Back to aggregate demand, which has been the #1 problem all along
Long time commenter David Pearson recently made a couple of excellent points in the comment section of this post:
The epitaph for “Credit Easing”?
“Although financial markets are for the most part functioning normally now, a concerted policy effort has so far not produced an economic recovery of sufficient vigor to significantly reduce the high level of unemployment.” (from yesterday’s Bernanke Princeton speech)
Then he made this observation:
Also, Bernanke argues-correctly, IMO-that “Credit Easing” already succeeded in moving funds out of safe-haven assets. This was a boon to corporate debt issuance, but unfortunately, it did very little for corporate investment (as companies simply used the proceeds to refinance existing debt or build up cash levels):
“Mr. Bernanke noted that by withdrawing supply of mortgage bonds on the market, the Fed was able to get investors wanting to hold safe instruments to move into higher-quality corporate bonds, thus lowering yields there.”
This would seem to support the view that actors have no problem moving out the risk curve. There are plenty of opportunities for companies faced with attractive investment projects to access funds from creditors and invest them. For goodness sake, Petrobras just placed a $70b equity offering! Yet some continue to argue that the issue is “demand for safe haven assets”-some sort of extreme risk aversion.
“Credit Easing” did not try to directly stimulate AD; it tried instead to produce a working monetary transmission channel. Bernanke’s work on the GD supports the idea that he would be extraordinarily concerned with getting this channel to function. Now it seems that he admits that it is functioning, and therefore he turns to the problem of a shortfall in AD.
Which has been the #1 problem all along. Of course there really was a severe banking crisis in late 2008, and I have an agnostic position on how that should have been addressed. But it’s not an either or question, the Fed needed to address AD, regardless of what they did for banks.
Just to head off any “can’t walk and chew gum at the same time” arguments, recall that there is no such thing as “not using monetary policy.” The Fed always can and does set a monetary policy stance. When the FOMC met in September 2008, it would have been no more difficult to set the fed funds rate at 1%, as compared to leaving it unchanged at 2% (which is what they did.) And in October it would have required even less effort to not pay interest on reserves, than to start up the IOR program.
I only met Mr. Bernanke once on my life, when he presented a seminar at Bentley discussing his theory that credit problems were a major factor depressing output in the early 1930s, even apart from the monetary shock. I asked him just one question; how bad would the Great Depression have been if the Fed had targeted steady NGDP growth, but the bank failures had occurred anyway? As I recall, he hemmed a hawed a bit in answering the question. I still don’t think the Fed has a good answer to my question from way back in the early 1990s.
A few of us were vigorously arguing for much more monetary stimulus in October 2008. The Fed wasn’t paying attention, nor were 99% of macroeconomists. I still haven’t heard a good explanation for why that happened. Why are we only now considering more stimulus? Was there no AD problem in late 2008, when NGDP was falling fast? And don’t say the Fed didn’t realize we had an AD problem—their own forecasts called for nominal growth rates far below any implicit Fed target.
Tags:
26. September 2010 at 09:43
The zero-bound meme caught hold and stuck, and it’s taken a while to dispel.
Arguing for monetary action has been a traditionally right-wing concern, but the entire post-Friedman right seems to have fled toward some kind of classicism. And concern about aggregate demand is traditionally a left-wing concern, but enthusiasm for monetary action there has never been very high.
26. September 2010 at 09:59
David, That’s as good an explanation as I have seen. But if correct, it calls for some soul-searching, oops, re-evaluation among both the left and the right. I hope you will also look at my newest post on the advantages of NGDP targeting over inflation targeting. If we combine these two posts with your argument, we are left with the following:
For 25 years we had a truce, as the Fed kept AD growing at a slow and steady rate. Both moderate liberals and moderate conservatives seemed fine with that. Then the left suddenly forgot that it is the Fed’s responsibility to keep AD growing, and the right suddenly forgot that a healthy economy requires fairly stable growth in AD.
26. September 2010 at 10:15
Yeah, the ‘truce’ reading is how I read it, too.
(a more cynical reading would be that neither side was particularly displeased at the prospect of powerless monetary policy; the left goes “great! Time to pull all the fiscal levers like we’ve always wanted” and the right goes “great! Time to hammer away at restrictions on the supply side like we’ve always wanted”. Which gives us the narrative we have now.)
As for NGDP targeting, you may have to wait until the next crisis at earliest. Changing tacks during a crisis isn’t politically likely or appealing for expectations reasons. And there’s no difference when RGDP is increasing smoothly (and NGDP targeting and inflation targeting are essentially similar). Only during the next real shock will the Fed, optimistically, adopt another policy. At earliest.
In the meanwhile, change the policymaking zeitgeist, I guess… good luck at that. You’ve shook the blogging world, but the journals might be a little more stodgy. 😛
26. September 2010 at 10:40
Seems to me the right-wing should embrace the monetary bulls. If QE works, it undercuts future arguments for deficit spending.
The right-wing, somewhere along the line, blindly adopted a “tight money and gold” fetish. There is plenty of chatter about gold in right-wing econ sites. Rising gold prices are seen as proof of Obama’s failure, a weak Fed, or the the general decadence of the United States.
And currently, I suspect some on the right-wing simply want Obama to fail, and starving the economy of money will get us there. (Richard Fisher?).
26. September 2010 at 10:47
David, Once again, I can’t really disagree with anything you say. I don’t think I’ll even try the journals route. Been there, done that . . . nobody paid any attention.
A younger spokesman for NGDP targeting will have to carry the torch in the journals, I turn 55 in a few days. I feel blogging is where I’ll have the greatest influence. Not influencing the Fed, but perhaps slightly influencing those who might have an influence on the Fed.
Benjamin, Yes, the lack of QE opened the door for much bigger government. And not just the stimulus, but bailouts of GM, more regulation, etc.
26. September 2010 at 11:34
I find 100% compelling the argument that the Fed is doing stealth monetization of the debt.
The Fed took in the toxic assets as collateral and filled the banks reserves, and in return the banks are buying the Treasuries.
Ultimately though, currency, credit and debt are BS, what matters is who owns the hard assets… and right now the wrong people own the hard assets.
Until Fed policy accepts that a decent percentage of ownership needs to change, and people with cash get to be owners and people with debt, get to lose ownership…. we’re all just spinning our wheels.
It would much more compelling if you framed your QE goals in terms of putting new money in the hands of guys with cash, and screwing down the guys desperately sitting on underwater assets, hoping for some relief.
Our sense of justice demands that the music stops and the right guys get to sit in the chairs.
26. September 2010 at 12:24
Not cutting rates to 1% in September 2008 was a mistake. But there was a much more important mistake – according to 3 month LIBOR, in late September markets were expecting effective fed funds rate that was much higher than the 2% set by the FOMC.
IOR program in October 2008 was not a mistake. Eventually this program was successful in lowering market expectations of effective fed funds rate to the levels that were set by the FOMC. I know it was also possible to lower market expectations of effective fed funds rate without IOR program, but with IOR, the stability of monetary transmission is greater. And when FOMC sets fed funds rate to zero, IOR will also fall to zero, so if you think that easier monetary conditions are needed, you should blame FOMC, not IOR.
26. September 2010 at 12:31
“how bad would the Great Depression have been if the Fed had targeted steady NGDP growth, but the bank failures had occurred anyway? ”
It all depends on bankruptcy costs. There are industries where bankruptcy costs are very low (commercial real estate sector is a good example), and a wave of bankruptcies does not cause a large AS shock.
I believe that bankruptcy costs of financial institutions were very high in Great Depression, and bank failures would have caused a large AS shock. An efficient resolution regime is needed for all financial institutions, this way failed financial institutions can reopen their doors the next day, their stakeholders should take appropriate haircuts that are needed to restore capital ratios, this way AS shock can be avoided.
26. September 2010 at 13:40
Scott,
I’m honored for the mention. In your post, you ask why Bernanke wouldn’t chew gum and walk at the same time. The following is a possible explanation.
Assume that Bernanke believes there is an explicit trade-off involved in AD stimulation: more NGDP growth now comes at a cost of future inflation uncertainty. He has said as much in his recent speeches, notably in the Q&A following the May Japan speech.
If one makes the above assumption, Bernanke’s actions are clearer. He has said, in response to an FCIC questioner, that his favorite work on the crisis is Gary Gorton’s. In his recent Princeton speech, Bernanke more or less repeated Gorton’s central thesis: the crisis was a classic “banking panic”, a run on the short term funding of the shadow banking sector.
Now, a classic run involves an information asymmetry; it is a “lemon” problem. Bernanke’s actions following the crisis are consistent with the belief that the vast majority of banking system assets–including those underlying Maiden Lane– were “money good”. “Credit Easing” was therefore intended to help banks hold these assets until the “lemon” problem dissipated.
In other words, the ’08 crisis was a particular type of market failure, one that we know how to tackle from years of research on bank runs. Here’s the key: Bernanke likely believed that the Fed had the know-how to repair this failure without resorting to un-anchoring inflation expectations. All of the actions he took and speeches he has given are consistent with this view.
Another way to look at the same question: “Recalculation” theorists and Austrians argue that the crisis was caused by mal-investment. New Keynesians/Monetarists believe it was a policy-enabled AD shortfall. Bernanke’s words and actions imply he believes it was neither. In “Essays on the GD” he wrote that the Depression was caused by a monetary policy error compounded by allowing a generalized bank run to follow. Since this time he did not, in his mind, commit the same policy error (defending the gold standard), this leaves us, again, with just a classic bank run, one that “Credit Easing” could easily address.
If the above explanation is true, it will be interesting to see how Bernanke explains the need for future AD stimulation. So far, he has told us inflation is too low because of “resource slack”. True enough, but hardly news. What has caused persistent “resource slack”? He doesn’t say. How could he, without also admitting to an important policy error? This creates a bit of a mess, especially for a rigorous thinker like Bernanke. Some other FOMC peers seem to have little need for such intellectual consistency. For Greenspan, it certainly would not have been a problem.
26. September 2010 at 14:44
@DP
Just as the Kocherlakota “unemployment is structural movie” is a replay of the same “movie” first screened in 1935 (according to Krugman: http://krugman.blogs.nytimes.com/2010/09/26/structural-unemployment-the-first-generation/
Bernanke´s “hiding” of his “policy mistake” is a replay of John Williams argument in the FOMC meeting of November 1937 that “the Fed should not reverse the decision to increase RR because that would show that the Fed had been responsible for the (1937-38) recession”!
26. September 2010 at 18:26
pearson’s take makes a lot of sense.
it amazes me how much of the economic conversation ignores the fact that we just had a financial crisis. it’s probable that if the government had been more consistent in handling the crisis that the economic collapse would not have been as deep. i’m not talking about not just the shadow banking sector, but institutions like WAMU — during the crisis, banks and non-banks were not able to access capital markets in part because of extreme uncertainty regarding government action. this uncertainly persisted until the final citigroup rescue and the stress tests, which were essentially government guarantees of the debt of large banks.
when the stress tests were published, it looked like the treasury had assumed two to three years of high yield spreads as they assumed that the banking system would be quite profitable in a pre-provision sense. so it’s possible they decided to guarantee the financial system and let it recharge knowing that investment would be depressed for several years because they believed the alternative was much worse.
27. September 2010 at 03:15
The taylor rule worked well for a couple of decades. We need to get back to it. And so, getting the relationship between the Federal Funds rate and riskier rates back to normal is a key. Focusing on the expected inflation rate (and so, the real Federal Funds rate) is a key. Noticing that money expenditures are like 15% below their previous growth path doesn’t fit into the taylor rule paradigm.
I don’t support adjusting the Federal Funds rate so that the core CPI is expected to increase 2% from where ever it happened to be.
But if making that scheme work tolerably is your goal, then targets for a growth path for money expenditures and quantitative easing are distractions.
For example, a growth path target for money expenditures allows actual inflation to vary. A 5% path would have it generally vary around 2%. Still, that the real interest rate is 2% (measured by the Federal Funds rate and the core CPI) is always 2% below the Fed’s target for the Federal Funds rate, would not apply. I think that is part of their paradigm.
The financial crisis messed up the effectiveness of the taylor rule. It is _normal_ for the Fed to say that anything that makes their particular policy approach ineffective to have _caused_ the bad macroeconomic performance.
For example, financial innovation makes it so that what were the proper settings for interest rate targets are no longer appropriate. The Fed used the settings that would have been appropriate if there had not been financial innovation. Inflation and/or production and employment results are undesirable. The Fed describes this as financial innovation _caused_ the undesirable inflation and/or output results.
The key to understanding this approach is that it is never the Fed’s fault. And since the Fed is never wrong, there is no threat that the President or Congress will compel the Fed to do something different. And, of course, it is really true that the politicians might compel the Fed to do something really, really, idiotic.
I support targeting the growth path of money expenditures.
27. September 2010 at 07:24
Morgan, More like reverse monetization, as the Fed pays higher interest on the money they inject (ERs) than the T-bills they buy back.
123, I’m not quite sure what you mean by an effective ffr. When were they expecting it to rise above 2%? And why not use the rate in the fed funds futures markets as the forecast.
I don’t understand how a higher IOR can lower interest rates.
123, You said;
“It all depends on bankruptcy costs. There are industries where bankruptcy costs are very low (commercial real estate sector is a good example), and a wave of bankruptcies does not cause a large AS shock.
I believe that bankruptcy costs of financial institutions were very high in Great Depression, and bank failures would have caused a large AS shock.”
Actually, we know this is not true, as there were a very large number of bank failures during the booming 1920s, and their effect was utterly trivial. Yes, the number increased during the early 1930s, but not enough to make a difference given the complete lack of macro effects from the 1920s failures. Worst case would have been a very small recession, and of course that’s overlooking the fact that those large bank panics in 1931-33 would never had occurred if NGDP had kept growing. So in practice, NGDP targeting is all we need.
David Pearson, You said;
“Bernanke likely believed that the Fed had the know-how to repair this failure without resorting to un-anchoring inflation expectations.”
I’m a bit confused by this, as Bernanke’s own Fed was forecasting excessively low inflation by late 2008 and early 2009. I was asking that they adopt a more expansionary policy precisely to keep inflation anchored. Any doubts about this interpretation were erased in late 2008, when Bernanke publicly endorsed fiscal stimulus. The purpose of fiscal stimulus is the same as the purpose of monetary stimulus–to boost AD, i.e. to boost inflation expectations.
That doesn’t mean you are wrong in explaining his reasoning, if you are right, then my response would be a criticism of illogical reasoning on the Fed’s part–(see my newest post on Svensson.)
I strongly agree with the last part of your comment. The Fed’s in a box, and doesn’t know how to get out w/o implicitly admitting they screwed up in 2008.
Marcus, Yes, for anyone who has studied the 1930s, the entire crisis seems like some sort of nightmarish recreation of exactly the same policy and reasoning errors that we did in the 1930s.
q, I’ve heard that argument before, and it’s possible. But recall that there is a huge cost to banks from the weak economy, as it leads to many more defaults on construction loans, etc, and that offsets some of the gains from the yield spread.
Also the “expectations theory” of the term structure says that playing the yield spread is a fool’s game. The rising yield curve is predicting rising rates, which will reduce the price of longer bonds over time.
Bill, Earlier I did a post where I argued that the Fed never gets blamed in real time–there is always some excuse. And that’s because the Fed tends to do what a broad consensus of economists thinks should be done, at any given point in time. And economists will never blame themselves.
Only future historians (and us in the minority) will blame the Fed.
27. September 2010 at 13:17
Scott, you said:
“I’m not quite sure what you mean by an effective ffr. When were they expecting it to rise above 2%? And why not use the rate in the fed funds futures markets as the forecast.”
Effective ffr is a daily volume-weighted average of rates on trades in fed funds market, it is calculated and published by NY Fed.
Fed funds futures markets suffer from survival bias – banks that fail during the period no longer affect fed funds rate after they fail, so fed funds futures understate the expected opportunity cost of reserves. If you are interested in incentives to hoard reserves, 3 month LIBOR is the place to look at.
You said:
“I don’t understand how a higher IOR can lower interest rates.”
IOR reduces hoarding of reserves. Let’s compare two different cases – ZIRP with 0.0% IOR, and ZIRP with a law that abolishes IOR. In the first case hoarding of reserves is lower, because if/when the time comes to raise interest rates, there is no need to lower the quantity of reserves. In the second case hoarding of reserves is higher. So please do not say “abolish IOR”, please say “temporary lower IOR to zero”.
You said:
“Actually, we know this is not true, as there were a very large number of bank failures during the booming 1920s, and their effect was utterly trivial.”
Even during booms economies are always receiving various AS shocks. Without bank failures 1920’s boom would be even more fantastic.
You said:
“Yes, the number increased during the early 1930s, but not enough to make a difference given the complete lack of macro effects from the 1920s failures. Worst case would have been a very small recession, and of course that’s overlooking the fact that those large bank panics in 1931-33 would never had occurred if NGDP had kept growing. So in practice, NGDP targeting is all we need.”
Bank failures have 2 disadvantages: AS shock if bankruptcy costs are high (this can be managed by a reform of bankruptcy process), there is also a risk that the volatility of AD will increase. Yes, NGDP targeting will turn things back eventually, but in the meantime it is a very challenging job to keep NGDP expectations stable and avoid big undershooting or overshooting.
27. September 2010 at 17:46
123, I completely disagree with everything you said about interwar bank failures. I don’t think you have any idea just how trivial they were, in a $100 billion economy. It was about as consequential as a bunch of dry cleaners going out of business. These are mostly tiny rural banks, usually in just a single farming community. No, the boom would not have been even more fantastic, it would have been about the same. And no, bank failures don’t make it hard to target NGDP. There would have been no liquidity trap under NGDP targeting, and the liquidity trap is the only thing that make sit hard to target NGDP. It’s very easy for the Fed to accommodate the increase in currency demand associated with bank failures.
You said:
“You said:
“I don’t understand how a higher IOR can lower interest rates.”
IOR reduces hoarding of reserves. Let’s compare two different cases – ZIRP with 0.0% IOR, and ZIRP with a law that abolishes IOR.”
I asked how a high IOR is easy money, and you respond with an example where a zero IOR is easy money. I fail to see the point. I understand that, I don’t understand why a higher IOR is easy money.
28. September 2010 at 04:08
Scott, OK, the boom would have been about the same, except in those unimportant rural areas. And yes, AS shock from prohibition was much more important in the 1920s.
I don’t think liquidity trap makes it hard to target NGDP – there are always plenty of assets that are not in the trap. Systemic banking panic makes it harder to target NGDP because it badly impairs price discovery.
You said:
“I asked how a high IOR is easy money, and you respond with an example where a zero IOR is easy money. I fail to see the point. I understand that, I don’t understand why a higher IOR is easy money.”
I gave an example where higher expected future IOR is easier money. Just last week inflation hawk Plosser said that IOR is a bad idea. Imagine that his proposal is implemented immediately. On one hand, this will cause fed funds rate to drop by 20bps – this will provide more stimulus. On the other hand, this means that reserves will be rapidly withdrawn after x years when the FOMC will raise interest rates – this encourages the hoarding of reserves. The second effect is more important.
In late September 2008, banks were hoarding reserves, they feared that reserves will be very valuable for them during next three months. With IOR program, the Fed reduced the risk that reserves will be rapidly withdrawn if inflationary risks materialize, for any given quantity of reserves, this reduced hoarding.
28. September 2010 at 11:57
Scott said “I asked him just one question; how bad would the Great Depression have been if the Fed had targeted steady NGDP growth, but the bank failures had occurred anyway? As I recall, he hemmed a hawed a bit in answering the question. I still don’t think the Fed has a good answer to my question from way back in the early 1990s.”
Aren’t the two intertwined? If NGDP didn’t collapse we would have had fewer failures.
29. September 2010 at 12:10
123, You seem to be restating the proposition that temporary currency injections are not expansionary, because they will be hoarded. I’ve always agreed with that, but consider it an entirely separate issue. However, it does point to the fact that the key issue is not how much reserves are out there, or what interest is paid, but rather the Fed’s nominal target.
Banks held much more reserves in October than September, so I don’t understand in what sense it reduced hoarding of reserves.
Thruth, That’s exactly the point. The Bernanke banking story was supposed to show that sound monetary policy isn’t enough, you also must address the financial problems separately. But the NGDP targeting counterfactual suggests that sound monetary policy might well have been enough, as any banking crisis would have been much smaller.
30. September 2010 at 04:16
Scott, you said:
“You seem to be restating the proposition that temporary currency injections are not expansionary, because they will be hoarded. I’ve always agreed with that, but consider it an entirely separate issue.”
It is not a separate issue. With IOR, expansion of monetary base is permanent. Without IOR, monetary base will contract when inflationary pressures emerge. Inflation hawk Plosser understands this and says IOR should be abolished.
You said:
“Banks held much more reserves in October than September, so I don’t understand in what sense it reduced hoarding of reserves.”
We can measure expected willingness to hoard reserves during next three months with 3 month Libor. 3 month Libor started going down from 10 October, 2008. Banks will hoard excess reserves until 3 month Libor will become smaller than expected risk-adjusted return on other assets.
30. September 2010 at 06:10
123, You are providing two options which are both doomed to fail, and then telling me those are the only two options available. I don’t agree, but if you are right I should just shut down this blog, as we are doomed to fail.
I believe the best way to measure the demand for ERs is to look at the quantity of ERs. I see no reason to assume bank ER holdigs are not in equilibrium. No one has a gun at their head forcing them to hold ERs, they can hold T-bills instead.
30. September 2010 at 07:02
Scott, I am telling you that NGDP targeting with IOR has lower macro volatility during periods of banking turbulence than NGDP targeting without IOR.
Of course banks are holding reserves in equilibrium. Yet look at these three different equilibria :
1) 0.0% IOR, 0.5% 3-month LIBOR, negative -2% expected risk-adjusted return on private sector assets
2) 0.25% IOR, 0.3% 3-month LIBOR, negative -1.5% expected risk-adjusted return on private sector assets
3) 0.25% IOR, 0.27% 3-month LIBOR, positive +1% risk adjusted return on private sector assets
We are now in equilibrium #2. Plosser wants to move to equilibrium #1. We should move to equilibrium #3 by switching to NGDP path targeting.
Also note that ER is a legal term, and not an economic one. Why do you think that some obscure regulation knows best how many reserves are really required?
30. September 2010 at 18:38
123, I do understand ER is a legal term. BTW, I oppose RRs.
I don’t think we know anywhere near enough about macro to talk about which scenario would have the most unstable NGDP path. Let’s start with the fact that if we really did 5% NGDP targeting, nominal interest rates never would have fallen down to zero.
So I’m not going to argue with your hypothetical, as I think there are too many imponderables to have any confidence either way.
1. October 2010 at 05:06
Scott, but there is a debate within the Fed – when the time comes to implement the exit strategy, what should be the first step – raise IOR or reduce the monetary base?
1. October 2010 at 18:17
123, My preference would be to get rid of IOR, and also RRs. I don’t see why you need either. The strongest atrgument for IOR is to overcome the inefficiency associated with RRs.
2. October 2010 at 05:00
Inefficiencies associated with the lack of IOR, and with RRs, are different inefficiencies. If RRs are abolished, without IOR transaction activity will shift from reserves to other media of exchange, and the Fed might lose the control of monetary conditions.
2. October 2010 at 05:14
Scott,
The amount of excess reserves is a function of the FED’s balance sheet. While an individual bank can reduce the amount of excess reserves it holds, the banking system in aggregate cannot. There are only two leakages to the this and they are both minor compared to the amount of ER: currency and required reserves.
IOR simply creates a floor under the FF rate. It isn’t anything magical. Without IOR the FF rate would go to zero as banks in round robin fashion would push overnight rights lower to get rid of ERs.
In fact, with IOR the FED has the ability to target a negative funds rate. Without IOR it would be impossible.
The main advantage of the IOR is that it allows the FED to separate the size of its balance sheet from the target it sets for the FF rate.
2. October 2010 at 08:32
123, Actually, it would make things easier. As we’ve seen, the demand for currency is more stable than the demand for reserves. More of the base would be currency.
Ed, You said;
“While an individual bank can reduce the amount of excess reserves it holds, the banking system in aggregate cannot. There are only two leakages to the this and they are both minor compared to the amount of ER: currency and required reserves.”
This is wrong. Banks can determine the amount of excess reserves they want to hold, even in aggregate. Believe me, if the interest rate on excess reserves fell from 0.25% to negative 4%, all those ERs would quickly spill out into RRs and cash. We don’t see that because the Fed doesn’t want that to occur. Indeed that’s exactly why they started paying interest on reserves–there were afraid of high inflation.
Ed, you said;
“In fact, with IOR the FED has the ability to target a negative funds rate. Without IOR it would be impossible.”
Yes, and I believe I was the first to propose the idea of negative IOR in an economic journal article (but not certain.)
2. October 2010 at 09:17
Scott,
I already mentioned those leakages (RR and currency) in my original comment. The total of RR is around $50B vs $1T ER. No matter what the level of lending I don’t think RR increases materially. But there should be a run on banks if IOR goes negative and demand for currency which yields 0 picks up. By and large FED balance sheet is equal to ER plus RR plus currency.
But that is the whole point why IOR is a useful tool for the FED. Argue the FED needs to lower the IOR. Don’t argue it needs to eliminate it. Should the FED decide to go to a negative target on the FF rate, IOR would be the perfect tool to accomplish that.
You have probably already read this NY FED paper, but it might be useful reference for your readers: http://www.newyorkfed.org/research/staff_reports/sr380.pdf
2. October 2010 at 10:23
Ed, I wasn’t disagreeing with your assumption that the amount of RR plus cash is unlikely to rise by a trillion under normal circumastances. Rather I disagreed with the reasoning. It’s not because banks, in aggregate, don’t control the level of ERs, they do. Rather the Fed will move the IOR around to prevent a massive disruption to the economy. So we probably aren’t far apart.
I don’t have any big problem with the IOR, I have a problem with it being adopted at a positive rate in October 2008. The impact (ceteris paribus in case “123” is reading) was contractionary. I agree they could go negative with the IOR, but I think the simplest system is to just eliminate both RRs and IOR and have 99% of the base consist of currency in normal times.
I’m not too worried about bank runs from depositors because of FDIC. I don’t see how eliminating IOR on ERs would create a bank run from bank creditors. If the IOR was so important that bank profits depended on it (which I doubt), then they could pay a high (inframarginal) positive IOR on RRs, and a negative rate on ERs. Since the level of ERs would fall close to zero, banks would still be recipients of a net subsidy from the Fed.
Yes, I have seen the NY Fed article. I seem to recall they cited one of my articles discussing negative IORs.
2. October 2010 at 13:09
Scott,
But banks in aggregate don’t control the amount of reserves in the system, the FED does. They can lend more, but required reserves on most lending is negligible. The only meaningful way for excess reserves to decrease is for depositors to demand currency. And with a negative IOR they have more of an incentive to do so. Even now, with low bank rates for depositors, currency has increased. So if banks had to charge you for your deposits, after all that is what a negative IOR means, then currency would increase as people withdrew funds from banks. At least currency rates are 0%, not negative. I shouldn’t have said bank run, as that has historic connotations that I didn’t mean to imply. It would make economic sense for people to hold currency over bank deposits at the margin if bank rates were negative.
IOR or IOER if you prefer simply creates a floor under the funds rate when the banking system has excess reserves. If there were no IOR, with excess reserves, there would be only one possible target for FFs and that would be zero. IOR is just a tool like the discount rate to control the FF rate.
2. October 2010 at 13:25
Scott, you said:
“Actually, it would make things easier. As we’ve seen, the demand for currency is more stable than the demand for reserves. More of the base would be currency.”
It would be very hard to control inflationary shock that could be caused by alternative source of liquidity.
you said:
“The impact (ceteris paribus in case “123”³ is reading) was contractionary.”
I am sure that for any given fed funds rate target IOR is expansionary. Bernanke was doing all he could at the time to reduce the opportunity cost of reserves and yet they were much higher than intended by the fed funds target.
you said:
“I don’t see how eliminating IOR on ERs would create a bank run from bank creditors”
IOR allows you to achieve simultaneously two different targets. One is interest rate setting consistent with macroeconomic stability, and second is the level of reserves consistent with low fears of financial instability. Without IOR, any shock that forces you to raise interest rates increases the risk of financial instability. The demand for reserves was very erratic in September/October 2008, and without IOR there is a great risk that on any given day the quantity of reserves will be too small.
2. October 2010 at 14:50
I made a mistake in my last comment. Obviously, reserve requirements apply to bank liabilities not assets. Nevertheless, it remains true that banks in aggregate cannot materially reduce ER no matter how they react. One banks effort to reduce ER forces them onto another bank.
2. October 2010 at 18:18
Ed, Suppose you have a monopoly bank, to make the analysis simple. The bank wants to hold less ERs, and thus buys some T-bonds. The seller has two options, take cash or deposit the money back in the bank. If they take cash, the level of ERs falls by that amount. If they redeposit it in the bank, most of the reserves go back into ERs, but some goes into RRs to back up the new deposit. Now the bank is still holding more ERs than they want. So they buy more T-bonds. Repeat over and over again. Each time cash and/or RRs go up, and ERs go down. It seems like a slow way to get rid of ERs, but remember, banks can buy bonds really quickly, the bond market is highly liquid.
Regarding the impact of IOR on monetary policy, the mistake is to assume that the stance of monetary policy can be characterized by the nominal interest rate. As Milton Friedman pointed out, very low rates often indicate tight money, not easy money. In 1937 the doubling of reserve requirements only increased short term rates by 1/4 point, but nevertheless the policy was deflationary, as it greatly increased the demand for base money. For example, if the supply of base money is fixed, then any action that increases the real demand for base money by 10%, will cause the price level to fall by 10%. And as we saw in 1937-38, that may be associated with ultra-low interest rates, as expectations of deflation tend to lower rates and overwhelm the liquidity effect. In addition, economic weakness tends to lower real interest rates. So I’m not convinced that the effect of IOR was to merely prevent interest rates from falling another 1/4 point. I don’t believe the stance of monetary policy can be accurately described merely by looking at interest rates, you have to look at the supply and demand for money. And IOR increases the demand for base money.
123, I don’t see why it would be hard to control inflation by adjusting the quantity of currency.
Regarding IOR, why not just have the Fed make discount loans? Wouldn’t that provide liquidity when a “lender of last resort was needed?
3. October 2010 at 04:26
Scott,
Banks are buying all the short term T-bonds they can. They cannot go out too far on the curve because they would be taking on too much price risk. But banks do continue to purchase money market instruments daily (tbills, libor, general collateral, repo etc) until rates on these instruments fall enough that the only option left is to place the money at IOR. This process plays out everyday. And it is why I call IOR a floor under the FF rate.
In terms of reserve requirments, remember that from 1994 banks were allowed to sweep and optimize their deposits into non-reservable accounts. Even during the real estate lending boom times of the last decade RR hardly grew http://is.gd/fIjV4.
In response to some of your other comments:
Of course I do agree people respond to real interest rates.
Operationally it is hard to see what IOR has done but prevent a further 21bps (effective FF does average .21) of easing. Whether that extra 21 would make a macro difference is hard to tell. A case can be made for a lower or even negative IOR, but I haven’t been addressing that issue.
Nevertheless, I hope in some way I have convinced you ER of close to $1T isn’t because of IOR. Even with no IOR and FF rate at 0% on a daily basis, ER would still be close to that level.
3. October 2010 at 05:14
Scott, what if you reduce the quantity of currency, but other monetary aggregates expand even more and inflation strongly overshoots?
The Fed did not have balance sheet space in September 2008 for more discount loans, this is why Bernanke wanted a tool that lets you expand the Fed’s balance sheet by as much as needed. This tool was IOR.
you said:
“Regarding the impact of IOR on monetary policy, the mistake is to assume that the stance of monetary policy can be characterized by the nominal interest rate.”
Monetary policy can be characterized by market expectations of future nominal interest rate together with market expectations of future quantity of reserves. IOR did nothing affect market expectations of future nominal interest rates, but it did have an expansionary impact on the expected quantity of reserves.
You are saying that IOR has changed both actual nominal interest rates and nominal interest rates consistent with monetary equilibrium. In fact, market expectations of nominal interest rates were driven by the FOMC and the risk of banking runs. The level of nominal interest rates consistent with monetary equilibrium also depends on the risk of bank runs. The goal of IOR was to reduce the risk of bank runs, this both reduces actual nominal short term interest rates and increases the nominal interest rate that is consistent with monetary equilibrium.
3. October 2010 at 11:47
Ed, You said;
“Banks are buying all the short term T-bonds they can. They cannot go out too far on the curve because they would be taking on too much price risk. But banks do continue to purchase money market instruments daily (tbills, libor, general collateral, repo etc) until rates on these instruments fall enough that the only option left is to place the money at IOR. This process plays out everyday. And it is why I call IOR a floor under the FF rate.”
I think you misunderstood my earlier argument, I’d never deny that at current levels of IOR banks would often prefer to hold ERs to T-bills, Indeed that has been a signiture issue on my blog for the past two years. I’ve argued that the IOR causes hoarding of reserves. If the IOR was negative 4%, then T-bonds would become much more attractive. What I was arguing is that banks can choose the levels of ERs they wish to hold. It is because they were afraid banks would try to sell off ERs that the Fed instituted IOR. Without an IOR there was a danger of high inflation.
I also agree that the IOR is a sort of floor on the fed funds rate. But I believe the ffr rate has dipped a bit lower on occasion, as some reserve holders do not received the IOR. I seem to recall these include the GSEs.
You said;
“Operationally it is hard to see what IOR has done but prevent a further 21bps (effective FF does average .21) of easing. Whether that extra 21 would make a macro difference is hard to tell. A case can be made for a lower or even negative IOR, but I haven’t been addressing that issue.”
I don’t think we are going to be able to reach an agreement here, as I don’t view low nominal rates as being easy money. I agree with Milton Friedman who argued in 1997 that the low nominal rates in Japan indicated that money had been tight, not that money was currently easy. The issue is not whether rates go down a bit or not, it is whether we can increase the supply of base money and reduce the demand. Right now the best technique would be to reduce the demand through a higher inflation target, But a lower IOR might also help a bit.
As far as the level of ERs, that depends on the stance of monetary policy; the higher the expected rate of inflation the lower the demand for ERs. The effect of removing IOR on reserve demand depends crucially on how it impacts inflation expectations. There is no mechanical economic model that can answer that question. But I certainly agree that it is possible banks might hold large levels of ERs at a zero IOR, indeed isn’t that the situation in Japan right now?
123, You said;
“Scott, what if you reduce the quantity of currency, but other monetary aggregates expand even more and inflation strongly overshoots?”
I have three responses:
1. That’s not very likely
2. The same problem could occur with any target.
3. If that happens you reduce the suppply of currency even further–until inflation expectations return to target. Of course if the markets know you plan to do this, then inflation expectations never move much in the first place. There was a big increase in the demand for US currency after the Soviet Union collpsed. It was not deflationary because the Fed smoothly accommodated it.
I don’t understand what “space for discount loans” means.
You may be right that the IOR was set up to reduce the risk of bank runs, but I’ve never heard the Fed make that claim. It seems odd, as it would be a good defense of their action. In any case, they could achieve the same net subsidy with a merely inframarginal IOR, that did not impact ERs beyond a certain level.
3. October 2010 at 12:58
Scott, yes it is not very likely. But imagine – the Fed attempts to reduce the stock of currency to zero, there is a deflation in black market goods, but 10%+ inflation in goods everybody purchases with credit cards.
You said:
“You may be right that the IOR was set up to reduce the risk of bank runs, but I’ve never heard the Fed make that claim”
IOR reduced the risk of bank runs not because there was embedded subsidy in IOR, but because it reduced fears that one day effective fed funds rate will be very high due to insufficient quantity of reserves.
Here are the first lines of IOR announcement:
“The Federal Reserve Board on Monday announced that it will begin to pay interest on depository institutions’ required and excess reserve balances. The payment of interest on excess reserve balances will give the Federal Reserve greater scope to use its lending programs to address conditions in credit markets while also maintaining the federal funds rate close to the target established by the Federal Open Market Committee.
Consistent with this increased scope, the Federal Reserve also announced today additional actions to strengthen its support of term lending markets. Specifically, the Federal Reserve is substantially increasing the size of the Term Auction Facility (TAF) auctions, beginning with today’s auction of 84-day funds. These auctions allow depository institutions to borrow from the Federal Reserve for a fixed term against the same collateral that is accepted at the discount window; the rate is established in the auction, subject to a minimum set by the Federal Reserve.”
The Fed was concerned with conditions in credit markets, and the concern was that these conditions were too tight. Conditions in credit markets were too tight because there was heightened risk of bank runs. Bernanke did not make Hamilton’s mistake that there exists an excess of reserves because daily effective fed funds rate is low, he knew that credit markets were signalling that the expected opportunity cost of reserves for the next 3 months is above 4%, meaning that there is a shortage of reserves. Hence a substantial increase of 84 day discount lending was needed to drive expected opportunity cost of reserves lower. To achieve substantial increase of lending Bernanke needed to expand liability side of Fed’s balance sheet. He knew that he could permanently expand reserves by switching to IOR.
Bernanke’s mistake in IOR announcement was in the word “substantial”. After one week he reconsidered and announced, that he would expand Fed’s balance sheet without any limit whatsoever until 84 day opportunity cost of reserves drops to desired levels.
3. October 2010 at 15:52
Scott, you said:
“In any case, they could achieve the same net subsidy with a merely inframarginal IOR, that did not impact ERs beyond a certain level.”
The reserves were hoarded not because of some ill thought IOR subsidy scheme in 2008. They were hoarded because the healthiest financial institutions could lend out reserves for 3 months at 4% to median banks in late September 2008.
I believe IOR on RRs and ERs should be equal. If FOMC would set ffr at negative 20bps, the Fed should pay negative 20bps IOR on both types of reserves.
4. October 2010 at 06:41
123, You said;
“Scott, yes it is not very likely. But imagine – the Fed attempts to reduce the stock of currency to zero, there is a deflation in black market goods, but 10%+ inflation in goods everybody purchases with credit cards.”
I don’t follow. Monetary theory doesn’t determine relative prices. It doesn’t matter whether you use cash or credit cards, relative prices reflect microeconomic factors.
I read that Fed statement as indicating that they were worried about ffr dropping to zero. That’s why Bernanke said removing the IOR would be expansionary.
You said;
“The reserves were hoarded not because of some ill thought IOR subsidy scheme in 2008. They were hoarded because the healthiest financial institutions could lend out reserves for 3 months at 4% to median banks in late September 2008.’
I don’t follow this. If you can lend the reserves out at 4%, why would you want to hoard them?
4. October 2010 at 12:53
Scott, you said:
“I read that Fed statement as indicating that they were worried about ffr dropping to zero. That’s why Bernanke said removing the IOR would be expansionary.”
I thought that Bernanke was worried that one day before the IOR announcement intraday high of ffr was 3,5%, three days before the announcement intraday high of ffr was 4%, and four days before the IOR announcement intraday high of ffr was 10%. Like Krugman, Bernanke understood that money is too tight.
Bernanke said removing IOR would be expansionary as long as ffr target range is 0-25 bps. Plosser said removing IOR will be contractionary when time for raising ffr comes.
You said:
“I don’t follow this. If you can lend the reserves out at 4%, why would you want to hoard them?”
Sorry. Banks hoarded reserves, because quantity supplied was too low. Quantity supplied was so low, that median bank had to pay 4% to make sure it will have enough reserves for next 3 months.
You said:
“I don’t follow. Monetary theory doesn’t determine relative prices. It doesn’t matter whether you use cash or credit cards, relative prices reflect microeconomic factors.”
Shortage of coins acts as a microeconomic tax on bus commuters in one high-inflation Latin American country.
5. October 2010 at 06:27
123, In the US all the various denominations are supplioed endogenously, once the overall bas eis determined. So there are generally no coin shortages.
I think the main problem was the quantity of fed funds. If rates were high, the solution was to increase the supply, not introduce IOR.
6. October 2010 at 05:15
Scott,
The announcement of October 13, 2008 said that the Fed will provide unlimited quantities of monetary base to stop the hoarding. Variable IOR is just an exit strategy preferred by doves, it reassures that enormous quantities of monetary base will not be withdrawn when hoarding stops.
You might have preferred unlimited auctions of monetary base without IOR, but inflation risk premiums would be too high without IOR. With unlimited offer of base money, it doesn’t really matter if current IOR is +25bps, zero, or -25bps, only the exit strategy matters.
you said:
“In the US all the various denominations are supplioed endogenously, once the overall bas eis determined. So there are generally no coin shortages.”
The extreme scenario is that without IOR, monetary base might contract to zero, but broad money aggregates might expand too fast.
7. October 2010 at 06:04
123, Monetary base contracts to zero. So Mexican drug dealers just show up at the Fed with truckloads of currency? How could that plausibly happen?
The Fed needs to do level targeting if they want to keep inflation expectaions anchored. The base should be endogenous.
8. October 2010 at 04:12
Scott, during financial crisis, the base should be determined by the need for LOLR operations, and NGDP expectations should be anchored by adjusting IOR.
Let’s say the Fed had NGDP level targeting in September 2008, together with prediction markets. Without IOR, prediction markets set the quantity of the monetary base. Half of the days, prediction markets overestimate the quantity needed, with no adverse consequences. During other days, prediction markets underestimate the quantity needed, financial market stress increases, this creates a feedback loop whereby economy becomes more sensitive to errors in estimating optimal quantity of monetary base.
With IOR, demand for LOLR operations determines the quantity of the monetary base, and prediction markets set the level of IOR that is needed to stabilize NGDP expectations. Errors in estimating optimal level of IOR cause no damage in both directions.
After Lehman, it took one month until the Fed understood that it is impossible to reliably determine the quantity of monetary base that needs to be supplied, so on October 13, 2008 they promised to supply unlimited quantity of monetary base according to demand in LOLR auctions. After Lehman, it took two years until Dudley said in public that level targeting is a good idea.
These two delays have caused enormous damage to the global economy.
You said:
“Monetary base contracts to zero. So Mexican drug dealers just show up at the Fed with truckloads of currency? How could that plausibly happen?”
The most “plausible” scenario where we get hyperinflation with zero monetary base is when Ron Paul introduces two bills after inflation reaches 10%. The first bill abolishes the Fed, and all currency in circulation is declared to be null and void. The second bill establishes the gold standard. Unfortunately, only the first bill passes the filibuster. We get hyperinflation after president Palin bails out American Express and every card-holder gets a new 200000 USD credit limit.
More seriously, here is Woodford:
“Even if reserve requirements were to become completely
negligible in the United States, owing to further improvements
in the ability of banks to serve the needs of their customers
without holding appreciable overnight balances in reservable
accounts, effective control of the funds rate should still be
possible through the adoption of a channel system for
implementing the Fed’s operating targets for the funds rate.12
Such a system should be highly effective, and would not require
any change in the way that the Fed formulates its operating
target for the funds rate. The effectiveness of funds rate control
as a means for influencing the overall level of spending and
hence inflationary pressures would be unaffected.
Implementation of a channel system in the United States
would, of course, require certain institutional changes. One of
the more obvious of these is that interest would have to be paid
on Fed balances, contrary to current practice. But this is a
development that would have much to recommend it, quite
apart from the way in which it would be possible to enhance
control of the funds rate through linkage of the interest rate
paid on reserves to the funds rate operating target.”
Here is the source:
http://papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID804633_code387943.pdf?abstractid=804633&mirid=1
8. October 2010 at 15:13
123, You persist in assuming that the banking system would behave under 5% NGDP futures targeting just as it behaved under expectations of FALLING NGDP. In fact, the system would have been in far better shape. But let’s assume I am wrong, what then? Then the MB would have increased enough to lead to 5% expected NGDP growth. If NGDP growth cannot hit 5% during a financial crisis, then it would have had to increase enough so that investors expected no financial crisis. So I don’t see the problem. If there are separte solancy issues having nothing to do with liquidity, then maybe you need a TARP–I’m agnostic on that issue.
Regarding Woodford’s 2002 call for interest rate targeting, I’m inclined to ask “How’s that working out for you now that rates are at zero and we need lots more stimulus?”
I know the Ron Paul thing was a joke, but there is always collector demand. 🙂
9. October 2010 at 06:04
Scott, solvency problem under 5% NGDP targeting would be much lower in the crisis of 2008. But what about the next crisis? Leverage will expand until the point where there is significant probability of widespread insolvency after the next real shock.
Liquidity was also a very significant problem. Diamond and Dybvig can threaten even very sound institutions. With quantity prediction markets, there is one chance in 32 that for a five days in a row monetary base will be too low to prevent bank runs. This is quite robust, but IOR is even better.
Woodford’s 2002 paper looks even better after the crisis. His warning that financial innovation makes IOR necessary looks prescient – the largest upside deviation of interest rate from targets was observed in the US, problems in the UK and Eurozone were smaller because had IOR. The pleasant surprise is Switzerland – even though it had two huge toxic banks whose assets are greater than national GDP, Swiss national bank targets innovative financial markets directly via 3 month LIBOR, and systemic banking panics are impossible in Switzerland by definition.
Sweden is a good example of Woodfordian monetary policy when zero bound is hit and more stimulus is needed. Their central bank sets expected future path of policy rate during each meeting, so if forecasts drop below desired target, they can extend the period during which rates will be equal to zero.
10. October 2010 at 04:51
123, I still don’t see how any of this answers my claim about NGDP futures targeting. Either bank panics are a problem for NGDP, or they aren’t. If they are a problem, NGDP targeting should automatically provide enough liquidity to prevent them. If not, then why would we care about bank failures? Indeed bank failures are a good way to reduce moral hazard, so I’d welcome failures as long as they didn’t cause a recession.
I would think after their massive losses in this crisis, banks would hold less leverage, not more.
I can’t imagine where the 1/32 comes from. It seems to be 2 to the fifth, but how can you know there is a 50% chance of the base being too low to prevent a run on any given day?
12. October 2010 at 12:54
Scott, the problem is that Diamond-Dybvig model has two equilibria with different NGDP. NGDP targeting that is based on the quantity of monetary base will set the probability of a bank run to a level where it is expected on average that NGDP will grow by 5%. Yet if the run happens next quarter’s NGDP will grow by 2%, and if run does not happen NGDP will grow by 6%. Yet the socially optimal probability of bank runs is different, only the runs that cannot be prevented by safe LOLR operations are socially optimal. The solution is to determine the quantity of LOLR operations according to the cost of LOLR support, and use the IOR mechanism for NGDP targeting.
You said:
“I would think after their massive losses in this crisis, banks would hold less leverage, not more”
After four years the most profitable banks in the banking industry will know that losses were caused by monetary shock, not by excess leverage.
You said:
“I can’t imagine where the 1/32 comes from. It seems to be 2 to the fifth, but how can you know there is a 50% chance of the base being too low to prevent a run on any given day?”
Oh, it is not on any given day. Only approximately once in a decade the probability of a run is 50%. NGDP targeting (quantity of base) will cause a Great Recession once in 320 years.
15. October 2010 at 15:22
123, If a bank run ever started to develop, the market would add enough money to where it wasn’t expected to develop. Recall that monetary policy is happening every day in my system, not once every six weeks (which your example seemed to assume.) It’s all in real time, so I can’t see how a bank run could be consistent with 5% expected NGDP growth.
In any case, I’m all for regulating banks by requiring large downpayments on any loans using FDIC or TBTF protected funds. That will limit leverage in the system. If we had done that over the past 10 years, this crisis never would have happened. If that doesn’t work, then abolish FDIC and TBTF.
You said;
“NGDP targeting (quantity of base) will cause a Great Recession once in 320 years.”
In other words, it would be successful beyond my wildest dreams.
18. October 2010 at 04:07
Scott, a run on shadow financial system has started when NGDP expectations were on target in August 2007. FDIC was not involved in the shadow financial system.
Expectations of stable NGDP path will increase optimal level of leverage until there is some slight probability of a crash.
My example assumed daily monetary policy, just like NY FED is intervening daily.
Great Recession every 320 years is not good enough. With Greenspan we had a Great Recession every 250 years 🙂 . With IOR NGDP targeting a Great Recession would never occur.
18. October 2010 at 04:24
You said:
“If a bank run ever started to develop, the market would add enough money to where it wasn’t expected to develop. ”
No. The market would add enough money to where it is expected that NGDP will stay on target on average. This average includes a variety of scenarios, some of them involve bank runs.
18. October 2010 at 18:23
123, Two points:
1. I’ve argued repeatedly that it is fluctuations in expected NGDP that really matter, and that transitory movements in actual NGDP aren’t a big problem. That’s because they don’t cause big changes in asset prices. And wages are influenced by expected NGDP.
2. The August 2007 bank run did not cause big macro problems. Even 12 months later many economists still didn’t think we’d have a recession. We needed a NGDP futures market to handle the period from December 2007 onward. It was December 2007 when NGDP forecasts probably started falling signficantly (and nominal bond yields were plunging–which is a tipoff.)
3. If we never have recessions all macroeconomists will lose their jobs. 🙂
19. October 2010 at 03:08
Scott,
Transitory movements in actual NGDP are not a big problem, nominal GDP drops on Sundays without any adverse consequences. However, variations in dispersion of expected NGDP can be very harmful.
The period from August 2007 to September 2008 is a very useful natural experiment. In Eurozone, there was a credible promise to use LOLR to prevent bank runs, high IOR was used to control inflation expectations, dispersion of expected inflation was low, there was no crisis in the Eurozone. In the US, fears of bank runs have led to a wide dispersion of expected inflation, in April 2008 many economists thought there was no recession and there is a high inflation risk, while behaviour loan officers was consistent with fears of below-target inflation. This shows that dispersion of inflation expectations have caused great troubles for inflation targeting regime. The problems with dispersion of forecasts are also applicable to NGDP targeting.
Macroeconomists will not lose jobs, because we will have RBC recessions 🙂
21. October 2010 at 05:05
123, Europe’s not a good example to cite, as they saw a bigger fall in GDP than we did, depsite not beening in the center of the subprime crisis.
Your discuss of the US is very speculative, and in any case relies on inflation data that is not relevant to my argument. I see no reason to believe that loan officers had different NGDP forecasts than NGDP market participants would have. The consensus of economists is often off course, that’s no surprise.
22. October 2010 at 06:52
Scott, I was referring to the period before Lehman, and monetary policy was better in the Eurozone than in the US at that time. European and American financial systems were equally exposed to subprime securities, and subprime financial crisis has started in Paris in August 2007. Since August 2007 financial stocks fell sharply both in Eurozone and in the US, but because ECB had IOR, Eurozone did not suffer from increased dispersion of AD expectations.
In 2008 there was an AD shock, so inflation expectation data and NGDP expectation data had tight correlation.
It is hard to tell whose monetary policy was better after Lehman, as differences in GDP drop can be attributed to different exposures to international trade shocks, Eurozone was also badly affected by the problems in the eurodollar market that originated at the Fed.
You said:
“I see no reason to believe that loan officers had different NGDP forecasts than NGDP market participants would have.”
Everybody has got different NGDP forecasts. If market is efficient, market forecast reflects the risk adjusted mean forecast.
23. October 2010 at 05:49
123, Europe had a much smaller real shock in housing construction (I believe.)
You still have showed that loan officers and market participants had dispersed NGDP forecasts, you simply assumed that to be the case.
24. October 2010 at 05:44
Scott, fortunately there is a market indicator that could help us resolve the question. If we take S&P 500 as a NGDP forecast, then VIX is a good indicator that allows us to observe how dispersion of NGDP forecasts evolves in time.
25. October 2010 at 06:57
123, Yes, I think the VIX would be correlated with more NGDP uncertainty. I didn’t mean to imply uncertainty did not increase in 2008–I agree it did. I just wasn’t comfortable with assuming a specific group like loan officers had a different NGDP forecast from the market, and that we could know whether they were over or under estimating NGDP.
27. October 2010 at 11:31
Scott, when the uncertainty increased, one group of people was tightening lending standards, another group was hedging inflation risk with commodities. Even if you would disagree with my specific assessment of facts, increased uncertainty about NGDP should lead to greater volatility of actual NGDP outcomes. I believe that it is easier for the markets to forecast the short term interest rates that are consistent with the NGDP target than it is easier to estimate the size of mbase consistent with the NGDP target, so I believe that IOR leads to lower dispersion of NGDP expectations and better actual macro performance.
27. October 2010 at 17:53
123, That’s a common misconeption about NGDP futures targeting. One of the great advantages–discussed in my 2006 BEJ paper, is that you don’t have to pick a specific monetary policy insturment. The market can decide wheether it thinks the base or interest rates or exchange rates are more informative. The market simply does OMOs, which is the same thing the Fed does when it interest rate targets. The market can look at the fed funds rate, and decide whether it wants to do more or less OMOs. So I think we agree.
30. October 2010 at 07:07
Scott, both mechanisms (IOR targeting and quantity of OMOs targeting) have the same advantage as they extract information from a wide variety of sources. The problem is that one mechanism fixes a daily interest rate, another mechanism fixes a daily quantity. I believe that the second mechanism is much less stable with respect to intraday shocks to money demand.
1. November 2010 at 00:10
My crazy 3 AM idea:
There is no aggregate demand as a scalar quantity. At least not one that makes any sense.
You cannot aggregate demand into a scalar quantity due to Arrow’s Impossibility theorem and Simpson’s Paradox and have it make much sense. I prefer to think of aggregate demand and supply as vector valued functions. Where they don’t meet up (wether you think this is from market inefficiencies or government involvement is secondary) you get a separation you can measure by using a cross product.
When we use the standard definitions of AS and AD we implicitly assume
1. that there are no inefficiencies (for example no minimum wage, etc) so AD always meets up with AS.
2. We ignore ignore that when we aggregate we lose the transitive property of preferences.
I propose a better way of thinking about it:
ADxAS=Innefficiency measure, where AS and AD are vector valued functions of quantities for the entire economy.
Now, on the other hand we can talk about demand for money and supply of money, but that doesn’t immediately give us AS and AD, because different goods should have different elasticities wrt to money.
4. November 2010 at 17:31
123, Haven’t we gotten along fine with OMOs and no IOR for many years? It’s only when the Fed let NGDP expectations fall that we got into trouble.
Doc Merlin. I define AD as NGDP–so that is something definite.
4. November 2010 at 22:23
Standard deviation of effective fed funds rate was low for many years. The system broke down, and IOR was necessary.
7. November 2010 at 04:54
123, So two things happened in late 2008; NGDP crashed and the fed funds rate became more volatile. I say the first was the big problem, and the second might well have been a symptom of the first. And IOR contributed to the first.
22. July 2011 at 03:57
Hier nochmal in Kürze, was man machen könnte – wenn man wenig Geld besitzt 🙂 : Ãœberflüssiges wie lästige Zeitungen usw abbestellen, alles weg, was nicht sein muss, das kostet nur und bringt nichts, auch Abo-TV usw