Does loosening monetary policy mean lower or higher nominal interest rates?
An article in today's Financial Times (H/T Brad DeLong) is a good illustration of the problems that arise when central banks frame monetary policy as a (conditional) time-path for nominal interest rates.
The Fed is trying to communicate two things. First, it is trying to communicate that it will buy long term bonds and this policy will be effective by pushing down yields on long term bonds, which should increase consumption and investment demand. Second, it is trying to communicate that this policy will be effective in increasing future inflation and real growth, both of which will push up yields on long term bonds. The Fed's "communications strategy" is self-contradictory. No single individual can believe both parts of that communications strategy at once.
“The Fed has been sending the message that its chequebook is ready and it will do what it takes to reflate the economy,” said Jan Loeys, head of global asset allocation at JPMorgan Chase. “What no one knows is whether inflation will start to show in two weeks or two years.”
Mr Loeys added: “We are seeing longer-term-thinking clients becoming increasingly wary of bonds and hedging against inflation. Shorter-term thinkers are still willing to buy bonds, on the presumption that they are nimble enough to get out when inflation comes to push yields up.”
Here's my take. There's one group of people who have ignored the first part of the Fed's message, and who are selling bonds. There's a second group of people who have ignored the second part of the message and are buying bonds. And each individual in the second group thinks he is quick enough to get back out before everyone else in the second group changes his mind and joins the first group. And they can't all get out before the others, of course.
What actually happens in the bond market depends on the relative size of the two groups, and how much wealth they are willing and able to bet on their beliefs. So the bond market cannot tell us whether the Fed's policy is working. We have to listen to people like Mr Loeys instead, who listens to the people actually trading in the bond market.
This is a bad way to manage a communications strategy. The Fed is telling the market it will push a policy lever down, that there is a transmission mechanism from this policy lever to the rest of the economy, and that pushing the policy lever down will pull the economy up. Unfortunately, there is another reverse transmission mechanism from the economy back to the policy lever, and if the economy pulls up, that will pull up the policy lever. So the market can't tell, just by looking at the policy lever, whether the Fed is really pushing it down, and if the transmission is working the way it is supposed to work.
It's as if some bad mechanic had hooked up the power steering the wrong way around. So if you wanted to turn right, you tried to turn the steering wheel left, at which point the power steering would kick in and force the steering wheel right overpowering your hands.
Expectations are more important than anything else in determining interest rates, output, and inflation. Those stupid bits of paper we call "money" would be worth nothing if people didn't expect them to be worth something. The whole purpose of central banks' having a communications strategy is to manage those expectations. You can't manage expectations with a self-contradictory message.
Here's an alternative communications strategy: "The Fed will loosen monetary policy by buying stocks; this will raise stock prices and make households more willing to consume and firms more willing to invest; this will increase aggregate demand and create a recovery. And, by the way, expectations of recovery will raise stock prices." That alternative communications strategy is internally consistent. Both parts of the message tell you that stock prices will rise. If it's credible, the Fed won't actually need to do anything, because stock prices will rise simply because expectations change. (Yes, I know the Fed maybe can't legally do this; but laws were made to be changed.)
And that's how the monetary policy transmission mechanism is supposed to work. The central bank tells people where it wants expectations to go, and expectations go there, and markets follow those expectations. All the central bank needs to actually do is make sure it doesn't subsequently do anything that would contradict those expectations, because that would harm its ability to manage expectations in future. The monetary policy transmission mechanism is not a set of mechanical levers. It's a communications strategy.
Worthwhile Canadian Initiative 
I don’t think it’s as paradoxical as you suggest. The front end will rally, the back end will widen. So the task of the market is figure out around what point the steepening is going to occur. Somewhere in the two to five year range, maybe?
OT: Your power steering metaphor is highly evocative for me. I accidentally once reversed power steering hydraulics. Almost broke my arm when I tried to drive it.
Also… Though I agree that buying stocks might be more effective than buying bonds, I think there are far better ways to deal with disinflation that don’t require changes in legislation. E.g.:
1) Fed prints money
2) Fed gives money to treasury (or buys debt from treasury with it – whatever)
3) Treasury sends cheques to all citizens
4) Disinflation? Repeat.
This will eventually cause inflation, which is the communication strategy you are looking for.
Actually it would, of course, require budget legislation. But still… Wouldn’t have to change the fed charter.
K: Nice save with the second comment.
I can’t speak on what the Federal Reserve/Treasury can and can’t do, but I’m pretty sure the functions of the Federal Reserve and the Treasury are roughly combined in the Bank of Canada, and I don’t see anything in the BOC legislation that allows them to cut a cheque to a Canadian Citizen. Therefore, to do #3 would require a bill be passed in Parliment. Besides, I would imagine that cheques cut to citizens would come of the Consolidated Revenue Fund, which is not BoC’s responsibility.
You can check yourself, starting at Section 18 and on:
Click to access B-2.pdf
Paul:”the functions of the Federal Reserve and the Treasury are roughly combined in the Bank of Canada”
In Canada it might be the department of finance sending out the cheques. And yes, it would require a budget law. But, anyways, I assumed we were talking about the US since Canada is not (apparently) in a liquidity trap.
The action is a success if the real yield declines (e.g. TIPS having a negative yield) while the inflation expectation increases. It is true the two can offset to show no change in the nominal yield. The most recent Debt issuance with a negative real yield highlights the success of the communication strategy – the real cost of borrowing is lower while the inflation expectation is higher.
MTJ: But if the economy is expected to recover, real rates should rise. Maybe the policy was a total failure, which is why real yields have fallen? Who knows?
K: “OT: Your power steering metaphor is highly evocative for me. I accidentally once reversed power steering hydraulics. Almost broke my arm when I tried to drive it.”
OMG! I didn’t even know it could be done!
“The front end will rally, the back end will widen. So the task of the market is figure out around what point the steepening is going to occur.”
That’s normally how it is supposed to work, when central banks operate on the short end. But if the central bank says it is operating on the long end, isn’t it trying to flatten the yield curve?
“1) Fed prints money
2) Fed gives money to treasury (or buys debt from treasury with it – whatever)
3) Treasury sends cheques to all citizens”
Yep. That’s the exact case of helicopter money. It’s what Australia did. But you can’t subsequently reverse it, if you need to, without tax increases.
Nick,
“K”, above, describes the money-financed deficit form of a “helicopter drop”. My question is, doesn’t this describe the current (expected) situation? Of course the Fed is not allowed to buy Treasury bonds at auction. No matter — it just buys the bonds that the banks bought at auction. The anticipated level of QE2 (around $100b/mo.) just about equals net monthly Treasury issuance to finance the deficit. My observation is that there seems to be a lot of talk about how QE gets “money into circulation”, while the fact that the Fed will be financing Treasury spending seldom gets mentioned.
BTW, should the GOP carry the House in the mid-terms, Congressman Ron Paul is expected to be the Chairman of the House Subcommittee charged with Fed oversight. Yes, laws can be changed, but it seems unrealistic to think that the President would have the votes for such an action in the coming Congress. Further, the Fed faces record-low popularity and more aggressive oversight from Congress as a result of its quasi-fiscal actions in 2008-2009. To ask Congress for the authority to purchase stocks in this environment is to invite all sorts of Fed-threatening amendments to the enabling legislation.
Finally, if your proposal has merit, why not argue that the fiscal authorities should undertake it? It is completely within their purview, and certainly the Fed has no more expertise in buying stocks than Treasury does. If Congress could authorize the Fed to buy stocks (which, again, is highly unlikely), they could just as well authorize Treasury to do so, and the Fed could finance it. This would at least leave intact our democracy’s system of oversight and accountability over fiscal actions without threatening, in the long term, the Fed’s independence. Surely, you can imagine that, following a Fed purchase of stocks, any future President or Congress, in an effort to maintain popularity, would urge the Fed to prop up or “juice” the stock market. Were Treasury to have to do so instead, it would be held accountable for such a blatant manipulation of prices for political benefit.
Nick:”But if the central bank says it is operating on the long end, isn’t it trying to flatten the yield curve?”
I am not aware that they are intending to flatten the whole curve, all the way out 30 years. If it were me, I would try to shape the curve into what I would consider to be a credible path for the short rate. I.e. If we keep the short rate at zero for the next 3 years, there’s going to be so much inflation/growth by then that rates will have to rise fast. So I’m going to buy 3 yr yields down to zero and let subsequent ones equilibrate to a natural level. Its a way of clearly communicating the path of front end rates. I like the idea of the Fed announcing for how long they plan to keep rates at zero, along with this strategy. “A trillion dollars of QE” is useless as a communication tool, since quantity of money has so little relationship to inflation, so it would be much better if they talked about the path of rates. I don’t know what they plan to do or say, but at the very least, I don’t think there is any inherent communication paradox in QE.
As far as helicopter money goes, reversal does require taxation. But that’s OK. Helicopter money is negative taxation. It’s only natural to reverse it with taxation. And it is a far more effective stimulus than bond/stock purchases since most of the recipients would be debtors rather than savers.
David Pearson: You’re absolutely right. Advocating a helicopter drop is the same thing as advocating fiscal stimulus. The only real economic difference between fiscal/monetary stimulus is who gets the money in the end.
OT: The threads on the hydraulic input and output lines were the same size on that car. It was like a Stephen King novel. Pure demonic possession.
“But you can’t subsequently reverse it, if you need to, without tax increases.”
Is that a bug or a feature? Stock purchases are easy to reverse, and if they’re expected to be reversed, they may not have much impact on stock prices. (In particular, investors may have a model of the economy in which the inflation rate consistent with high-enough-to-produce-recovery stock prices is one that is too high for the Fed to tolerate.) Of course, the Fed can always overwhelm the market and force stock prices up, but that might require a huge increase in its balance sheet, which it might find unpalatable. The fact that the Fed can make more of a commitment in the helicopter drop case (because tax increases are more difficult than stock sales) is a way to overcome the “expectations trap.”
Negative TIPS yields make my brain hurt. Why would anyone in sign-up for that?
The Fed has really made a dogs breakfast out of this. How many people will say to themselves: “Hey honey! The Fed is sorta maybe kinda trying to tinker with the long end of the yield curve … let’s go buy a washing machine!”. It strikes me as ridiculous. They’re living in their own little echo chamber and mistaking it for reality.
David: I tend to agree. I find it hard to see the difference between:
1. Fed helicopter drop
2. Fed lends money to Treasury, which drops it.
3. Fed lends money to individuals, who lend it to Treasury, which drops it.
It may make a difference to expectations of future policy. Is it Fed or Treasury which gives future guidance? And we do live in a world in which central banks are supposed to be in charge of AD, rather than departments of Finance. And this is the same answer I would give if you suggest Treasury buys stocks, instead of the Fed. Whose communications strategy should we be listening to? A communications strategy is more credible if there is one voice, and that one voice has both current and future responsibility for the policy in question. That means, given history, an independent central bank.
K: Actually, I’m not exactly sure just which bit of the term structure the Fed wants to operate on. And if it were successful, and expected to be successful, that would change the whole term structure again, in ways that are hard to foresee. Just precisely when will real income and inflation recover? Again, at a minimum, the whole communication strategy gets very muddled.
Andy: That’s a topic for another post, that has been rattling around in the back of my mind for some time. The Fed wants a strong response to its entrance, but an easy exit. But the easier its exit is seen to be, the weaker will be the response to its entrance. I need to draw some supply and demand curves for the Fed’s entrance and exit.
Patrick: Negative TIPS? So invest in canned food instead! But if the Fed announced a clear inflation target, I think people could understand that. “Look Honey, the Fed says stuff will cost more next year. Let’s go buy a washing machine now”.
Andy: Really good point. Helicopter drops are extremely convincing.
Patrick:”Negative TIPS yields make my brain hurt.”
Me too. Especially in the absence of NGDP futures. Low rates may still cause some people to invest, but consume? I doubt it, until we see some serious real inflation – not just expectations thereof. And that isn’t likely to happen absent consumption. I think it’s a big mistake to confuse financial market inflation expectations with consumer expectations. And in this case, it may be the markets that are wrong, and consumer “real rates” may be much higher than the TIPS yield suggests.
Nick:
As a limiting strategy, the Fed could buy treasuries in order starting from the shortest maturity and working their way out the curve. They announce that they are going to do this, and that when they are done, they are going to keep the short rate at zero out until the maturity at which they stopped. Then, at every increment they buy that yield down to zero. Each incremental purchase is stimulative so presumably it would cause steepening of the rest of the curve. They stop when they like the shape. There is a very coherent communication strategy: The Fed chooses the first part of the curve. The market chooses the rest. Instead of setting the short rate, the Fed sets the end time of zero part of the curve. I think the market could understand this very well: as usual they are watching one degree of freedom determined by the Fed.
I am not saying that the Fed wont muddle it. I am just saying that there is no inherent inconsistency in communication of QE. You do however, need to leave some part of the curve in control of the market, in order to prove the effectiveness of the stimulus.
Nick:”Look Honey, the Fed says stuff will cost more next year. Let’s go buy a washing machine now”.
I wonder. I suspect consumers will react after they get burnt. Not before.
Yes, we need helicopter drops. No, stock purchases are not the answer. What if we relax the assumption of EMH, and say that stocks are already very expensive, like in 2000 (or perhaps in 2010). This is what happens:
“The Fed will loosen monetary policy by buying stocks, this will raise money expenditures and real profits. Expectations of higher real profits will raise stock prices. Our presence as a significant buyer will depress the required return on stocks, even though it is already too low. Our success will increase the probability that our exit strategy will start soon, this will increase the required return on stocks.” The result – stock prices are a very poor indicator of policy success.
This is what happened in Japan:
“The Ministry of Silly Walks will loosen fiscal policy by building bridges to nowhere, this will raise money expenditures and prices of construction materials. Our success will increase the probability that our exit strategy will start soon, this will depress the prices of construction materials.” The prices of construction materials are a very poor indicator of policy success.
In all these four cases the real indicator of policy success is M3. In case of helicopter drops, M3 will increase automatically. In case of QE2, M3 will increase a little bit, and velocity of M3 will increase a little bit, but it is hard to make quantitative predictions. The same applies to stock based QE and infrastructure investments.
K, good points. This is what basically happens in Sweden. Swedish central bank publishes the forecast of short term interest path with every policy decision – they are doing permanent QE2 without any need to buy actual government bonds. And differences between the central bank interest rate forecast and actual market data are very informative – they give us a constantly updated market opinion about which parts of central banks forecast are suboptimal.
Nick, buying stocks has one very significant advantage vs. QE2: bond bubbles are much more dangerous than stock bubbles.
Although the Fed can’t legally purchase stock directly, it could target stock prices, using bond purchases as a vehicle by which to manipulate them. (In fact, the Fed is normally in a position of using one market to manipulate another: it doesn’t operate in the interbank lending market, but it uses the Treasury market to target the federal funds rate.) Stock-price targeting has interesting implications. If the Fed followed such a policy consistently in pursuit of its mandate, then private sector stock purchases (in aggregate) would represent bets on a weaker economy, as investors anticipated an increase in the Fed’s stock price target. That should give us some degree of automatic stabilization.
I don’t find the washing machine example very compelling: it’s hard enough to get adults to believe in Tinkerbell over longer horizons, but at a one-year horizon, depending on the characteristics of the Phillips curve, she may fail to exist even if people do believe in her — which makes it that much harder to get them to believe. If you’re on a flat section of the curve, you have to move along it by convincing people that it’s less flat than it actually is. Over longer horizons, it becomes more and more plausible that we will eventually reach a steeper section of the curve, and the Fed can promise more inflation. It won’t make people buy washing machines sooner, but it could make them buy stock, and it could make businesses buy durable productive assets that otherwise would not have been profitable to buy.
Nick:”And we do live in a world in which central banks are supposed to be in charge of AD, rather than departments of Finance.”
I agree. So right now the Fed can’t do helicopter drops. But this is not an argument against helicopter drops. It’s an argument for giving the Fed authority to do helicopter drops.
Nick,
What if the Fed just purchased 3-month T-Bills? Those rates are already at zero so nobody would judge the policy by whether or not yields went down. They could combine these T-Bills purchases with a commitment to move the TIPS breakeven inflation curve back to its 2004-2008 range (2.3%-2.5%). This would have the positive feedback effect without the legal constraints of stock index purchases.
Nick, As usual, very good post. There are lots of assets the Fed (in theory) could buy; stocks, real estate, foreign exchange NGDP futures. In each case the price would move in the “right” direction.
Andy Harless makes a very good point. As long as they can target an asset price that is observable in real time, they don’t actually have to buy that asset. Indeed in my view they could buy zero interest T-bills (although that might not be enough.) Right now stock and foreign exchange prices are observable in real time. Inflation expectations are as well, although there are some questions about the accuracy of TIPS spreads, and I don’t know much about the liquidity of the CPI futures market. There is no NGDP futures market. So it’s one of those “lesser of evils” choices.
Regarding expectations (mentioned by several commenters), in my view it’s not about the average guy or gal. The Fed needs to convince markets that it is aiming for a higher NGDP three years out (compared to what is currently expected.) If they can do that, stock, commodity and commercial real estate prices will rise right now. That will start the ball rolling, and the average guy will respond to higher asset prices, not to the Fed. (Or at the very low end, rising employment opportunities as higher asset prices lead to more capital formation.)
BTW, what’s your take on the recent posts by Krugman? Doesn’t he seem a bit unimaginative? To me it seems obvious that the markets are reading the whole QE discussion as an implied increase in the Fed’s inflation target—or perhaps a commitment to hit the target, where previously that had been lacking, or perhaps a movement in the direction of level targeting. Krugman’s usually got a pretty sophisticated take on expectations. I can’t imagine why he keeps insisting it’s all about maturity transformation. He’s right that by itself that wouldn’t do much.
“As long as they can target an asset price that is observable in real time, they don’t actually have to buy that asset.”
If we added one clause to that statement, “as long as they can target an asset price and have the power to change it …”, we would have the “bazooka theory.” That theory has failed dismally in practice. Targeting something without the power to affect it amounts to the “air bazooka theory”: pretending to have a bazooka when everyone can see that you do not. I think that Nick will like this view, because the air bazooka theory is a common thread in many of his posts. But advocating this theory amounts to claiming that pretending to have a bazooka will work better than actually having one. That is a doubtful claim. Why should manifestly empty threats have the power to change expectations?
Scott: “Indeed in my view they could buy zero interest T-bills”
This the same as doing repos, only the money comes back after a month instead of a day (repo is just QE on overnight government debt). Even the Fed doesn’t believe there is any point doing repos at the zero bound. I can’t see how buying T-Bills would be any different.
I just realized the real reason that we need NGDP futures. It’s not so that we and the Fed can observe the market expectation of NGDP. It’s so that the Fed can manipulate those futures in a big way. Lets say they bought next years future up to 5%. Then anyone with a project that depends on 5% NGDP can invest in that project and short the future. In effect they can hedge out their macro exposure. They don’t have to take the Feds word that NGDP will grow. It’s more like arbitrage. And if the economy does not produce 5%, investors will receive a direct stimulus from the Fed. It would work in the other direction too. The Fed could manipulate futures down and people would invest in futures rather than real projects. Arbitrage ought to be a far more powerful force than manipulating expectations and gets us around the problem of Fed credibility.
And, Nick, yes it’s a bit like buying stocks. But it’s much better because 1) it doesn’t transfer wealth disproportionately to savers and big business and 2) it is way more transparent and direct both as a communication tool and as a hedging tool.
I agree with Phil. But even if they had a bazooka, and everyone saw they had a bazooka, it wouldn’t do any good if everyone knew they wouldn’t ever use it.
The Fed needs to take a lesson from Tony Soprano on credibility and managing expectations. Sometimes you need to back-up your threats with real muscle. Most of the time you don’t have to resort to muscle if you have a reputation, but every once in a while somebody tries to call your bluff. That’s when you need to remind everyone that you really are the baddest dude in town.
The specific machinery doesn’t matter all that much so long you have the ability to force people to do what you want them to do. I was only 6 years old at the time, but didn’t Volcker teach us that? The Volcker Fed said “we’re going to stamp out inflation”. Everyone said: “That’s nice. I’ll believe it when wages and prices stop rising”. Then the CB applied muscle; they raised interest rates to 18% (or whatever it was) and everyone started believing them. To this day, does anyone doubt a competent CB’s ability to stamp out inflation at will?
At ZIRP with high unemployment and idle capacity the CB can’t call our bluff with interest rates, and all the wishy-washy talk of bending the yield curve and keeping rates low for a long time is not going to convince anyone of anything (at least not in any reasonable time frame). The Fed really needs to call our bluff and apply some muscle.
The problem with “targeting and not buying” is that markets front-run.
Let’s say the 10yr is at a 2.6% yield, and the Fed announces that it will buy bonds to hit a target of a 2.2% yield. Traders will quickly vacuum up the bonds, so that the 10yr quickly hits 2.2%. At that point, Treasuries are “trading inventory”, waiting to be unloaded on the Fed. Along comes the Fed and says, “well, the yield is at our target, so we won’t actually have to buy many of those bonds.” Traders, realizing that there is no big buyer for their inventory, dump the bonds, and the yield quickly rises back to 2.6%.
Something like the above has occurred in 10 and 30yr Treasuries in the past few months. This is an indication that the bond market is not giving QE much credibility, particularly since Dudley and other Fed officials have stressed that an important stimulative impact of QE is lower long term rates. What lower rates? They have just round-tripped, and they are at levels they traded at before the Bernanke Jackson Hole speech.
David Pearson: Yes. If the Fed promises to do QE, they may actually have to do it…
David Pearson makes a good point. It is a contradiction to hope to influence expectations and yet assume that people will not trade on those expectations.
But the Fed wants people to trade on those expectations. It’s not a problem! And if they trade against the Fed’s desires (i.e. rates go back to 2.6%), the Fed should punish them, by actually doing the QE they were threatening.
David and Phil: OK, I get your point. You are saying that the QE yield target is inconsistent with resulting probable path of the short rate and that therefore the market wont respect the QE target. There are two answers to this:
2) The Fed can overwhelm the market whenever it wants. If it wants 2.2% yield, it gets 2.2% yield.
1) The Fed should never set curve shape targets that are inconsistent with the path of short rates. That’s why they should always only determine the initial part of the curve and let the market do the rest.
No, when the Fed wants a 2% yield, and that isn’t a plausible yield, then what the Fed gets is not a 2% yield but no yield as it owns all the securities outstanding. Of course in reality the markets would stop being liquid long before then, and therefore they would stop transmitting expectations.
The index aggregates expectations, but changing the index will not change the expectation.
When you push too hard on the speedometer to speed up the car, you get a broken speedometer, not a faster car.
RSJ: OK, but I said probable, not plausible. The distribution of plausible yields is very wide. 3yX3y swaption vols are around 30% – thats 1 standard deviation log-normal. Two standard deviations is forward yield range of 1.6% to 5.4%. And that doesn’t even account for the smile, which is huge. And if the Fed says the yield is 2.2% because 1) we’re going to force it there and 2) that’s going to be the path of the short rate, then it suddenly becomes a lot more plausible. So no, the Fed can’t put the yield anywhere. But it can put it anywhere it wants, which is to say that it can use the curve as a very convincing way of asserting its intended path of rates.
The Fed can overwhelm the market whenever it wants. If it wants 2.2% yield, it gets 2.2% yield.
It is highly unlikely that the Feds has this kind of market power.
At the long end of the yield curve the Feds “power” is close to non-existent as opposed to the shell game on the interbank market.
Any possible change in the long term T-notes can be explained by psychological herd behavior rather than the Feds actual power.
David:
Of course the Fed is not allowed to buy Treasury bonds at auction. No matter — it just buys the bonds that the banks bought at auction.
The banks are not the major buyers at T-bond auctions, non-banks are.
My observation is that there seems to be a lot of talk about how QE gets “money into circulation”, while the fact that the Fed will be financing Treasury spending seldom gets mentioned.
Your remark is based on the assumption that the same group of non-bank entities that have recently bought government securities from the Treasury would be willing to turn around and re-sell those securities to the Feds, not to someone else, on the open market.
That’s a rather far-fetched proposition .
vjk:”It is highly unlikely that the Feds has this kind of market power.”
Based on what? What’s going to stop it? Restrictive balance sheet limits I’m not aware of? The most charitable interpretation that I can make of your comment is “The Fed doesn’t currently exercise that kind of market power.” Is that what you meant?
“At the long end of the yield curve the Feds “power” is close to non-existent as opposed to the shell game on the interbank market.”
There’s only $846Bn of outstanding treasury bonds (everything from 11 to 30 years). Do you really think the Fed couldn’t control the long end of the curve? (Not that I think they want to. It’s the short end they want to control.)
K:
The total bond market size is about $32 trillion. The global bond market size is about $100 trillion.
There is no special yield curve for Treasuries only.
Not that I think they want to. It’s the short end they want to control
If they want to control the short end, why would they buy long term ? Does not make sense.
Besides, they can control the short end only upwards, its already being at 0-25 bps.
K, the Fed can offer to buy them at 2%, and they will buy all of them at 2%.
In the money markets, there is a certain demand for actual cash, both to settle transactions as well to prevent bank overdrafts or to meet bank reserve requirements. That demand is inelastic over the short run. There is no substitute for cash when cash is needed to settle a payment.
But in the capital markets, there are plenty of substitutes for long bonds, or agencies, or what have you. There is no requirement for the private sector to try to outbid the CB, whereas a bank short of reserves must try to outbid the CB. The private sector can just sell the asset to the CB. Sure, there will be front running and noise traders, but these are short term and unreliable as policy transmission mechanisms.
So the private sector does sell 800B in treasuries to the CB, but assuming that they were indifferent in holding those bonds, then 1/2 of the proceeds of the sale will be used to purchase put options and the other half will be used to purchase call options, and no new investment occurs, neither do asset prices rise. Broker Dealers end up sitting on a lot of cash that is deposited with banks, creating 800B in excess reserves.
In the capital markets, you are not going to get asset prices to rise by retiring assets — you need to change the expectations of the market participants, so the 3/4 of them go long and only 1/4 go short. But trying to change the index that measures those expectations is not going to work.
“The total bond market size is about $32 trillion”. And almost none of it beyond 10 yr.
“If they want to control the short end, why would they buy long term ? Does not make sense.”
Do you know what they are going to buy? Last time they bought a lot of agencies. Duration generally around five years or less. And no, I wasn’t proposing they buy t-bills at 25 bps. I think they are going to buy mostly less than 3-4 yr debt, as I’ve discussed above, in some of the first comments. If they start buying the long end they get into the conflict that Nick discussed in the post. It’s highly counterproductive.
RSJ: “K, the Fed can offer to buy them at 2%, and they will buy all of them at 2%.”
No, they won’t. Only if no one believes that 2% is a reasonable average path of the short rate. In reality the range of opinions is extremely broad. There are plenty of people who expect us to fall into an extended liquidity trap, QE or not. And lots who think we’ll get rampant inflation. And if the Fed sticks to targeting yields that are consistent with its intended path of rates, it’s extremely unlikely that an overwhelming majority of the market will dissent. And if they do, the Fed has enormous power to punish them, either by manipulating yields or through the ultimate path of the short rate.
K:
Do you know what they are going to buy?
I vaguely recall there was some talk about 10 years and above.
My understanding was they intended to lower interest rates at the long end of the yield curve.
That’s futile for two reasons: the Feds cannot influence much rates beyond the interbank market rate; the outcome of the exercise is going to be even more bloated bank cash accounts at the Feds (which already stand at $1tril in “excess”) and not much more.
So, it is unimportant whatever maturity they are going to buy.
A simple answer to the Feds inability to influence long terms is sheer capital market size (bonds combined with other substitutes as RSJ said). There other theoretical factors of arguably less importance predicting the Feds impotence to influence long term.
All of the above may be bunk so one may be wrong.
But, on the other hand, judging by the Japanese QE experience re. long term and the QE1 voyage re. the same, one may be right.
K:
What is “the ultimate path of the short rate.” ?
By the “ultimate path” I just meant the path that is ultimately followed (out of the multitude of possible paths).
vjk:”I vaguely recall there was some talk about 10 years and above”
So you had proof they’re idiots, but you lost it…
“the Feds cannot influence much rates beyond the interbank market rate”
You repeating this won’t make it true. I see no reason why the Fed can overwhelm the overnight market, but not the market for term debt. Especially since they have the power to discipline those who disagree with them through the setting of the path of the short rate (for the last time). You need to acknowledge this point.
“the outcome of the exercise is going to be even more bloated bank cash accounts at the Feds…”
I never said it was going to work. My contention is that there exists a QE strategy with a coherent communication strategy. I don’t know if the Fed is going to follow such a strategy, but I’m not going to a priori assume they’re stupid. But even if they do follow the best possible strategy, it may not be sufficient to pull us out of a this liquidity trap (assuming that’s what it is).
“A simple answer to the Feds inability to influence long terms is sheer capital market size”
How about the potential sheer Fed balance sheet size?
The expectation hypothesis has been shown to be inconsistent with the data for about two decades now. From countless studies. It’s hard to think of a statement that has more solid econometric backing than “EH is false”.
Here is a reasonable survey: http://research.stlouisfed.org/wp/2003/2003-022.pdf
The fact is, you have no idea about what the actual relationship is between the short and long term rates. You have a theory that is contradicted by the data, but these contradictions are ignored for theoretical convenience. Actual markets are not disciplined by theoretical convenience.
Institutionally, the CB can control the zero rate because banks must outbid the Fed when they are short of reserves, and by draining, the CB has the power to place them in a net deficit position with respect to reserves. Therefore they will pay whatever the CB decides.
That is a very special case.
But no one is required to get into a bidding war with the CB for assets of maturity > 0. No one is forced to hold these assets, or can be required to have a short position in these assets. If the CB offers to overpay, then that is not discipline, it is free money, and those who hold the asset will sell as much to the CB as it is willing to buy. When it stops buying, then the CB loses control of the rates. If it buys all the assets, then still there will be an implied risk-free rate at that maturity, and you have no plausible mechanism to control this rate.
In the interest of full disclosure, I am long bonds, and I think they will fall, Japan style. The fed doesn’t need to try to make them fall.
I am in the RSJ camp. His point is exactly why Sumner’s idea of manipulating NGDP futures prices won’t work. If the fed is prepared to make an unlimited market, it can indeed make the futures price whatever it wants, but there is no reason for the underlying NGDP to fall into line with the futures price. This is the general case for asset prices, not an exception.
RSJ, isn’t there a substitution effect there? When the Fed buys treasuries and drives down yields, the spreads on risky bonds grows and eventually money flows there as well.
RSJ: I’m not familiar with the literature on the EH, but the EH as stated in the paper you cite, is a hypothesis that forward rates are literally expectations of future short rates under the historical measure. They aren’t. This isn’t even consistent with an arbitrage free rate model. In an arbitrage free world, discount factors are the reciprocal of the expected value of a bank account under the risk neutral measure. And the historical measure is related to the risk neutral one by an arbitrary conditional time varying drift. This drift (as well as convexity effects resulting from a literal expectation of rates) can likely explain the rejection of a simplistic EH under an empirical study. I don’t think this is very interesting, and it certainly doesn’t prove the existence of arbitrage in fixed income markets.
Anyways, when I use the word “expectation” I usually mean “in an arbitrage free model under the risk neutral measure”. So when I very loosely say “the yield is the expectation of the path of the short rate” I don’t mean in a trivial econometric way. It should really be read as “the price of bonds is determined by an expectation of a function of the path of the short rate”. And if the Fed specifically announces the path of the rate, then the market will arbitrage the yield to a level that is consistent with that path. Yes, the path has to be credible and it has to be respected, which means the Fed shouldn’t try to lay out a 10 year curve. It should start from the front and work its way out the curve as necessary.
As far as your argument that the Fed has special power over the short rate, I need to think more about the mechanism. But here is why I’m initially skeptical: If the Fed’s control over the short rate was just a technical trick and not a result of the unlimited balance sheet of the Fed, then the Fed funds rate would become disjoint from the real short term funding costs of arbitrage players. This is not the case: arbitrage players have near unlimited access to funding at fed funds plus a small fixed spread. This is because the banks are willing and able to source those funds in the fed funds market. If equilibrium were achieved though an non market mechanism the banks would be either unwilling or unable to do this. Furthermore, yields and the short rate path are related by “expectations”. This means that moves in one will cause significant flows/pressure on the other through carry trades. Over the long run, the average pressures in one ought to be about the same as the average pressures in the other. If the Fed has the power to directly effect the real funding costs of market participants then I think I think it ought to have the power to directly effect the market for expectations of functions of the future path of those funding costs. That’s my gut feeling, but no, I don’t have empirical evidence.
Finally, I agree with Nick and you and most of the rest of the people here, that indiscriminate buying across the curve by the Fed (if that’s what they choose to do) likely does not have a coherent associated communication strategy, and is therefore likely to be very ineffective.
Phil Koop:”Sumner’s idea of manipulating NGDP futures prices won’t work”
I said that, above. But Sumner may have said it before me, somewhere else.
“there is no reason for the underlying NGDP to fall into line with the futures price”
Is too. The reason is arbitrage. People can invest in projects even if they 1) depend on a high level of NGDP, and 2) don’t believe that level will be achieved. They can do this by investing in the project and short selling the future at the high level of NGDP futures price caused by the Fed. The Fed effectively guarantees the level of NGDP. So arbitrage will lift NGDP to the futures level. At least in principle. I’m not so sure if it would work very well in practice. But you can’t say that there is no reason.
RSJ,
I think in the case of a Fed S&P target, Nick would argue that Fed buying creates private demand for the asset. In other words, the Fed will not end up owning all the stocks, because by buying stocks, it will make private buyers also want to buy stocks. The link between the S&P level and private demand for stocks is the “wealth effect” on private spending that raises expected corporate dividends.
So what’s wrong with the above premise? By buying stocks, doesn’t the Fed make private holders “richer”? Isn’t there a “wealth effect” of rising equity prices on private consumption spending?
Of course, Fed-induced capital gains on stocks, all else equal, would support more current consumption by equity holders. The problem is that this increased spending is offset by the lower spending of savers that now face lower future returns from too-high equity prices. In effect, the Fed’s purchases may merely be viewed as a transfer of wealth to current holders from future holders forced to accept lower returns. The wealth effect benefit to current holders is only half of the equation needed to calculate the impact on total spending.
Of course, the above all holds true for permanent Fed bond purchase programs. In fact, a Fed bond price target would be no different than a stock one. This is because, using a dividend discount model, it makes no difference whether you raise the price of stocks through asset purchases, or raise the “value” of the same by targeting a lower l.t. bond yield and thereby reducing the discount factor. The problem in each case is that future savers face lower returns, and therefore must save more to support the same future consumption.
There is an exception to the above: the special case where rising prices beget expectations of HIGHER future returns. This is what we had during the housing bubble. Actors saw rising house prices as evidence that housing capital gains would continue well into the future, so they needed to save less to support the same future income.
Nick,
Agree with your comment on the real rate, however the inflation expectation will only increase if the policy is a success at reflating. i.e. Fed is pushing down real rate, if a success the real rate and the inflation expectation will increase. The real rate impact is messy but the inflation component will only go up if successful at reflating.
I’m losing track of the comments here. Just want a add a few points:
Whether or not the Fed actually has or needs a bazooka, that can do real things regardless of expectations, the communications strategy is a very important part of its arsenal. And a communications strategy that has that internal inconsistency can’t work as well.
A promise, or conditional promise, to use a bazooka in future, if credible, can be as effective or more effective than using a bazooka today.
There are games, games of pure coordination, where a credible communications strategy is all you need. Cheap talk does the job. We all want to meet somewhere. Where shall we meet? A single voice saying “Grand Central Station” is all that’s needed.
If people did expect recovery, then nominal and real interest rates would rise, and the equilibrium monetary base would fall. With a credible communications strategy, the Fed would actually need to point the bazooka in the exact opposite direction. In a sense, it’s not 50% communications, nor 100% communications, it’s 150% communications (if what you are communicating is interest rates and the monetary base).
The main effect of stock prices on AD is not via wealth effects. It’s Tobin’s Q. Higher real stock prices reduce the cost to firms of funding new investment via stock issuance. The P/E ratio on stocks is just as or perhaps more important than the dividend/price ratio on bonds, especially government bonds.