If that expected value falls below 0%, they will continue to hoard cash. Now, are you saying that if investors feared 10% p.a. deflation, that they expected value of inflation would not change? What about 20%? It seems you assume that investors will attach a 100% probability to whatever inflation target the Fed comes up with, and a zero probability of deflation no matter how deep deflationary fears run.
Its actually quite telling. It seems the Fed never has to “signal” its credibility by making explicit trade-offs. Statements such as “we will stomach high inflation before we see deflation set in,” or “we will see unemployment go much higher before we accept over 2% inflation” never have to be made. I think the Fed certainly agrees with you. I doubt that the real world will afford the Fed that luxury.
6. October 2009 at 17:25
Scott,
Thanks for the response. So, when velocity picks up, the market expects the Fed to be too tight? I was under the impression that the Fed would find it both practically and politically difficult to be very contractionary, but then I am not the market.
6. October 2009 at 19:14
Its precisely because the base is hard to unwind that gives it its potent inflation potential. The term of the Fed’s balance sheet has shifted dramatically. As the TAF+SWAPs unwound, the Fed replaced them with lots of 5+10 yr paper.
Some months ago I speculated that the Fed’s base expansion was not worrisome because the illiquid portion of the base remained smaller than the putative ‘normal’ level. This is now not the case. If draining the base were easy, the Fed wouldn’t be discussing unorthodox alternatives such reverse repos.
6. October 2009 at 19:18
Scott:
On thinking about it, I have really quibble with your TIPS numbers. The 5yr treasury market is significant distorted by open markets desk right now. What premium is the Fed paying for its QE? 100 basis points is easy to believe. In which case, the 5yr inflation would be 2.4%. Which in turn suggests a market expectation of 3.7% inflation on average after the first two years.
7. October 2009 at 05:41
As Tim Duy points out, after the last recession we faced a “dearth of investment opportunities” (I actually don’t agree with that statement, but let’s assume its true) and rather than invest in productive assets, we invested in houses. As he also points out, investing in houses is actually more consumption than investment. So, is it not correct that rather than a lack of consumption, which actually hasn’t dropped all that much in this recession (with the exception of housing), the problem we face is a lack of investment? If that is true, how exactly do we expect to increase investment with the dollar falling on a daily basis? We can see from the flow of funds report that private investment is flowing out of the US right now. So whether you blame it on Fed policy or fiscal policy, what we are doing right now is making our situation worse. So which is it? Monetary policy or fiscal policy? If the problem is fiscal policy, I am skeptical that we can fix it with monetary policy. Attempts to inflate will just make the problem worse by further devaluing the dollar and increasing the flow of investment out of the country.
So, Scott, how do we increase investment in the US? Is there any policy the Fed can pursue which will accomplish that real goal?
7. October 2009 at 06:26
Joe Calhoun,
There are problems with definitions in this regard. One possibility is to consider consumption to be “consumer goods”, this is not quite correct though as consumers may invest directly. I buy a car to use many times, people bought houses for the same reason. So, many instead define consumption as something that is used up when use by consumers. Then we can talk about “consumption goods” as those bought for fairly much immediate consumption – even this has problems though.
Now, it doesn’t really matter which we use as long as we’re consistent and aware of this problem when reading other articles.
The problem in places with overbuilt housing is that the exchange-value and the use-value of houses has fallen. The exchange-value part is simple, prices of houses have fallen. The use-value part is because many houses no longer fulfill the place in people plans they once did. People want to spend their money on more urgent needs than large houses and they want to live in places other than where the large houses are.
This is what we really have, a drop in the value of Capital. And of course all the other macroeconomic fall out that has created.
7. October 2009 at 06:27
Jon said: “What premium is the Fed paying for its QE? 100 basis points is easy to believe. In which case, the 5yr inflation would be 2.4%. Which in turn suggests a market expectation of 3.7% inflation on average after the first two years.”
If the Fed has had a 100 bps impact on 5yr Treasuries, wouldn’t that also show up in the TIPS? It sounds like you are arguing the Fed isn’t actually influencing prices/expectations with QE but merely passing on subsidies to random sellers?
7. October 2009 at 16:35
Thruth: why would the Fed necessarily have the same impact in TIPS? It is as you say, the Fed is driving down the price of particular things. The Fed’s balance sheet is simply too small to move the risk-free rate of the entire 5yr maturity market even if they succeed at moving particular ‘risk-free’ components.
7. October 2009 at 17:04
Current,
Thanks and I agree that defining consumption goods is not as cut and dried as it seems, but when I say productive assets, I mean assets which are used to produce other goods or more intellectual assets. The question is the same though, if the dollar is falling and private capital is flowing out of the US, how will we be able to make up this investment deficit? It seems to me that a monetary policy that pursues a higher inflation target or a higher NGDP target at this point is counterproductive because it causes a further drop in the value of the dollar.
I would also add that to me the drop in the dollar starting in 2002 had an effect on investment after the last recession. The dollar started falling in 2002 and the only reason gross private domestic investment rose was due to “investment” in real estate, primarily residential real estate. If you look at new orders for non defense capital goods excluding aircraft (a good proxy for capital spending) it dropped from 7.5% of GDP in 2000 to 5% of GDP in 2001 and never recovered. Why did people prefer to invest in real assets (real estate and commodities) while the dollar was falling? Was it a lack of other opportunities or was it a function of a falling dollar? It seems a very rational investment decision to invest in hard assets if the dollar is falling and while I don’t think most investors thought of it in those terms, that is what happened.
If the dollar continues to fall, investors will continue to choose hard assets over more intellectual assets. The capital that today is flowing to gold, oil and other commodities does nothing to aid our economic recovery. If we pursue a monetary policy and a fiscal policy that continues to weaken the dollar, recovery will be a long time coming.
So the question for Scott is the same; how do we increase investment in the US and is there any way for monetary policy to assist in this effort?
7. October 2009 at 17:26
Some comments:
1) I can’t find the data right now, but I recall a recent survey of economists showing a high level of polarization on inflation expectations – with many more economists at the extremes than is typical.
2) This is reflected in current market activity and volatility – on the one hand, we have Bill Gross and the Bond Market declaring that deflation is now the bigger threat. On the other hand, we have a massive anti-dollar carry trade (led by the likes of Peter Schiff). In an almost Hegelian manner, the “consensus” has declared deflation now but Big Inflation later. The Fed has bought into it.
3) The ultimate driver of Big Inflation expectations is the Federal Debt. The question is just how unsustainable is it? Most of Western Europe has a larger govt. debt/gdp ratio. But, they also have a higher savings rate (which funds the debt) thus sustaining their currencies. Of course, our savings rate is now increasing (total private debt just dropped another 12 billion last month…)
One quite plausible future is that the Fed’s near-term tight policies lowersw AD expectations, thus dropping business investment and household consumption, but that government (through automatic and discretionary spending) compensates for this through spending. However, the spending may be sustained by the higher domestic savings rate. In the long run this is unsustainable, but that could be longer than many expect (judging by European govt debt/gdp levels).
In the short run, however, the polarization in expectations may have left us in a tipping point situation – a smackdown between the Bond Market and the Carry Trade. Rather than respond in a credible, unified manner, the Fed has betrayed an unctious fragmentation of ideological opinions and no discipline in reigning in disparate viewpoints.
Worth reading:
http://www.alongthemargin.com/archives/stephanie-pomboy-immediate-threat-is-deflation-not-inflation/
It isn’t just the economists that are polarized; it’s reflected in an implied divergence between the stock market and the bond market. The notion of asset price inflation and consumer price deflation seems very odd, and yet that is where the markets seem to be going – presumably driven the deflation now, inflation in a few years narrative.
7. October 2009 at 17:40
Quick note about inflation expectations:
Theoretically, there should be an arbitrage opportunity between TIPs and futures. I recall a similar issue coming up in CDS contracts here:
http://baselinescenario.com/2009/04/06/inflation-prospects-in-an-emerging-market-like-the-us/#comment-9390
One possibility (mentioned) was a nominal floor skewing the TIPS projections, but one can imagine relative market liquidity also being a factor. An average is probably a reasonably thing to do.
8. October 2009 at 02:50
I see what you mean, but I don’t really agree. Certainly a house doesn’t provide a tangible asset, like a car production line does. But it does provide a useful service.
Compare it to a warehouse for example. The warehouse provides the service of protecting goods from the elements. Surely a warehouse is a productive asset? Because of this service it renders a rent can be charged for it. Now, why is a house any different?
I know what you mean here, a house cannot be used as a means of production, unless it’s converted. But, it’s still a form of investment. It’s a consumer capital good that provides a service to the consumer over time.
(One of the odd things about recent events is that they have made it much more attractive for those of us in Northern Europe to move to the US. House prices in the UK and Ireland and very high, but they are fairly high in other European countries too.)
I agree in general. However, this doesn’t argue against monetary equilibrium, that is stabilizing NGDP or stabilizing some sort of index of overall purchasing power. Only against price rises above that.
I agree with you that the comparison between Europe and the US may not be favourable to Europe. I don’t like this business that many commentators use of considering the “savings rate” as an important indicator. Savers can decide to change their savings behaviour quite fast, so I don’t think trends in the saving rate are that important.
The problem here is uncertainty about the future. We all know that the debts must be maintained or partly paid. That must be done by taxation or by seignourage. The important questions are how? and who shall pay?
This is how I look at it for my own finances. I must find out if I will be one of the sectors of society taxed to pay of the debts or maintain them. This is what each of us must do, and I think this is where the problem lies. If the various groups in society correct assess that they are going to be taxed in the future then they will prepare for that. So, the payment of the debt will be spread out and cause no major crisis.
However, if the groups I mention plan badly this may not happen. If every part of society thinks “they will never dare tax me” then when tax rates have to rise it will come as a surprise.
8. October 2009 at 05:50
Alex, The reason the markets expect low inflation is because they are worried the hawks have too much influence. So yes they make forecasts conditional on expectations of Fed policy, which is precisely the problem–they don’t like what they see (and expect). The way to fix that problem is for the hawks to stop talking, and start listening to the doves. The doves have been right so far, and the market still thinks they are right.
I don’t know much about SDRs, but it is probably a minor issue.
David, You are mixing up two issues. One is the long run decision of what sort of NGDP or CPI rate to target. The other is what do we do right now. We have implicitly adopted a roughly 5% NGDP target, or perhaps a roughly 2% inflation target. My point is we should keep aiming for that target, regardless of the current shocks hitting the economy.
If in the long run we decide to pick an entirely different target, such as 0.7% inflation, we can do so. But it should not be in reaction to any deflationary shocks. If we had had a 0.7% inflation target all along, then we should have stuck to that one in this recession. So there is no contradiction, a recession is a poor time to change long run strategy.
You said;
“The Fed promises to create 2% inflation. Investors will assign some probability to that promise, and determine an expected value for future inflation. If that expected value falls below 0%, they will continue to hoard cash.”
This has things backward. I think the Fed should target market expectations. So if the Fed has a 2% inflation target, they should increase the monetary base until the public expects 2% inflation. The system you describe is the one Svensson favors.
You said;
“It seems you assume that investors will attach a 100% probability to whatever inflation target the Fed comes up with, and a zero probability of deflation no matter how deep deflationary fears run.”
I don’t assume this at all, I just assume the market expectation equals the the Fed’s target. There is a huge difference between a market expectation, and being certain that a particular inflation rate will actually be achieved.
Lee, You said;
“I was under the impression that the Fed would find it both practically and politically difficult to be very contractionary, but then I am not the market.”
Bernanke was just reappointed, despite deflationary policies. I think most of the pressure is to raise rates. There is a constant drumbeat of criticism of 2004, when a policy of low rates supposedly led to the housing bubble. The Fed will be under tremendous pressure not to hold rates low for too long.
If we get a robust recovery in NGDP, there will be little criticism when the Fed actually does raise rates, just as there was little criticism in 2004-06.
Jon, I agree it is not easy, but I don’t see where a few capital losses from selling 5 and 10 year bonds would be that big a deal. We are already throwing $100s of billions out the window, what’s a few billion more?
I’m sure you are right on one level, the Fed is less cavalier than I am, and they do worry about it. So I suppose they will keep the interest on reserves program going forward. But losses are losses, whether they are realized or not. Hopefully the Fed isn’t too irrational about these things.
Jon#2, I don’t believe it would be anywhere near 100 basis points. But even so, I also provided the CPI futures number, which was 1.71%, and should overstate inflation expectations.
But I also think people need to step back and think about the implications of their inflation arguments. If you are right, then it would actually be a bad thing if the fiscal stimulus “worked.” If inflation expectations are too high, we should be rooting for fiscal stimulus to fail, rooting for it to not boost AD. I’ve never heard anyone make that argument, even any of the conservatives opposed to fiscal stimulus. Indeed they generally argue the opposite, that it would fail to boost AD, and then imply that would be a bad thing.
I suppose all this sounds off topic. But I have been perplexed by this asymmetry since I began this blog. Why are so many people worried about inflation, but very few are worried that fiscal stimulus will reduce unemployment from 10%? Worrying about one but not the other is entirely irrational. You may be rational, but I just want to spell out the implications of the fear that inflation will be too high.
Joe, You said;
“As Tim Duy points out, after the last recession we faced a “dearth of investment opportunities” (I actually don’t agree with that statement, but let’s assume its true) and rather than invest in productive assets, we invested in houses. As he also points out, investing in houses is actually more consumption than investment.”
I don’t agree with either point. Business investment went way overboard in 1996-2000, so there were very few productive opportunities left in 2002. So housing was a much more productive use of capital than building more factories or stores or office building. Of course later on the housing boom also became overextended, but that wasn’t due to monetary policy, it was due to foolish decisions by lenders. You can’t legislate against stupidity. The Fed should focus on aggregate demand, and let the market decide where to allocate those resources.
And housing certainly is an investment good, indeed far more so than most business investment, much of which depreciates far faster than housing. The extra housing that was built will mostly last for 100 years or more. After 3 or 4 years of being empty, population growth will fill most of them up (except the few that were torn down in the Southwest) and people will enjoy another 96 or 97 years of living in those supposedly wasteful investments. Yes, they were wasteful, but only because they were built a few years too soon.
Right now investment is too low because the economy is weak. If we boost AD, then investment will pick up.
Current, I agree.
Thruth and Jon, Can I assume that the Fed bought lots of conventional 5 years, but little or no indexed 5 years?
Joe#2, I would point out that many feel that there was overinvestment in equipment around 2000. But actually I do agree with your view that we should invest more. My solution is a Singapore-style forced saving plan, to force Americans to save much more. I would offset this in two ways. Sharply lower taxes, especially on capital. This could be financed by cutting entitlements for the middle class. The forced saving program would replace many entitlements. And second, wages would rise as companies would reduce health care coverage, only providing catastrophic coverage. The other health bills would be paid out of personal health savings accounts that everyone would be forced to hold. No more fighting with insurance companies over small and medium expenses, just the huge ones.
This should raise investment long term, and reduce consumption, especially consumption on health care, which is at an unhealthy 16% of GDP
Statsguy, I agree with the polarization argument, I saw the same survey. And I agree this may be influencing the Fed, but consider the irony here:
“The ultimate driver of Big Inflation expectations is the Federal Debt.”
So the TIPS market currently expects low inflation, but the Fed second guesses. OK, but why? Because they think the national debt will have to be monetized. OK, but who would do that? So the Fed is second guessing markets because they know better than markets just how irresponsible they are likely to be.
On a more serious level, I don’t worry at all about inflation, I only worry about inflation (or deflation) expectations. Transitory inflation doesn’t work its way into wages and core inflation. And I think the observability of TIPS spreads in real time makes anything over about 3% expected inflation unthinkable. So my only fear is intentional inflation, monetizing the debt. I don’t think it likely, but it is possible.
BTW, I know you didn’t make that specific argument, but it seemed sort of implicit in your collection of entirely valid individual points.
You said;
“Rather than respond in a credible, unified manner, the Fed has betrayed an unctious fragmentation of ideological opinions and no discipline in reigning in disparate viewpoints.”
This is a great point. I just spoke with someone at the Boston Fed who also thought money was too tight last year. It made me realize just how diverse the Fed is. I often (wrongly) treat it as a monolithic institution in this blog. But what your observation really points to is the need for an explicit target, so that there are no more hawks and doves, just people with differences of opinion about how to get X% inflation, or X% NGDP growth.
Statsguy#2, Thanks for your two links, I will look at them. That’s a good point about arbitrage, I had never thought of that.
Is the nominal floor on TIPS that you refer to the fact that the total inflation adjustment on TIPS can’t fall below zero? If so, we can avoid that problem (for all practical purposes) by using longer term TIPS with just two and five years remaining.
Current, I am actually not too worried about higher rates affecting me, if I was working class I’d be very worried. To raise European levels of revenue for a welfare state, you need huge payroll and consumption taxes. I retire in 8 years, and am a high saver, not a spender.
But maybe I’ll be one of those who is unpleasantly surprised by what’s coming down the road.
8. October 2009 at 06:48
Well, I already have those tax levels. I’ll probably retire in ~2043 or earlier if I can. What worries me though is that they will get worse, because of the need to maintain or pay off deficits.
The problem is how to account for that? It’s obviously necessary to save a great deal. What isn’t clear is where to put the money, or the important question of which country to retire in.
On a macro level what worries me in Europe is that certain sorts of savers will probably lose and others gain. Those who allow themselves the option of moving country will be able to avoid the worst of consumption taxes. But, what effect of the finances of those countries will that have?
8. October 2009 at 11:19
this is slightly off topic but could someone explain why it makes any difference if oil is priced in dollars vs. any other currency. isn’t the dollar pricing merely an index? i can understand that if saudis preferred to hold yen vs. dollars that that would affect echange rates, but how does the mere dollar denomination affect exchange rates? Is the denominal problem real?
8. October 2009 at 11:32
Scott you say:
I think its a question of short-term, long-term thinking. I am worried that policy is insufficiently stimulating because I agree that there is a lingering employment problem. But I am also concerned about the inflation potential out years ahead–I think the Fed will be very reluctant to realize losses.
I’d prefer the Fed to have a smaller balance sheet but more aggressive policy overall (i.e., interest on ER). I believe such a policy would be more stable.
One point about inflation targets, the TIPS spread reflects expectations of the CPI report not expectations of inflation per se. Again the 1991 methodology boosts CPI by a small amount over the current BLS numbers–it puts current inflation at 2%! We know that the Fed does not believe in commodity indexing, but do they believe the current BLS CPI? One way to render Fed policy rational is to believe that they do not, that they prefer the older methodology.
8. October 2009 at 11:43
I’m very sceptical of this idea too. The currency used in a spot market is not very relevant for wider macroeconomic effects. What matters is the currency in which future contracts are made.
As far as I can see any effect comes from bonds bought with oil money or oil contracts.
8. October 2009 at 11:43
Scott,
I agree with most everything you say about how to address the lack of savings and investment and if either of us was dictator we could get that done. However, since that isn’t the case, my point is that in the absence of such policies, there is a limit to what can be accomplished with monetary policy.
I think a lot of our current difficulties can be traced to the decline of the dollar that started in 2002. Whether that was the fault of monetary policy or other policies of the Treasury is debatable I suppose, but at least some of it was due to faulty monetary policy. If the dollar was overvalued in 2000/2001, it was certainly undervalued by 2006 and I think it should have been stabilized around 2005. If we had done that, I think the worst part of the housing bubble would not have happened. I think it would have been hard for the Fed to have stabilized the dollar without the support of the Treasury but surely they should have seen it as an issue. I don’t remember Greenspan or Bernanke ever mentioning the dollar during that time or even much since then.
The value of the dollar is to me the most important indicator of overall economic policy. An overvalued dollar produces problems as we saw in the late 90s and an undervalued dollar produces problems as we’ve seen lately. If we don’t get policies, monetary and otherwise, that stablize the dollar soon (say around the levels of last summer) we are going to have more serious problems than we’ve just seen. We can’t devalue our way to prosperity.
8. October 2009 at 14:47
Hey rob. Well, ideally, won’t really make a difference, as money is fungible so it doesn’t matter if people use dollars or yen or gold brick certificates.
Here is the story though:
However, in part, the dollar’s value, like the value of anything else comes from the demand for dollars. At the moment since oil is sold in dollars, demand for oil means that people want to buy dollars to get the oil, and want to hold dollars because its useful for buying oil. This makes the value of the dollar higher. A higher value dollar means we can deficit spend and how really low interest rates without causing too much problem for ourselves. Its like free money given to the american people/government.
That is the story anyway, not sure if its true or not.
9. October 2009 at 01:29
Doc Merlin,
You are right about normal transactions that go on in dispersed markets. A banknote is a loan to the issuer. But, there’s more to it than that.
Let’s say we have a dispersed market with long-term contracts in dollars. In this case the currency has two effects, firstly, as you say, those trading are loaning the issuer money. Since we are talking about current account money they are loaning a commercial bank money. Secondly, since the contracts are denominated in dollars and long-term then fluctuations in the price of a dollar will redistribute between both parties.
If we are talking about short-term contracts though then fluctuations in the value of the dollar don’t make any difference. If I agree to pay you in 30 days time then neither of us will lose or gain much from the change in the dollar over that period. Also, if the market is centralized then little currency is actually needed. The various market participants often operate by a system of cancellation. Rather than transfer bank account money for each transaction they run up debts against each other. They then pay them on a regular date, most of them being paid by canceling debts against each other and the rest being done using money.
So, the currency used by the spot markets really isn’t so important. What is important is that used by the longer-term contracts I mention and those used in more dispersed markets.
10. October 2009 at 10:04
Current, Those are good points. I am currently in a very high tax situation because I pay a brutal marriage tax penalty. I pointed out that if we divorced and started “living in sin” we could cut our taxes by $5000 a year. But you know how sentimental wives are.
Still it’s good to know the Feds are doing the best they can to discourage people from getting married. (Yes, I’m being sarcastic.)
rob, I’ve always wondered the same thing. Especially now that oil prices bounce around in the world market. Back in the 1970s, I seem to recall that OPEC had more price fixing power.
jon, You said;
“I think its a question of short-term, long-term thinking. I am worried that policy is insufficiently stimulating because I agree that there is a lingering employment problem. But I am also concerned about the inflation potential out years ahead-I think the Fed will be very reluctant to realize losses.”
Fair enough, but shouldn’t this call for people to recommend a more expansionary monetary policy now, and then a less expansionary monetary policy later.
I don’t know about other people, but when I read articles it’s almost like people assume monetary policy affects inflation and fiscal policy affects real growth. Of course both affect AD.
Regarding your last point about the CPI, it is also a fair point, but if that’s the Fed view then they should logically believe the economy faces no AD problem, and also that fiscal attempts to lower unemployment are misguided. Do they think that? Maybe, but it would be an incredible bombshell if they admitted it. I certainly think the economy needs more AD. As do most economists and most politicians and most of the public.
I’ll do the rest later.
11. October 2009 at 05:45
Joe, You said;
“I agree with most everything you say about how to address the lack of savings and investment and if either of us was dictator we could get that done. However, since that isn’t the case, my point is that in the absence of such policies, there is a limit to what can be accomplished with monetary policy.”
I completely agree, monetary policy can only address issues of aggregate demand, which is why I favor targeting NGDP, not RGDP.
I do not agree that monetary policy should focus on the exchange rate. We tried that in 1929-33 and it led to the Great Depression. If NGDP growth expectations are telling you to go one way, and the exchange rate is saying the opposite (like right now) always go with NGDP expectations. Right now the dollar is weak, which would normally call for tighter money, but we need much easier money as NGDP growth expectations are also weak.
I believe if they target NGDP growth the dollar will always be at its proper level. It may fluctuate dramtically at times, but those fluctuations will help stablize the economy.
Doc Merlin, In theory your story is true, but the amount of base money used in oil purchases is so trivial that it doesn’t make any practical difference.
Current, I agree.
11. October 2009 at 12:29
Scott,
I am not suggesting that merely stabilizing the value of the dollar is sufficient to ensure economic growth. The value of the dollar is not just a function of monetary policy. My point is that if we pursue loose monetary policy in the absence of other policies that would support the value of the dollar, the effect will be mostly inflationary. I’m sure you are right that we can create a rise in NGDP with monetary policy, but should that be the goal if accomplishing it causes a steep drop in the purchasing power of the dollar? Surely there is a point where such a monetary policy is causing more harm than good.
14. October 2009 at 15:38
Joe, Yes, we should try to have low but steady NGDP growth even if it means a steep fall in the value of the currency. The Aussies did that last year, and the AUS$ fell sharply. They are the only developed country to miss this recession. And they also missed the 2001 recession. Meaning their last recession was in 1991. The AUS$ is not very stable, but their economy is very stable.
Admittedly their economic structure is different, they are more commodity-based than we are. But commodities were hit hard late last year, so their avoidance of recession is impressive.