We always knew that interest rate targeting could never work in theory, because it left the price level indeterminate. But it seemed to work well in practice, and kept inflation close to target, so we eventually learned to overcome our theoretical squeamishness and embrace it as part of the reality of how modern central banks operate. But now interest rate targeting has failed in practice, and failed badly. It cannot keep inflation, and expected inflation, on target. We want to loosen monetary policy. And because monetary policy is interest rate targeting, we can't, because interest rates on safe liquid assets are already at zero.
OK, this is probably the weirdest post I have ever written. I am going to argue that interest rate targeting is not what central banks really do; it's a social construction of what they really do. Interest rate targeting is not reality, it's a way of framing reality.
That was weird enough, but I'm now going to get really weird. The failure of monetary policy is not caused by anything central banks are actually doing; it's caused by central banks' way of framing what they are doing, and by the rest of us accepting that same framing. The current recession was caused by those (and that includes especially central bankers themselves) who think that central banks use an interest rate as the control instrument. It's the framing of what central banks do that caused the mess, not anything central banks are actually doing. The social construction of reality is what dunnit!
Let me give an example of what I mean by "framing". It would be quite conventional to say that the Bank of Canada sets an overnight rate target as a function of various indicators, including inter alia the exchange rate. Write that reaction function as i=F(S,I), where i is the overnight rate, S the exchange rate, and I a vector of other indicators. Now, just invert that reaction function and write it as S=G(i,I). Exactly the same thing mathematically, but see how the framing has changed. Now the Bank of Canada is targeting the exchange rate as a function of various indicators, including inter alia the overnight rate. The Bank of Canada is doing exactly the same thing, but the account it gives of what it is doing, the framing, has changed radically. And the Americans would get mad at us if the Bank of Canada targeted too low an exchange rate!
Now here's a real-life example of how the Bank of Canada has changed the way it frames monetary policy, and why framing matters. The Bank of Canada used to construct and publish a "Monetary Conditions Index" which was a weighted average of the overnight rate and exchange rate. And it would talk about what was happening to the MCI whenever it changed the overnight rate target. This did not mean it kept the MCI constant over time. It would adjust the overnight rate target and so adjust the MCI whenever it felt that the indicators warranted it. But the Bank of Canada eventually stopped publishing the MCI, and stopped talking about the MCI, because it found the MCI interfered with the Bank's "communications strategy". People were thinking of the stance of monetary policy in terms of the MCI, and the Bank wanted people to think about the stance of monetary policy in terms of the overnight rate target.
When it stopped publishing and talking about the MCI, the Bank wanted to change the framing of monetary policy. And it succeeded, even though anyone with a calculator can construct the MCI from public data.
As any philosopher or sociologist can tell you, the way reality is socially constructed can have real effects. "I may lawfully nourish myself from this tree; but the fruit of another of the same species, ten paces off, it is criminal for me to touch. Had I worn this apparel an hour ago, I had merited the severist punishment; but a man, by pronouncing a few magical syllables, has now rendered it fit for my use and service…" says David Hume. (One fruit tree is on his land, the other isn't; the guy said he could buy the clothes). When one man body-slams another, is it a fight or a hockey game? A British visitor can't always tell the difference, and not because his eyesight is any worse than the Canadians who know how to frame what they are seeing. One framing causes a cheer; the other causes a call to the police.
How could a stranger, with good eyesight and a time series graph of all macro variables, really tell which one of those variables the Bank of Canada was targeting? OK, there is a way, if he has a very high-frequency data set. The stranger would notice that the overnight rate tended either not to move, or else move in discrete jumps, about 8 times a year. But if Fixed Announcement Dates were held monthly, or weekly, or daily, or hourly, the stranger would see nothing. In the limit, with continuous FADs, the "fact" that the Bank of Canada targets the overnight rate would be only a socially constructed "fact". Since I don't think hourly FAD's would make much difference to policy, especially under current circumstances, when nobody expects the overnight rate to change anytime soon, I think I'm safe in saying that interest rate targeting is at least 95% a social construction of reality.
A socially constructed reality, like who owns what, is a game-theoretic equilibrium held in place by players' shared expectations of how each would react to an out-of-equilibrium move, something they never of course observe in equilibrium. They follow the rules because of what they think would happen if they didn't.
A given time path of overnight rates would be highly inflationary under one set of expectations about inflation and real output growth, and highly deflationary under another set of expectations about inflation and real output growth. A time path of interest rates cannot measure the stance of monetary policy. A time path of interest rates does not define a coherent monetary order. An interest rate reaction function, with feedback from expectations to the time path of interest rates, might define a coherent monetary order. (Equilibrium would be maintained by players' shared expectations of what would happen out of equilibrium.) We thought it did for a decade or two, but events have proved us wrong. We have hit the zero limit where it can react no further, so all we have left is the time-path itself, rather than a reaction function.
If expectations of monetary policy coalesced around the time path of some nominal variable (one with $ in the units), we could escape the liquidity trap. Just let that time path for the nominal magnitude grow over time at a fast enough rate and the equilibrium overnight rate will rise above zero. If monetary policy were framed as the central bank setting some nominal variable, expectations of monetary policy could then coalesce around that variable, monetary policy would be loosened, and the overnight rate, as some endogenous response to monetary policy, would rise above zero. Under an alternative framing, a loosening of monetary policy would mean an increase in the overnight rate.
But we are stuck with framing monetary policy as setting an interest rate. So there is no way the Bank of Canada can try to say that an increase in the overnight rate counts as a loosening of monetary policy. Because given the way expectations are determined by the framing, it would be a tightening of monetary policy. The Bank of Canada is fettered by its own social construction of what it is doing.
Oh hell, it's late. I'm posting this anyway.
Worthwhile Canadian Initiative 
Nick,
It’s pretty clear they’re not using the overnight rate once they’ve hit the zero bound. So that needs to be qualified. They’re using credit/quantitative easing until they can get the system back into the usual interest rate zone. Monetarists may not like interest rate targeting, but I’ve never seen a monetarist propose a change to the architecture of the reserve system that would facilitate something that is essentially different. The system is set up now for the central bank to control the level of reserves, which affects the overnight rate. How do you set it up without the necessary presence of that influence? How do you set it up so that an “efficient” market determines the overnight rate without direct central bank influence? How do you set it up so that the central bank can influence any other variable without influencing the overnight rate at the same time? Have monetarists thought about the inherent architecture problem of a system in which the central bank doesn’t influence the overnight rate?
E.g. # 1 Scott Sumner’s NGDP targeting uses NGDP futures, but the OMO mechanism still influences the overnight rate. It’s still monetary policy implemented through interest rates – just a different indicator used as the trigger for the interest rate mechanism. E.g. # 2 FX intervention as a target will still automatically affect the overnight rate through the central bank’s effect on the domestic currency reserve system. It’s just a different indicator being used to trigger a specific action that has an effect on domestic reserves and the overnight rate. Even with FX intervention, the central bank must make a decision on “sterilization” and resulting reserves and resulting overnight interest rates. In both examples, the transmission mechanism involves central bank real time intervention in domestic reserves with direct effect on the overnight rate. Scott’s trigger is NGDP futures. Yours is FX. E.g. # 3 The current actual system is “eclectic” in terms of what is targeted “in front of” the overnight rate. But the necessary transmission involving reserves and interest rates is the same in all cases.
And BTW; central banks aren’t targeting interest rates primarily – they’re targeting inflation primarily and interest rates secondarily, now supplemented with credit/quantitative easing. The question is whether the interest rate is an independent variable or a dependent variable in the full transmission process. E.g. it’s a dependent variable in NGDP targeting and FX targeting; and it’s even a dependent variable in “eclectic” targeting.
Monetarists face a fundamental architectural conundrum in changing the essential nature of central bank interest rate transmission as their way of implementing any monetary policy, if that’s what they want to do. Otherwise, they’re targeting interest rates, either directly or indirectly.
One thing I missed in the post was the dual target: the overnight rate as the short term target and the inflation rate as the long term (fixed) target. They lost the social construction for the inflation target, but kept the social construction for the overnight rate target, unfortunately. The two frames weren’t mutually credible, and the wrong one lost.
But is it a question of changing the architecture? Or is it, as I argued here, a question of the social construction of the building, rather than the bricks and mortar?
When does a church become a house, or vice versa? Do you need builders? Or do you need a priest to consecrate/deconsecrate? Think of it as a church and it becomes a church. Think of it as a house and it becomes a house.
The overnight rate would still be affected by monetary policy. It’s endogenous with respect to monetary policy. But a perceived loosening of monetary policy might now mean a rise, not a fall, in the overnight rate.
Put it another way. Ultimately, with a credible monetary policy framing, the only transmission mechanism is via expectations. You see this now; the bank announces a lower interest rate target, and interest rates just drop. That’s far more important than anything the Bank actually does, or would do if they didn’t drop.
Here’s a question: when the average person sees the central bank maintain historically low interest rates, and he sees more and more people borrowing 5x, then 7x, then 10x, and sometimes 12x their annual income to buy houses, then vacation homes, and finally a portfolio of investment properties, all at prices that have doubled or even tripled in less than a decade, is that a “framing issue” ? Could it be that monetary policy is “too loose” because average joe and all the speculators perceive it to be, regardless of any inflation target ? Shouldn’t central banks worry about the amount of debt in the economy, as greater debt levels carry greater risk of insolvency, and require larger swings from consumption to savings to service it when the economy turns ?
Why not do debt targeting? Set interest rates to corral government, business, and household debt into “safe” ranges where systemic risk is low.
“Put it another way. Ultimately, with a credible monetary policy framing, the only transmission mechanism is via expectations.”
I think that you got it right.
rp said: “Why not do debt targeting? Set interest rates to corral government, business, and household debt into “safe” ranges where systemic risk is low.”
BINGO!!!!! IMO, we have a winner!
Here is the fed’s problem. All their cheap labor policies and positive productivity growth have produced price deflation. What if the fed incorrectly believes that they need to produce currency denominated debt to keep the price deflation from happening?
Once again, I’m assuming lower interest rates don’t get excess savers (think Goldman Sachs employees) to spend.
“Put it another way. Ultimately, with a credible monetary policy framing, the only transmission mechanism is via expectations.”
What if expectations and reality are different???
rp said: “Shouldn’t central banks worry about the amount of debt in the economy, as greater debt levels carry greater risk of insolvency, and require larger swings from consumption to savings to service it when the economy turns?”
Think about who benefits from higher and higher currency denominated debt levels if price inflation remains muted.
This is all a bit philosophical for my liking. An individual country should set monetary policy in terms of whatever variable they feel gives them the best chance of hitting their monetary policy goal (ie consumer price inflation, perhaps with some deviations to utilise the temporary influence of monetary policy on real activity), which could be a short term interest rate or an exchange rate if that is not resisted by the other currency area involved. But whatever the control variable, the policymakers’ model of the transmission mechanism that relates it to the goal variable has to be developed objectively in terms of including all known influences on the goal variable, regardless of political expediency, and being honest about the influence that the policymakers actually have.
If you ask me, this is why monetary policy failed. It took insufficient account of asset prices, both in the transmission mechanism (many asset prices are tied to consumer prices, but it may take several years – beyond the typical monetary policy forecast horizon of two or three years – before the tie tightens) and in being unrealistic about the degree and acceptibility of asset price changes associated with the policy framework. A major reason for this is that it was not popular either to take action aimed at braking asset price rises (even if the ultimate aim was consumer prices) or even to give sufficiently explicit warnings about what would happen when the tie tightened. In that sense, interest rate (or exchange rate) targeting was a social, rather than a scientific, construction. And I am afraid that a period of policy erring on the tight side, not the easy side, is necessary to reduce the social influence.
Nick:
Excellent.
Add your pole story.
It isn’t that we have to move the bottom one way to get the top to move the other exactly.
It is we have to move the bottom of the pole relative to where people think it should be, to get the top to really move relative to where people think it will be.
Or something like that.
Bill
“Put it another way. Ultimately, with a credible monetary policy framing, the only transmission mechanism is via expectations. You see this now; the bank announces a lower interest rate target, and interest rates just drop. That’s far more important than anything the Bank actually does, or would do if they didn’t drop.”
Sorry, Nick. You’re cheating here. And the last part is quite wrong.
You’re cheating because this is an instance of operational expectation – not inflation or policy rate expectation. It’s operational expectation because participants know that if they resist the natural market effect of the policy rate announcement, the central bank will simply pump in additional excess reserves to force rates down through the process of banks competing to “get rid” their excess reserves individually – i.e. spend their way out of them by purchasing short term liquid and near risk free assets. They can’t do this collectively of course, so it forces the rate down. But they all know this is an unnecessary alternative to simply going with the flow, so the trading rate drops immediately. Think of it as arbitraging a less efficient and slightly slower working counterfactual – the smart guys who know exactly what must happen in any event arbitrage the dumb guys who forget how the transmission mechanism must work one way or the other. Of course, not too many participants are that dumb, so it works even faster.
So this is an operational expectation that is at least one step removed from or a sub-derivative of your topic of policy rate expectation versus other forms of monetary transmission. Correspondingly, the announcement is not more important than what the Bank would do if trading rates didn’t drop – the latter is in the effect the reason market rates follow the policy rate, as described above. It is inherent in the announcement transmission mechanism, not separate from it, through smart/dumb arbitrage.
The operational mechanism and expectation for the policy rate transmission is separate from the expectation for the policy rate itself. The former is more Chartalist territory; it’s operational and factual; the latter is more monetarist territory. Think of it as separating operational from strategic expectation – operational response to a policy rate announcement versus expectations for the policy rate itself – completely separate analysis.
Winterspeak’s final question on the previous post is relevant here. Banks are not reserve constrained. They are capital constrained. This is ALWAYS the case, not just sometimes. But that doesn’t mean they aren’t reserve motivated. Chartalists get this slightly wrong. If banks are recipients of huge excess reserves provided by the central bank, they will attempt to find alternatives (to receiving interest reserves) that DON’T USE UP CAPITAL – i.e. risk free or near risk free assets such as treasury bills or extremely high quality short term credit (very difficult to find these days). Moreover, ARBITRAGE ensures rates adjust very quickly and any balance sheet expansion due to excess reserves is limited by the speed of the market arbitrage rate adjustment. E.g. if the Fed eliminated the 25 basis point reserve interest compensation, rates on treasury bills and near risk free assets would move so quickly that the banking system balance sheet effect and money supply effect would be quite contained. I’m afraid Scott Sumner gets this point wrong. Any tactical advantage of excess reserve deployment is only temporary due to risk free rate arbitrage; this plays a back seat role to the strategic constraint of capital requirements for any risky lending by banks. This is why the Treasury injected TARP capital into the banks, and why excess reserves are primarily a Fed liability tool for its own balance sheet expansion rather than an intended commercial bank lending motivator.
FYI . . . Warren Mosler’s comments on JKH’s last paragraph:
JKH: Banks are not reserve constrained. They are capital constrained. This is ALWAYS the case, not just sometimes. But that doesn’t mean they aren’t reserve motivated. Chartalists get this slightly wrong. If banks are recipients of huge excess reserves provided by the central bank, they will attempt to find alternatives (to receiving interest reserves) that DON’T USE UP CAPITAL – i.e. risk free or near risk free assets such as treasury bills or extremely high quality short term credit (very difficult to find these days).
Warren: Banks are recipients of excess reserves for their own account only if they sell assets (voluntarily) for any reason. But it can be
presumed a seller sells something voluntarily at a give price because at that price he’d rather have the money than the thing he sold. He also may be doing a ‘swap’ where he then takes those funds and spends them on another financial asset that, at current prices, he’d rather have than the asset he sold. In that case the seller of the second asset sells at a price where he’d rather have the money than the asset he sold, and so on down the line, until, before the Fed window closes at 3pm, prices adjust to where the last seller of assets would rather hold the funds as overnight reserves than the assets he sold. Most of these assets have similar capital requirements, and all contribute to the overall leverage ratio. If that’s too high, banks need to sell assets to pay down debt, but that’s another story.
JKH: Moreover, ARBITRAGE ensures rates adjust very quickly and any balance sheet expansion due to excess reserves is limited by the speed of the market arbitrage rate adjustment.
Warren: Yes, but balance sheet expansion is not a function of excess reserves.
JKH: E.g. if the Fed eliminated the 25 basis point reserve interest compensation, rates on treasury bills and near risk free assets would move so quickly that the banking system balance sheet effect and money supply effect would be quite contained.
Warren: there wouldn’t be any effect per se no matter how quickly or slowly rates adjusted.
JKH: I’m afraid Scott Sumner gets this point wrong. Any tactical advantage of excess reserve deployment is only temporary due to risk free rate arbitrage; this plays a back seat role to the strategic constraint of capital requirements for any risky lending by banks. This is why the Treasury injected TARP capital into the banks,
Warren: The Tarp was nothing more than regulatory forbearance.
JKH: I want to take back the last part of what I wrote, specifically “…or would do if they didn’t drop.” Under a natural interpretation of what I wrote, I was indeed mostly/partly wrong. I will return to this point when I can think up a good analogy, to make it clearer what I was trying to or should have said.
Thanks, Scott.
(Ooh; that was quick. Please note I said SLIGHTLY wrong (IMO).)
Warren’s the expert on the Fed and related things.
Technical point – maybe I’m out of date here, but individual banks (or their non-bank dealer clients) don’t/didn’t necessarily need to sell assets directly to the Bank of Canada in order for the banks to be the recipient of system reserve injections – at least that’s the way it used to work in Canada. The CB could either leave excess reserves created by government expenditure and its credit to the deposit liability side of the commercial banking system, or do foreign exchange swaps with the FX fund – an internal transaction that created excess domestic balances and reserves covertly – away from either the domestic money market or the FX market. Either way, the result was an off-market pro rata injection of excess reserves (pro-rata according to individual bank shares of system and shares of government deposit accounts.) Maybe they’ve changed that. I don’t know. But it offered an operational option to adjust reserves away from otherwise required OMO’s. I’d be surprised if they’ve surrendered that flexibility. And I don’t know if the Fed has it, but Warren obviously knows, and his comment suggests they don’t.
I can’t see much disagreement with Warren’s first point, though.
In any event, agreed the direct sale of assets from a bank to the central bank doesn’t change the size of the system balance sheet. But a sale from a bank customer (i.e. a non-bank dealer) to the central bank does. I presume there are still some non-bank dealers that can deal directly with the Fed, but I’m not up on it.
I disagree with the point that excess reserves can’t drive balance sheet expansion, but only to a very limited degree. That was my point, and not necessarily inconsistent with the fact that banks are not reserve constrained. It just means once in a while they are reserve liberated, but only when it doesn’t conflict with the fact that they are capital constrained – i.e. when they attempt to replace excess reserves with the purchase of risk free assets from the non-bank sector, which expands assets and liabilities of the banking system but doesn’t draw on capital to any significant degree – at least on a risk weighted basis. Here the issue of nominal equity ratios as opposed to risk adjusted ratios becomes constraining – which supports Warren’s point – but there is usually some room to manoeuvre here. If there is no such room, then I would concede Warren’s point totally. Perhaps that’s it.
Agreed the TARP equates to regulatory forbearance, and that regulatory forbearance would arguably have been the cleaner route. But TARP was government supplied bank capital in the absence of equivalent explicit regulatory forbearance.
Hi JKH
Warren’s the one that’s quick . . . I don’t know how many conversations he has going on at once during any given day (I’m assuming a good deal more than I have going on), but it’s amazing how quickly he returns emails.
I don’t really see anything to disagree with in either of your posts. Your point on aggregate balance sheet expansion fits my understanding. Warren may have been thinking of an individual bank.
Also, regarding your original post, my impression is that we’ve always emphasized the arbitrages you speak of there . . . are you seeing it differently?
Regarding the technical point, yes, some dealers are non-banks, some are banks, in the US.
Scott,
I certainly don’t see a disagreement here. I think there’s a marginal effect that I attempted to describe. It’s a slightly different idea than “not reserve constrained”, but more supplementary to it than inconsistent with it. And what I’m talking about would get exhausted by rate arbitrage very quickly. “Slightly wrong” wasn’t quite the way to frame my point.
The most interesting perspective I think would be to simulate a bank market response to a decision by the Fed to eliminate the 25 basis point interest payment on reserves. My point is that there could be some thrashing about marginal activity before reaching interest “equilibrium”, which may have a marginal effect on balance sheets. But this would be very limited, as the arbitrage should be near instantaneous (leaking and front running might well occur, of course). The capital constraint, risk weighted and/or nominal, plus the interest rate arbitrage would full stop out the market and money supply effect very quickly, and the banking system balance sheet would settle in very close to what it is with the current 25 basis point rate. My guess is that the banks might end with a smattering more of treasury bills on their balance sheets, but not much more before those rates settled in at near zero – i.e. slightly closer to zero than where they are today.
Some monetarists regard the fact that the Fed is paying interest on reserves as a huge deal. I believe they’re wrong, and I think you and Warren would agree with that. Yet this monetarist interpretation seems to have gotten more mainstream traction than is warranted – at least when presented as supporting some argument that the Fed is being “too tight”.
P.S.
The banks individually and collectively are well aware of course that the Fed controls the aggregate supply of reserves in the system. And they’re well aware that their individual attempts to substitute interest earning assets for reserves in the kind of scenario I suggested can only succeed in the sense of attempting to push away those indestructible reserves to some other bank. In other words, they’re aware that the collective result is in that sense somewhat counterproductive. In addition to the fact that the rate effect gets arbitraged out very quickly, I believe that a collective mindset awareness actually plays a role in a game of such limited reserve arbitrage opportunity.
JKH:”Yet this monetarist interpretation seems to have gotten more mainstream traction than is warranted – at least when presented as supporting some argument that the Fed is being “too tight”.”
I don’t know it this is really true or not but I certainly hope that it’s not. It’s a horrible and sad day for econ when Sumner is mainstream.
For what it’s worth though, I agree with you that the interest on reserves is no big deal and I think it’s a pretty stupid argument to suppose that if it was eliminated credit would flow like a tidal wave.
RebelEconomist said: “This is all a bit philosophical for my liking. An individual country should set monetary policy in terms of whatever variable they feel gives them the best chance of hitting their monetary policy goal (ie consumer price inflation, perhaps with some deviations to utilise the temporary influence of monetary policy on real activity), which could be a short term interest rate or an exchange rate if that is not resisted by the other currency area involved.”
This is not just for Rebel but everyone. Instead of using a short term interest rate (which I consider manipulating debt levels on the lower and middle class) or an exchange rate to achieve consumer price inflation, why not use real earnings growth for the lower and middle class???
Wouldn’t this allow the lower and middle class to both pay down currency denominated debt and spend?
P.S.
One of the consistent and recurring themes of Chartalism is that it identifies a surprisingly wide scope of things where mainstream economics literally gets things backwards. This recurs throughout the topics of bank credit and money, bank reserves, and government expenditure and “financing”.
Yet another example is the rationale for the Fed’s provision of excess reserves to such an extraordinary degree during the credit crisis.
Mainstream economics tends to believe these reserves are being supplied in an effort to cause the commercial banks to lend.
This is absolutely incorrect.
They are being supplied only as the Fed’s chosen “funding” consequence attributable to its own lending.
Mainstream economics gets the asset-liability dynamic backwards.
The purpose of these extraordinary excess reserve levels is central bank liability driven – not commercial bank asset driven.
Hi JKH
I agree with everything you’ve said in this thread (aggregate balance sheet expansion, arbitrage, interest on rbs). I didn’t mean to appear defensive about the “slightly wrong” comment (I know we appear that way often, and it may be true sometimes as we routinely are misinterpreted even as most everyone is sure we are wrong). I just wanted to see what Warren thought and see where that took us; no worries at all.
Dear Adam . . . I would agree with JKH regarding how interest on rbs has been interpreted by and large (but by no means everyone, or maybe the misinterpretation is mostly by those with the loudest voices). Quite unfortunate, as you said, but that’s what happens when the vast majority of the economics profession believes that the details of monetary operations aren’t something worth understanding.
I was not going to comment on this thread because the argument that interest rate targeting was about a social construction and framing device was so absurd, that Nick could not possibly mean what it seemed he did. And then we got this: “Ultimately, with a credible monetary policy framing, the only transmission mechanism is via expectations.”
Surreal.
My neighbors down the street bought a house they could not afford back in 2006, because they were worried about being priced out forever. They are now underwater, husband has lost his job, wife’s is looking shaky. They are doing their best to stay current on their mortgage payments, and have cut back on everything else so they can sock away every cent in case both end up unemployed. They are in California, so have about 40% of her earnings taken away by the Govt.
I assure you, Nick, their problem is not one of “framing”. It is one of not having enough money to service their debt out of income.
Instead of futzing about with “expectations” via something as operationally impotent and remote as interbank lending rates, you could focus on simply recapitalizing households. Maybe that is too straightforward.
JKH said: “They are being supplied only as the Fed’s chosen “funding” consequence attributable to its own lending.
Mainstream economics gets the asset-liability dynamic backwards.
The purpose of these extraordinary excess reserve levels is central bank liability driven – not commercial bank asset driven.”
Could you expand on that?
I am starting to speculate that there are excess reserve levels because the fed ran out of treasuries to swap with “garbage debt”.
I am also starting to speculate that excess reserve levels are actually future gov’t currency denominated debt that has not been issued yet.
Scott,
I always appreciate your comments.
Nick,
Sorry, didn’t mean to hijack the thread to go tangential; but everything’s connected to everything else.
Too Much Fed:
The Fed created excess reserves through its lending programs. Excess reserves appear as a liability on the Fed balance sheet – “funding” of a sort – except the Fed created its own funding in the process of its lending. That’s the extent of the economic role of the now $ 1 trillion in excess reserves. From the commercial bank perspective, these reserves are essentially dormant, risk free assets, earning something like a risk free rate. They play no role in commercial bank credit expansion. That’s the role of bank capital and bank lending.
You’re right about the Fed running out of treasuries. That’s really equivalent to saying they reached a point where they were forced to expand their balance sheet in order to lend/acquire any more assets (once they ran out of treasuries).
The Fed will eventually unwind the excess reserve position. It will do that in response to unwinding its unconventional lending programs. Some people think that will take a long time.
JKH: think of police doing crowd control, or traffic control. For nearly all people, nearly all of the time, all the police have to do, if they have sufficient credibility, is to tell people to move, and they do. Yes this “request” is backed up by people’s expectations of what the police would do if someone didn’t move as requested. But we very rarely see the police actually having to do anything. It’s all a counterfactual conditional. Again this assumes “credibility”, which means that the police control the framing, so everyone expects everyone else to do what the police ask them to do. So if one person considers disobeying, he expects to be the only person doing so, and easily handled by a couple of cops. Of course, this analogy usually breaks down at the point where the BoC can print as many “reserves” as it likes, if the crowd gets out of control. But excess reserves don’t seem to do much today.
Winterspeak: “I was not going to comment on this thread because the argument that interest rate targeting was about a social construction and framing device was so absurd, that Nick could not possibly mean what it seemed he did.”
Thank God! Someone actually understood what I was saying ;). I did say it was “weird”, didn’t I?
Bill: Thanks! Yes, next post on the pole analogy.
Nick,
“think of police doing crowd control”
I like the analogy!
Until this:
“Of course, this analogy usually breaks down at the point where the BoC can print as many “reserves” as it likes, if the crowd gets out of control. But excess reserves don’t seem to do much today.”
The analogy doesn’t break down. You may think it breaks down, because you may be confusing today’s abnormal reserve environment with a normal one.
An environment of “normal” excess reserves is one in which the central bank doesn’t have to pay interest on reserves in order to “defend” the policy rate at the lower bound. The lower bound is defended normally by the degree of restriction on excess reserves. In such an environment, flooding the system with excess reserves is guaranteed to drive rates lower and flatten any resistors in the process.
Today’s environment is one of “abnormal” excess reserves, in which the central bank has “flooded” the system permanently with excess reserves, while defending the lower bound by paying interest on reserves. Through the interest rate arbitrage process discussed earlier in this thread, the central bank has absolutely no problem in guiding short term market rates lower when it drops both its policy rate and the rate it pays on reserves.
And there’s no problem in increasing rates either. The central bank squeezes excess reserves in the normal environment, and just pays a higher rate of interest in the abnormal one.
You’ve underestimated the power of your analogy!
Just remember, this is an operational analogy about expectations in the market RESPONSE TO ACTUAL CHANGES in the policy rate, not expectations about the path of future policy rate changes or other variables such as inflation.
And of course, if the police order the crowd to do anything unreasonable, they can be readily overwhelmed…….a very good analogy indeed!
“if the police order the crowd to do anything unreasonable, they can be readily overwhelmed..”
BTW, do you allow your cops in the UK to carry guns yet?
🙂
JKH said: “The Fed created excess reserves through its lending programs. Excess reserves appear as a liability on the Fed balance sheet – “funding” of a sort – except the Fed created its own funding in the process of its lending. That’s the extent of the economic role of the now $ 1 trillion in excess reserves. From the commercial bank perspective, these reserves are essentially dormant, risk free assets, earning something like a risk free rate. They play no role in commercial bank credit expansion. That’s the role of bank capital and bank lending.”
Did the fed use excess reserves to overpay for “garbage debt” to protect bank capital (bailout their banking buddies)?
Why excess reserves and not currency (aren’t they both liabilities of the fed)?
I still want to consider the idea that excess reserve levels are actually future gov’t currency denominated debt that has not been issued yet.
Here is one last thing. Does paying interest on reserves transfer the interest payment from the taxpayers to the banks?
NICK: You are an absolute champ. I am just so humbled and elated by the grace you show in your comments section. I am being 100% serious.
Too Much Fed: Paying interest on reserves does fund banks. To the extent that it is done instead of giving money directly to taxpayers, it is a transfer payment. In general, though, it is not correct to think that taxpayers fund Government spending. It is the other way around, Government deficit spending gives taxpayers the money they need to pay taxes (and net save).
Too Much Fed: To further elaborate Winterspeak’s point on interest on reserves balances, if anything, if the qty of reserve balances circulating with interest payment is greater than without (which very much IS the case now, obviously), then it results in a net payment TO the Tsy, not FROM the Tsy. This is because there are 2 possible sources for the additional reserve balances. The first is an open market operation in which the Fed replaces the non-govt sector’s holdings of a Tsy earning market rates with reserve balances earning 0.25%. The second is the Fed is lending at a rate higher than the rate paid on reserve balances (which it has done in every instance). In either case, the net cash flow to the Tsy INCREASES.
So are you claiming that the prevailing rate with respect to the natural-rate is irrelevant?
Too Much Fed,
An alternate way of saying Scott’s viewpoint is that the Fed is highly leveraged (short short term, long long term) and hence more profitable than less leverage. The profits are given to the Treasury. Of course if the Fed decides to sell the Treasuries it holds, it can affect the price and in the end could make a loss but that will be damaging and the Fed will never do that.
Jon: “So are you claiming that the prevailing rate with respect to the natural-rate is irrelevant?”
That is not an easy question to answer. “No” is the short answer, but it’s not very helpful. Let me try another way.
Given the current framing of monetary policy, if you asked someone what they expect the stance of monetary policy to be over the next year, they would give an answer in terms of what they expect the time path of i to be. They would also have expectations of the time path of other variables, a vector X, over the same time-period, conditional on their expectation for i.
Suppose monetary policy were re-framed in terms of some other element of X, namely x. People would then describe the stance of monetary policy in terms of their expectations of the time path of x. They would also have expectations of the time path of the other variables in the vector X, and of the time path of i, over the same time-period, conditional on their expectation for x.
But an announced “loosening” of monetary policy would mean a reduction of i relative to previous expectations, under the first framing, but might mean an increase in x relative to previous expectations under the second framing. (Let x be stock prices, for example).
A “loosening” of monetary policy under the first framing would see i fall and x rise, so i and x move in opposite directions. But a “loosening” of monetary policy under the second framing might see x rise and i rise too, so i and x move in the same direction.
Now there is a lower bound on i (namely i=0) that is binding at present. But there is no upper bound on x. So the Bank of Canada cannot credibly promise a loosening of monetary policy under the first framing (because that means violating the lower bound), but can credibly promise a loosening of monetary policy under the second framing.
I need to try to construct a formal model to clarify this.
British police are armed more often these days, JKH, but they mainly shoot innocent Brazilians.
Scott Fullwiler said: “This is because there are 2 possible sources for the additional reserve balances. The first is an open market operation in which the Fed replaces the non-govt sector’s holdings of a Tsy earning market rates with reserve balances earning 0.25%. The second is the Fed is lending at a rate higher than the rate paid on reserve balances (which it has done in every instance). In either case, the net cash flow to the Tsy INCREASES.”
So what did the fed use to buy the non-treasury debt?
Plus, aren’t the banks getting an extra 0.25% that they would not have gotten in the past? I also read somewhere that raising the reserve requirement would accomplish the same thing as paying interest on reserves but would not be as profitable for the banks.
Ramanan said: “An alternate way of saying Scott’s viewpoint is that the Fed is highly leveraged (short short term, long long term) and hence more profitable than less leverage.”
Can you rephrase/expand on that for me?
Good questions, Too Much Fed
1. Regarding non-Tsy debt, that’s much like lending to the non-govt sector and likewise earns a higher return than paid on rbs. Sorry for the confusion.
2. Yes, banks are receiving an extra 0.25%. BUT, the Fed could not keep $800 billion in excess balances (or whatever the current amount is) and still achieve an overnight rate above zero, which it currently wants. So, the choice is really b/n paying no interest on a very small qty of excess balances AND equivalently reduced qty of assets earning a positive net return, or the current regime with a LOT of assets earning more than the rate paid on rbs. Given these choices, the current approach is reducing the govt’s deficit by more than would otherwise be the case, albeit quite modestly.
3. Ramanan’s point is an excellent one. It’s essentially what I’ve said in point 2. Just think of liabilities = leverage, and rbs are liabilities.
Nick, your not really helping anything with your post here, you can’t just choose anything to be x. There has to be a transmission mechanism.
The central bank can set i (short term inter-bank lending) or the exchange rate but these things both work because they are in price mechanisms that enter agents maximization problems and so changing one of them directly changes behaviour.
They also work because there is an arbitrage relationship that allows the CB some control of them. In normal times the CB can control i, now the problem is that i can’t be made negative but the CB can reduce the exchange rate. But of course there are limits to CB control of the exchange rate as well, it can’t raise it as high as it pleases.
The problem though has nothing to do with social framing, it has to do with the fact that most things (like inflation or nominal expenditures) are not actually under the control of the Fed, there is no arbitrage relationship. Only Scott Sumner is dumb enough to think the fed can actually control NGDP growth or inflation. That’s why we usually think of the fed choosing i to target inflation. For the fed i is a choice variable, inflation is not.
It all goes back to the discussion on the other threads, the CB can control i through an arbitrage relationship and i has the virtue of a direct transmission mechanism. It’s these two properties of i that are the reason why monetery policy is operated through i, it’s not an arbitrary social construct and any properly specified formal model will tell you that.
Adam P: I’m just getting my head around a small model. There are two assets: bonds and shares. A bond pays you $1 forever (or maybe it’s a bill that pays you $1 next period). A share pays you 1% of a companies profits. Companies’ profits depend on GDP. So lets simplify and say a share (or a share mutual fund) promises to pay you 1% of GPD forever (or it might be analogous to a bill, and pay you 1% of GDP next period).
Assume people are risk-neutral, so the real expected rates of returns on bonds and shares must be the same, by arbitrage.
The central bank can use the price of bonds (bills) as an instrument (that’s simply the inverse of the nominal interest rate). The central bank can also use the price of shares as an instrument. (I would say this is a framing choice, not a real choice, but never mind). But there’s exactly the same transmission mechanism in either case (via arbitrage).
But my shares (at least in their one period “bills” variant) are exactly the same as Scott Summner’s NGDP futures contracts.
Continued: And if the central bank was using the share price as an instrument, an announced “loosening” of current and future monetary policy would mean raising the price of shares at an increasing rate. The only equilibrium would be where expected Py grows at an increasing rate too. The real rate of interest may rise or fall (depending on Euler equation effects of future y, and on the slope of the future SRAS curve), but expected inflation rises, and so i can rise. And it can rise off the lower bound.
Okay Nick. I think what you’re saying is that interest-rate targeting is like epicycles.
We all believe in this Earth Centered model of the solar system. But it appears to require a very nuanced model to represent all of the behaviors seen, e.g., retrograde motion. So our experts toil and toil to provide a complex model (the epicycles).
But if we just change our perspective, and write down a sun-centered model of the solar system, the behavior of the planets is easy to describe.
So your claim then is that interest-rate targeting is our modern epicycles.
Jon: that’s closer to what I’m saying, but not exactly right. We could view it either as Earth centered or as sun-centered, as you say. (In the long run, it’s actually easier to view it as money centered, rather than interest rate centered, but that’s not my main point).
Where your analogy breaks down is that it’s as if the motion of the planets depends on expectations of monetary policy, and expectations are different if the central bank can communicate what it is going to do in terms of an earth-centered policy than in terms of a sun-centered policy. There are things the Bank can say in earth-language than are unsayable in sun-language.
Adam P: but I accept your main point though. The Bank cannot choose anything whatever to be its instrument. It cannot choose an exogenous variable (obviously), and it cannot choose anything as an instrument that it can only influence weakly, and with a long lag (otherwise an inflation-targeting central bank could just choose inflation itself as the instrument, and we lose the distinction between target and instrument).
The instrument must be some variable that is more or less observable in real time (or close to it), and responds (or can respond) quickly to what the Bank does (or can do).
I value your comments on this, by the way. I am not saying “You are flat out obviously wrong!”. You articulate my own inner doubts about what I am saying. I have a vision of how monetary policy framed in terms of some other instrument can do things, via influencing expectations of the time-path of that other instrument, that cannot be done with an interest rate instrument.
But I think you can see that if the Bank could buy and sell the CPI bundle directly (ignore practical problems like storing haircuts in the basement of the Bank of Canada), then the Bank of Canada could use CPI as an instrument, and by a credible announcement of a more steeply rising path of CPI, which would constitute a loosening of monetary policy, the nominal interest rate would rise off the zero floor. (This sounds like fiscal policy, but not if haircut market interventions were offset by OMO, so the quantity of haircuts stored in the basement didn’t get too large.)
Still trying to think through a simple economic model to articulate my vision better.
Adam P. Yes, and the instrument must have some sort of connection with the rest of the economic system. If the Bank of Canada took over Statistics Canada, it could re-write the CPI numbers published by Stats Can to report 2% inflation. But that doesn’t count.
But I think that share prices meet all these criteria. They are as connected to IS and AD as much as bond prices are; they are observable in real time; they adjust as quickly as bond prices; the Bank of Canada can influence share prices as easily as bond prices; it can buy or sell shares as easily as bonds; it can announce a time-path for share prices as easily and as credibly as it can announce a time-path for bond prices.
“It all goes back to the discussion on the other threads, the CB can control i through an arbitrage relationship and i has the virtue of a direct transmission mechanism. It’s these two properties of i that are the reason why monetery policy is operated through i, it’s not an arbitrary social construct…”
Good, Adam.
“and by a credible announcement of a more steeply rising path of CPI, which would constitute a loosening of monetary policy, the nominal interest rate would rise off the zero floor. (This sounds like fiscal policy, but not if haircut market interventions were offset by OMO, so the quantity of haircuts stored in the basement didn’t get too large.)”
Your OMO mechanism is opposite to Scott Sumner’s on NGDP futures, Nick. Why? The targeting mechanism is parallel.
And neither you or Scott has addressed the fact that monetary architecture must change if you expect the short rate to rise when the central bank injects reserves under OMO.
Nick said: “But I think that share prices meet all these criteria. They are as connected to IS and AD as much as bond prices are; they are observable in real time; they adjust as quickly as bond prices; the Bank of Canada can influence share prices as easily as bond prices; it can buy or sell shares as easily as bonds; it can announce a time-path for share prices as easily and as credibly as it can announce a time-path for bond prices.”
I think you are ignoring wealth/income inequality.
Why not announce they are targeting wages to be both positive in real and nominal terms?
What do you think the chances are of central banks ever doing that? IMO, they are doing just the opposite.
Scott Fullwiler said: “The second is the Fed is lending at a rate higher than the rate paid on reserve balances (which it has done in every instance).”
Can you expand on that one?
And, “2. Yes, banks are receiving an extra 0.25%. BUT, the Fed could not keep $800 billion in excess balances (or whatever the current amount is) and still achieve an overnight rate above zero, which it currently wants.”
I think they should try something else to raise the overnight rate.
And, “So, the choice is really b/n paying no interest on a very small qty of excess balances AND equivalently reduced qty of assets earning a positive net return, or the current regime with a LOT of assets earning more than the rate paid on rbs. Given these choices, the current approach is reducing the govt’s deficit by more than would otherwise be the case, albeit quite modestly.”
At first glance, I’m not getting that one.
Nick, your mistake is here when you say “it can announce a time-path for share prices as easily and as credibly as it can announce a time-path for bond prices.”
The CB CAN’T announce a time path for share prices (forgetting the question of which shares) or bond prices (which bond?). There are sever limits CB ability to control the price of even a single share, for example suppose the Fed wanted to raise the price of GE stock to $20. GE closed yesterday at 15.75, buying GE shares could bid up the price but pushing the stock substantially above the equilibrium price would require buying the whole company.
This is both horrible policy and would not have the desired real effect, GE would not be encouraged to invest more because their private cost of capital has gone down. GE would NOT be able to issue more shares to raise capital at $20 a share unless the Fed bought the new shares as well, but if the Fed wants to directly fund investment by GE then it could just loan the money at low or zero interest, but now we’ve totally elimnated the market directed allocation of capital. It’s not monetary policy anymore and it’s a really bad idea.
YOU CAN’T HAVE THE FED CONTROLLING SHARE PRICES AND MAINTAIN THAT THE USUAL TRANSMISSION MECHANISM FROM SHARE PRICES TO REAL ACTIVITY STILL EXISTS!
Anyone who has thought about this agrees that expectations are vital to the transmission mechanism’s operations but expectations, prices and resulting real activity are all jointly determined, JOINTLY DETERMINED. We use the language of simple causaul chains, expectations to prices to activity, but that is within a general equilibrium system where in fact all three are jointly and endogenously determined. None causes the other.
Once you start having the fed directly allocate capital the whole system changes. This is one of the reasons, though not the most important, that Sumner’s NGDP futures scheme is so incredibly stupid (I kinda wish he understood basic macro theory well enough to see this).
The CB controls an overnight INTER-BANK lending rate and/or exchange rate through explicit arbitrage relationships (and even then tat control is not absolute) and THE CURRENT RECESSION IS IN NO SENSE A FAILURE OF INTEREST RATE TARGETING AS THE MONETARY POLICY TOOL.