Some people argue about whether the macroeconomy is inherently stable or unstable. I don't think that's a very useful question. Because…..it depends. And one of the things it depends on is monetary policy. And that is a useful discussion to have, because we can actually do something about monetary policy.
Don't adopt a monetary policy that would make an equilibrium
unstable. Because if you miss that equilibrium by even the tiniest
amount (which you almost certainly will) the economy will move further
and further away from equilibrium.
I thank Steve Keen for (inadvertently) reminding me to do something that George Selgin had wanted economists like me to do. It's not often any of us get the chance to write a post that will make both Steve and George happy at the same time. (Well, Steve should be happy, because I'm talking about unstable equilibria, but you never can tell.)
There's nothing new here. Just the old stuff, but perhaps told in a slightly different way.
1. Don't use monetary policy to target unemployment.
Targeting a real variable like unemployment sounds like a really good idea. Because it's real variables like the unemployment rate, and real income, that really matter to people. Much more than nominal variables like the price level, or inflation rate. So why not tell the central bank to target unemployment?
Like most economists, I think that if you loosen monetary policy then inflation goes up and unemployment goes down. At least, temporarily. There's some sort of short run trade-off. But I'm less sure about the long run. If you permanently loosen monetary policy, so that inflation is permanently higher, will unemployment go down, go up, or stay the same?
It is easy to build a theoretical model in which there is no long run trade-off. The Long Run Phillips Curve is vertical. Now any macroeconomist with any ingenuity could tweak that model to make the Long Run Phillips Curve slope in either direction. But I wouldn't trust those tweaks in either direction. The data don't help much either. When I look at the data, I think I see that countries with high average inflation tend to have high average unemployment rates. Which means the Long Run Phillips Curve slopes the "wrong" way. But I don't totally trust my lying eyes. Because it might just be that screwed up countries tend to get everything wrong, and it's not that their loose monetary policy is causing high inflation and high unemployment. It might just be something else that is screwed up in those countries, that is causing both the high unemployment and the loose monetary policy.
But if you want monetary policy to target unemployment, you had better be very confident that the Long Run Phillips Curve slopes the "right" way. Because if it slopes the wrong way, you have got an unstable equilibrium. If you miss the equilibrium by just the tiniest bit, and set a slightly too low target for the unemployment rate, you loosen monetary policy to bring unemployment down, inflation will rise, the unemployment rate will eventually rise too, so you loosen monetary policy further, and so on, getting further and further way from the equilibrium (assuming it even exists).
Even if the Long Run Phillips Curve does slope the right way, and a stable equilibrium exists, it might be very different from what regular folk might think of as stable. A Russian doll is stable, but it sways back and forth a lot in a gusty wind. If the Long Run Phillips Curve slopes the "right" way, but is very steep, small shifts in the Phillips Curve will cause very big fluctuations in inflation if you target unemployment.
Most economists say the Long Run Phillips Curve is vertical, and so monetary policy can't target unemployment in the long run. They don't really mean that. They mean they don't really know which way it slopes, and vertical is probably a good approximation. And they don't want monetary policy to target unemployment, because the risk of the equilibrium being unstable, even if it exists, is far too high for comfort. And even if it is technically stable, it will move around a lot, so it won't look "stable" in the ordinary sense of the word.
As older readers will remember, targeting unemployment has been tried before. And it took a couple of decades to recover from the experiment and get both inflation and unemployment back down again.
Lets not go there again. It's a really bad idea.
2. Don't use monetary policy to target a nominal interest rate.
But that's what the Bank of Canada does, isn't it? Well, yes and no. Mostly no. What the Bank of Canada does is target 2% inflation. It only targets a nominal interest rate for a very short time period, of about 6 weeks. Every 6 weeks it adjusts the interest rate target, as needed, to try to keep inflation at the 2% target.
There (probably) exists a time-path for the nominal rate of interest that is compatible with an equilibrium in which inflation stays at (roughly) 2%. But:
We don't know what it is. It won't be a constant. And, most importantly, it will (almost certainly) be an unstable equilibrium. If we miss that equilibrium, by even the smallest amount, the economy will move further and further away from the equilibrium.
Suppose the central bank sets the rate of interest too low. Demand will be too high, and inflation will start to rise, and keep on rising.
It gets worse. As inflation rises, expected inflation rises too, the real interest rate falls for any given nominal rate, and so demand increases still further.
None of this is news to New Keynesian macroeconomists. They know you will get an unstable equilibrium if the central bank holds the nominal interest rate constant. So they insist that the central bank must adjust the nominal interest rate quickly enough and by a large enough amount to try to turn an unstable equilibrium into a stable equilibrium. And if people are confident that they can and will do this, expected inflation will stay anchored at the 2% target, which helps prevent one of the destabilising forces.
Whether that feedback rule for the nominal interest rate will always operate quickly enough and strongly enough and be credible enough to convert an unstable equilibrium into a stable equilibrium…..is another question. The answer looks a lot different at the Zero Lower Bound than it did before. Maybe it's time to stop using even very short term interest rate targets?
Worthwhile Canadian Initiative 
Yippee! I mean, thanks, Nick.
Thanks George. You were quite right to say we should say this. I hope Steve likes it too.
Nick Rowe,
Does it ever depress you that your post, which I imagine will surprise all too many people, is basically a very good summary of Milton Friedman’s “The Role of Monetary Policy” (except with the order of interest rates and unemployment reversed) now over 44 years old?
If unstable equilibria don’t occupy the mind of mainstream economists, then presumably mainstream economists must think that Friedman ’68 is a very weird paper. Actually, given the (total?) lack of maths in Friedman ’68 & the fact that it was in the AEA, it SHOULD be a bit weird for those only familiar with the modern AEA.
Milton Friedman: heterodox economist?
Hey Nick, so what I read above is that implicitly the real economy, the barter economy, is really really stable: you need to push really really hard and keep on pushing to change a real variable to keep it where it us until it breaks.
The unemployment rate is quite real; the nominal interest rate (i) is part real (r), and part nominal (pi) or i ~= r + pi. The lesson is: don’t touch those real variables?
W Peden. Well, it might depress me. But then I think of the Carleton University motto: “Ours the task eternal”, which provides solace. There will always be jobs for teachers.
The only case that comes to my mind where a “mainstream” (though, not really mainstream) economist forgot about the unstable equilibrium stuff that’s clear in Friedman ’68 was the infamous Kocherlakota episode. (And he has so much improved recently, it proves there’s always reason to hope).
It’s usually an implicit understanding. We don’t talk about it much, because we know not to go there.
Martin: maybe. But a barter economy would be so very different in so many ways, even if you did assume some hypothetical zero transactions cost economy. But you could still get instability in a barter economy. If people had really stupid expectations (they might), or if fiscal policy were really stupid. Possibly other cases too.
But yes, the main lesson is indeed; don’t use a nominal variable (monetary policy) to try to target a real variable. (Interest rates are actually a bit weird, on the nominal/real distinction. Because all interest rates, both nominal and real, have the units 1/time. And 1/time is real units. True nominal variables have $ in the units. I got my head around this once, but now I have forgotten what I figured out.)
Yes, you’re right about barter. I first read you as a bit of Wicksell, but then I instinctively switched back to Walras + Money with non-neutrality in the short run, and neutrality in the longer run.
Hence why I thought of barter; I have this incessant need of fitting square pegs into round holes. And from your response, I am guessing that I am wrong to read you as simply Walras + money?
Don Patinkin’s Money Interest and prices is about as close as you can get to Walras + money. You could do a lot worse than walras + money. But I don’t think it’s really me.
Nick,
Bog standard NK economics right until the penultimate sentence. But “Maybe it’s time to stop using even very short term interest rate targets?” is a non-sequitur. I’ve been reading you for long enough to know what you’re probably on about, but still… what are you on about?
K: I thought it was a sequitur myself? Ditch the short term interest rate target, and replace it with something with $ in the units. NGDP futures market, TSX-300 index, commodity bundle futures index? Dunno.
Ditch the short term interest rate target, and replace it with something with $ in the units.
So, we should close the discount window and shutdown Fedwire? Just walk away from helping banks cover intra-day and overnight funding shortfalls? Let the free market do it?
If we don’t, then how much do we charge banks who are short of short term funds? And isn’t that our target?
rsj: it would probably be best to keep the discount window open at a floating penalty rate for the lender of last resort function.
Nick,
How is that rate determined? If it is floating, then what other metric sets it?
rsj: you want me to convert Bagehot into some simple formula? Other people could give a better answer to that question than I could. Some rate, such that banks would never use it to any significant extent, if at all, in normal times, but would use it freely in a crisis. Perhaps no such formula can be found, because the lender of last resort function is inherently a judgement call, if risk is involved, which it is.
Not to interfere with rsj’s line of questioning (which I think is sensible)… lets say the cb stops interfering with the market for reserves (you’d have to have a perfect settlement process every night) . Now you no longer have an instrument for the cb to effect policy. So it’s just another loud mouthed think tank. Targets are great, but Chuck needs weapons. The interbank rate is a weapon, an NGDP futures instrument could be a weapon. An NGDP target is not a weapon. Chuck must say how he intends to use his weapons in order to effect his targets.
Some rate, such that banks would never use it to any significant extent, if at all, in normal times, but would use it freely in a crisis
Nick, it doesn’t work like that. We have a credit based money system in part because are allowed to lend first and obtain reserves later.
Now, I understand that you do not believe that, that banks need reserves prior to lending, etc. But in our world, there is a payment system in which the CB, as a matter of course, lends billions of dollars per hour to the banking system in order to keep this whole thing working. These are not overnight loans, or emergency loans that we can spend a lot of time thinking about, but instantaneous intra-day loans necessary to keep everything working.
And banks need to know how much they will be charged in order to conduct their normal affairs and keep this credit-based demand-determined endogenous-money game going.
See, for example, http://www.newyorkfed.org/research/epr/08v14n2/0809bech.pdf
K: central banks can buy and sell government bonds, or gold, or other financial assets. Buying and selling very short term IOUs from commercial banks (overnight lending) isn’t the only thing they can do.
rsj: the commercial banks settle their own daily transactions on the books of the Bank of Canada, and the whole balance nets to approximately zero every night. Nearly all overnight lending is between the banks themselves. Let that rate float. Target something else. We haven’t always done it this way. We don’t always have to do it this way.
rsj: “Now, I understand that you do not believe that, that banks need reserves prior to lending, etc.”
No, you don’t understand that.
rsj,
If the system settles perfectly, as it does in Canada most every night, no bank ends up with a net reserve position overnight. So there is no cost of reserves. (The fact that the interbank policy rate still is the anchor for all rates in the economy is critical, but the mechanism unfortunately is really badly understood by most macro economists). The point is I don’t think Nick needs to buy into the nitty-gritty facts of the existing monetary system. But he does need to propose some policy under which the CB proposes to exchange/lend its reserves/currency and he needs to tell us the basic rules of the settlements process as he envisions it. Without that the CB really is nothing but yet another think tank.
Let me clarify: “no bank ends up with a net reserve position with the Bank of Canada overnight”. Of course, banks have reserve debits and credits between them.
Nick: “central banks can buy and sell government bonds, or gold, or other financial assets.”
Agreed. But if they do that they have to have a policy on IOR. If that policy is IOR=0, then the interbank rate will be zero.
The details of how the Large-Value Transfer System works, which is what Nick and K are talking about, can be found in a detailed pdf-format book on the Bank of Canada’s website.
Quick and dirty: The Bank of Canada sets a target Overnight Lending Rate and a target Overnight Deposit Rate. If the external borrowing rate is below the deposit rate, there is an instant, large and absurdly obvious arbitrage opportunity. The market will take advantage of that and the abnormality is corrected. The flip side is if the lending rate is below the general overnight market deposit rate; the settling banks will engage in a carry trade by borrowing from the Bank of Canada and depositing it in the overnight market until the abornormality is arbitraged away.
These arbitrage features are there by design, they are the teeth of monetary policy. It’s simple, robust and you have to have a mighty large crisis to break the system.
The interbank overnight rate is the ultimate short rate, the risk-free rate. All other rates, for mortgages and GIC’s, are based on this rate with additional basis points for additional term risk. In Canada mortgage “terms” are for six months to five years; the mortgage is rolled over many times under a conventional 25-year amortization. The amortization is not the term.
Nick,
You are basically saying that central banks should not be market makers for bank reserves. But that is the foremost role of central banks. It’s why they are called reserve banks. And the market maker sets the price.
Banks are market makers for money for the non-financial sector. But banks themselves need a market maker for each other’s liabilities. Historically, private reserve banks sprang up to play this role. But even then, the price of reserves reflected the plumbing — e.g. shortages of reserves among certain banks — and not economic conditions in the non-financial sector. This free banking era was financially volatile. To settle that volatility, we created government reserve banks that made markets in reserves. They advertised the price, and they served the banks that made markets to households.
Now, in a Walrassian model, there is a fairy that makes all markets simultaneously. The auctioneer announces the price. With such a fairy, there is no need for a central bank, but also no need for banks. And no need for money. So you create a fictional need for money by sticking it in the utility function, to balance the fiction of the global market maker, and that frees up reserve banks to do other things than make markets in reserves.
But in our world without fairies, central banks will always first and foremost make markets in reserves, meaning they will conduct policy by setting the price of reserves.
This is actually a great debate, because it completely captures the endless disputes about differences between reserves and deposits, about why there is a demand for deposits separate from a demand for reserves, and about why we need both banks to make markets in deposits as well as central banks to make markets in reserves.
And then there is the meta criticism that if you get the institutions wrong, your policy advise (e.g. stop making markets in reserves) will be all wrong.
Hmmmm. Why do I get the feeling that people don’t want to talk about unstable equilibria?
Ooooo, those scary hyperinflation days of the 80s and 90s. Not that we should repeat the 70s but it isn’t inflation that is scary but accelerating inflation in an environment unused to it. I have my doubts we want stable inflation though. Rather we may want an inflation that rises gently over the cycle destroying the debt created in the prior recession to make fighting the next recession easier and reduce the likelihood of a Minsky moment, then slowing inflation as real debt falls. Just a guess though.
Hmmmm. Why do I get the feeling that people don’t want to talk about unstable equilibria?
We are talking about unstable equilibria by attacking the model. Attack the model and you attack the conclusion. Model’s don’t need to be perfect, but getting the basic and most important role of the central bank right is a requirement for a model that tries to predict the effects of CB policies, no?
But if you want to talk about unstable equilibrium, let’s talk about the greatdestabilization when we dropped the ball on fiscal policy and focused solely on monetary policy to stabilize aggregate demand. The consequences of this policy were much more devastating on a global GDP loss or global employment loss basis than the stagflation period at the end of the previous regime. It’s an order of magnitude worse blow up, so I’m not sure how you can argue that one form of policy is stable and the other is not. Canada has sailed by relatively well, but let’s see how well it does when the housing bubble bursts.
If you want to talk about unstable equilibria, then explain why you need to cut rates by more to speed up the economy than you need to raise rates to slow it down by an equivalent amount. How is that a stable equilibrium? How are stagnant wages and a growing compensation/productivity gap stable? How is an economy based on asset price appreciation stable?
I will take some inflation instability any day, provided that male long term unemployment falls to the level that it was during the bad old years of stagflation, or that we get some stability in median wage shares.
Sorry about the typos. Those links should be separated by a space. And the contraction. Ugh, it’s late here. I will resume battle tomorrow!
Nick
I think that NAIRU and the Wicksellian natural rate are both good definitions of macroeconomic equilibrium (in a Keynesian-Wicksellian sense, they’re perhaps isomorphic) and that it may be difficult, if not impossible for a monetary authority to hit either of them. I think that targeting the NAIRU equlibrium is unstable because while targeting the Wicksellian one is simply difficult, not unstable. While I’m not in favour of targeting asset markets, I think an NGDP target makes a lot of sense, esp. if you happen to choose one which is sufficiently above 0%, sufficiently above the typical ‘natural’ short rate and sufficiently below the typical ‘natural’ cost of capital. 5% is a good first approximation.
How to go about targeting the 5%? I’m not quite sure. I would probably recommend some combination of IoR and helicopter drops/ mop-ups in a way that responds to ‘the forecast’. How to develop the forecast? Again I’m not quite sure. But there cold be several ways, including simply taking a poll. But in general, I’d not target asset markets.
To K’s and rsj’s line of questioning, you’re actually quite safe. Monetary policy and liquidity policy are separable. ffr = ior + svr (scarcity value of reserves). Even if you conceive of monetary policy as setting ‘the’ interest rate, this is the ior and setting this can be taken out of the central bank (even if it is the central bank that pays the interest and holds the reserves). An open market committee could do it. The repo market could do it.
The channel in which the ffr operates – the bounds of svr given by 0 and the discount window rate – is then free to be set by the lender of last resort/ discount window operator. The authority that sets the svr has the narrower view of financial stability and the payments system. It does not need to bother itself with hitting macroeconomic inter-temporal sweet spots.
The monopoly issuer of reserves and the decipherer of the numeraire rate of the economy do not need to be the same. The monopoly issuer only controls svr. The Friedman rule sets svr equal to zero, but for financial stability and ‘liquidity as a put option’ reasons, this marginal cost pricing of reserves need not be followed. In systems where the ior traditionally been 0, the analysis has been confused because it looks like the ffr = svr but given how most of these systems operate at ‘peak liquidity’ it’s better to look at them as ffr = ior and the svr = 0.
Lest anyone think I’m talking crazy stuff, I’m basically invoking Charles Goodhart (and Willem Buiter) nearly idea for idea.
Monetary policy is bonkers. Period. Full stop. At least its nonsensical to use it without corresponding fiscal measures, and for the following reasons.
1. It’s distortionary. That is boosting an economy just via interest rates (i.e. borrowing) makes as much sense as boosting an economy just via people with brown hair.
2. QE benefits the rich far more than the poor.
3. The idea that there is a relationship between interest rates and the actual availability of credit is a joke.
4. Low interest rates promote bubbles.
5. At the start of a recession, extra borrowing and investment is exactly what is NOT NEEDED because there is a surplus of unused capital equipment.
6. The effectiveness of interest rate adjustments is hindered by movements of foreign currency into and out of the relevant country, e.g. high interest rates designed to damp demand attracts foreign hot money, which negates the intended effect.
7. Prior to the crunch people were borrowing like crazy to fund property speculation. In a free market, interest rates would have risen which would have choked off some of the speculation. But central banks in their infinite stupidity use interest rates to control inflation, and inflation wasn’t too bad prior to the crunch because the extra resources going to speculation were resources REMOVED FROM elsewhere. Net effect on inflation was almost nil.
I.e. using interest rates of regulate economies is a stroke of genius, I don’t think.
“Let that rate float”
The CB determines the amount of reserve balances through its own balance sheet management. And unless it deliberately sets the price of reserves, the overnight interest rate will either be zero (reserves in excess of system settlement requirements, causing banks to drive the rate to zero through competition to invest excess funds in things like treasury bills), or potentially unlimited on the upside (reserves short of system settlement requirements).
In the latter case, there has to be a rule and a pricing for overdrafts (i.e. overnight loans from the central bank), since at least one bank is short of funds by assumption. Otherwise, there is no system at all – it just collapses into chaos, because there is no pricing for settlement funds, and there is no pricing discipline motivating a system settlement process that intends to avoid more expensive central bank funding in the normal course.
Administered pricing of reserve balances is a necessary consequence of the institutional design of the system, with commercial banks settling on the books of the central bank.
So there must be pricing of settlement funds and the central bank must set it. And so the central bank sets the interest rate. It has to. It has no choice.
(The only realistic example I’ve seen so far without this is the case of a currency board, but this is an intricately interlinked FX dependent system rather than a true domestic currency settlement system. Maybe somebody can point out others.)
BTW, it seems to me that market monetarists (including Sumner) who complain that the Fed caused the recession by allowing NGDP to fall and by keeping money “too tight” describe that tightness as a failure to lower the policy rate quickly enough. So they are rate focused in that context. I think your use of the term “target” to describe a potential central bank “interest rate target” is confusing, suggesting a false choice. Market monetarists would be admitting the co-existence of two targets in this case (the interest rate and NGDP), although in different modes, because they acknowledge that the Fed set the rate on the way down to zero, in their argument favoring an NGDP targeting antidote.
In other words, treasury bills trade at the rate they do because the Fed sets the policy rate where it does (with the spread adjusted for all sorts of complicating institutional frictions), and it can’t be any other way. It certainly can’t be vice versa.
BTW, Volcker still set the funds rate – because in responding to money supply figures, he still had to make the decision on where to stop it out on the upside – which he made though active choices on reserve management and the effect of reserves on funds pricing. Just because he didn’t announce a rate doesn’t mean he didn’t determine it by central bank actions in reserve management. It was not a market determined rate in that sense. It required judgement in translating money supply outcomes to appropriate reserve and reserve pricing settings. (And the discount window rate was a known, central bank determined number).
Nick, didn’t you do a post on a gold rule for interest rates, along these lines?
Nick,
There’s something strange about the way you’re flirting with the use of the term “targeting” to apply to what the central bank does with its policy rate. This is the catalyst for a false choice. It’s not a choice between the policy rate and NGDP, or between the policy rate and money supply, or between the policy rate and gold. It’s a choice between things like CPI, NGDP, money supply, and gold. Volcker chose money supply. You choose NGDP. Bernanke now chooses a history dependent employment path, apparently constrained by inflation in some important way. But every one of these requires that the central bank sets the policy rate in response to those things. And using the term “targeting” to describe this just muddies the waters.
JKH: “And using the term “targeting” to describe this just muddies the waters.”
I agree the terminology is muddying. But the Bank of Canada uses the same word to talk about its “inflation target” and its “target for the overnight rate”. The key distinction is that the latter is a temporary conditional target that is changed, as needed, to try to hit the former, longer term, unconditional target. In the olden days we would sometimes talk about “intermediate targets” to describe a temporary conditional target. But if I said that the overnight rate is the Bank’s “intermediate target” (which it is in a sense) people would get even more confused, and say “no it isn’t; it’s the instrument” (which isn’t strictly correct).
That’s why I said “But that’s [target the rate of interest] what the Bank of Canada does, isn’t it? Well, yes and no. Mostly no.”
The main position I am attacking in the post is someone who says “The Bank of Canada should “set” the rate of interest at (say) 4% and keep it there for decades”. Only at the very end of the post do I raise the question: well if decades is too long, is even 6 weeks too long? (But it’s not really the length of time per se; it’s more the conditional/unconditional aspect, and the role it plays in the communications strategy, that matters.
JKH: “Nick, didn’t you do a post on a gold rule for interest rates, along these lines?”
I can’t remember. I think I have touched on it a few times, in the past. I seem to remember a dream about a girl with golden eyes. It’s really just Irving Fisher’s old “Compensated dollar Plan”. The central bank adjusts the price at which it will buy and sell gold, in order to keep the Price level (in Irving Fisher’s case, but it could be NGDP or whatever) on target, and lets the market set interest rates.
Ralph: “I.e. using interest rates of regulate economies is a stroke of genius, I don’t think.”
So, you agree with me?
But you really do need to get out of the head-space (sorry for the hippy-talk) of thinking that monetary policy is is is interest rates. It isn’t. And it could be even less about interest rates than it is now. Instead of the central bank trying to set interest rates where desired saving = desired investment, and changing them every 6 weeks if it thinks desired S or I has changed, it could do something else, and let the market adjust interest rates up or down.
Nick,
“The main position I am attacking in the post is someone who says “The Bank of Canada should “set” the rate of interest at (say) 4% and keep it there for decades”.”
Good grief. Does anybody actually think that way? That can’t be the issue, surely.
Sorry, I really wasn’t clear in my language that seemed to criticize your language, and that shouldn’t be the criticism exactly.
It’s not so much the double use of “target”, but the question of “setting” the rate that’s at the heart of this.
And whether it’s 6 weeks, or indeterminate (it used to be) or unannounced (it used to be) also is not the issue.
The way to think about it IMO is to use a base case of a continuous time rate setting process, and layer those administrative arrangements over that. The continuous time response is then a function of what it “targets” in your sense of the example of NGDP as a target. I’d say gold, money, and the CPI are examples of alternative targets in this sense.
My point is that the central bank ALWAYS sets the rate as a functional dependence on whatever targeting process is in place as above. It MUST do this, given the institutional mechanism of how the commercial banking system clears on the books of the central bank and uses settlement balances (or “reserves”) to do this.
I really think you monetarists miss the entire point that your pure economic model way of thinking about things actually depends on the fact that you almost always assume this institutional arrangement in your description of the world. Therefore, you should recognize how it works explicitly as a constraint on your modeling.
I suppose you can model the world in a barter system or a system with no commercial banks and no clearing system and no pricing of settlement balances in that system, but that is another world – perhaps useful for illustration, but essential that you distinguish it that way. As soon as you assume banks clear through a central bank, that all changes. Pricing changes – because banks are motivated to price all their assets and liabilities according to signals given by the reserve setting and reserve pricing at the margin.
BTW – I think the Krugman/Keen debate was all tied up in this. I had a look at two of the Brainard papers going back, which was PK’s base of reference, and it’s very interesting that the commercial bank reserve clearing system (and commercial banks altogether) are entirely omitted from those papers. And I think that’s where Krugman went off the rails in his notion that banks don’t complicate the issue at all – it’s easy to say that when the original paper doesn’t even include banks.
Nick,
If we are going to talk about instability (I’d like to) I don’t see how we can escape introducing a model (preferably a toy model) of fractional reserve banking. We need to talk about what (if anything) the CB has to do in an ideal world. If you see a mechanism whereby the CB can buy assets using its liabilities, but can avoid determining the short rate on its liabilities, maybe that would be a good starting point for a toy model.
I’m sure people will object that institutional details matter. That’s fine, and to the extent that they do, we should start with a simple model, see how the dynamics play out, and then add complexity and see if it makes any difference. Without a common framework model, we are just going to get bogged down in the same old arguments, each person having her own mental model, and everyone talking past each other.
Ritwik: “I think that targeting the NAIRU equlibrium is unstable because while targeting the Wicksellian one is simply difficult, not unstable.”
Hmmm. Why do you think that? If it weren’t for the distinction between real and nominal interest rates (i.e. if the central bank were targeting a real interest rate on indexed bonds) I would say the two would be isomorphic. But if it targets a nominal interest rate, I would say it must be much more likely (almost certain, i.e. except in a really weird model) to be unstable. Because you have that additional positive feedback channel from inflation to expected inflation to a lower real interest rate.
Very interesting stuff on Goodhart and Buiter. I haven’t read them on this (unless I’ve forgotten), but I’m thinking along those sort of lines. You could even separate central banking into three jobs: running the payments system (and you could maybe make an analogy there to a dealer in the stock or bond market, who quotes buying and selling prices and keeps a small float and adjusts those prices continuously); acting as lender of last resort in a crisis (much more of a judgement call; and doing monetary policy (via say OMOs or gold or whatever purchases).
I know there has been some (very little) discussion (e.g. between my late colleague TK Rymes and David Longworth, ex-BoC, IIRC) of what would happen if the Large Value Transfer System were run by the commercial banks themselves with the Bank of Canada staying out of it.
rsj: “If you want to talk about unstable equilibria, then explain why you need to cut rates by more to speed up the economy than you need to raise rates to slow it down by an equivalent amount.”
You are asking me to explain a “fact” that I don’t think is a fact. And the fact that interest rates have been falling over time doesn’t make that thing a fact.
(I thought you MMT guys wanted 0% interest rates, anyway? Or is that just some of them? 😉 )
JKH: “Nick,
“The main position I am attacking in the post is someone who says “The Bank of Canada should “set” the rate of interest at (say) 4% and keep it there for decades”.”
Good grief. Does anybody actually think that way? That can’t be the issue, surely.”
I thought some of the MMT guys were saying that (only it was 0% not 4%). Maybe I totally misunderstood them?
And what is Ralph Musgrave saying above, if he doesn’t like central banks moving interest rates up and down because that’s putting the burden of adjustment on people who are interest-sensitive?
I mean it would be nice if I were only attacking an imaginary straw man here as a pedagogical thought-experiment. But am I?
Nick: “I thought some of the MMT guys were saying that (only it was 0% not 4%). Maybe I totally misunderstood them?”
Actually, I thought you were saying that. If you want to buy assets and you have zero IOR, then you are setting rates at zero (I think you might be a closet MMTer!) If, on the other hand, you want non-zero IOR, you are going to have to have a policy on that rate. And that policy matters.
K: Interest on reserves is a fairly recent policy. It used to be 0%. (I think it maybe is still 0% in some countries?). But interest rates were not 0% when interest on reserves was 0%.
Nick,
No, rates don’t have to be zero just because IOR is zero. Depending on the efficiency of the clearing mechanism, banks may hold a small amount of excess reserves at IOR=0. In Canada, that small amount is generally exactly zero.
But!
The CB cannot introduce more reserves at their discretion in such a system. Lets say you are a bank and the CB buys $1Bn of T-bills from you. What are you going to do with that $1Bn of non-interest earning reserves? You are going to lend it at the interbank rate, right? Except, I (another bank) don’t want it either. Nobody wants it because it’s excess. The most I, or anybody else, will pay to borrow it, is whatever we can earn on it by holding it overnight. I.e. zero. Therefore the interbank funds rate drops to zero. You cannot control both excess reserves and the difference between the funds rate and IOR at the same time. If you want to control excess reserves you must have FF=IOR. And then you need a policy for FF. This has nothing to do with esoteric, particular institutional arrangements. It’s inescapable economics of the clearing system.
JKH: “BTW – I think the Krugman/Keen debate was all tied up in this.”
Yes and no. It’s better to think of that as a short run/long run debate. PK was talking LR, and SK was talking SR. But SK’s SR model (where the rate of interest is exogenous) doesn’t make sense in the LR, precisely because the equilibrium is unstable (it’s my second case above). And since (IIRC) the debate was about what caused the increase in the level of debt over a couple of decades or more (?), we definitely aren’t talking about the SR where the rate of interest is exogenous. So SK’s explanation doesn’t make sense over that sort of time-scale. And PK’s objection is perfectly valid in the LR.
The Brainard/Tobin stuff, even though PK brought it up, wasn’t really relevant to the debate, IMHO.
@K:
What if the central bank is just fine with interest rates plummeting to zero in response to the monetary injection?
What if the central bank just out-and-out tells everyone that the $1Bn T-bill purchase is permanent. The money is there, y’all have to figure out what to do with it.
Would it make sense that eventually desired bank lending might rise enough that the excess reserves become required reserves, and then interest rates would return to a positive value?
Would it make sense if interest rates plummeting to zero actually sped up that process?
Alex Godofsky,
Yes it makes sense. But the quantity of excess reserves is utterly irrelevant. The only thing that has any relevance whatsoever, is the time at which excess reserves will be allowed to return to zero, i.e. the time at which the short rate will be allowed to rise above zero. And after that, the only thing that matters is the short rate (excess reserves will be zero). And since the quantity of reserves is irrelevant before the rate hike, the only thing that matters in that period is the fact that the short rate is zero. From which we can conclude that the only thing that matters at every point in time is the value of the short rate. The path of the quantity of excess reserves is either
a) irrelevant (when r=0); or
b) informationless (i.e. 0) when r>0
JKH: ironically (actually, it’s more funny than ironic) my own “very short run” analysis, when I’m yammering on about banks creating hot potatoes and planned expenditure being greater than expected income, is actually much closer (at least in spirit) to Steve Keen (AFAIK) than it is to something like Tobin or the official New Keynesian story.
Alex Godofsky: “What if the central bank just out-and-out tells everyone that the $1Bn T-bill purchase is permanent. The money is there, y’all have to figure out what to do with it.”
Actually, that part of what you said doesn’t make sense. The existence of $1Bn of reserves has no impact on the behaviour of any bank. They simply don’t care. It costs them nothing, and it earns nothing. So they don’t have to figure out anything. They know the CB will buy it back in time to start raising rates again. (Or the CB will institute IOR=FF (a floor system) and the banks will never have to care about the quantity of excess reserves.)
Nick: “Steve Keen”
There you go. I knew you were an MMTer.
@K:
Even when r=0 doesn’t the quantity of excess reserves matter in that it gives us some idea of when r>0 will happen?
Let’s hypothesize a central bank that completely ignores interest rates and conducts monetary policy purely through setting the total quantity of reserves. Every six weeks they meet and decide on a new quantity. They pick the quantity that they think is most consistent with their NGDP level target (or inflation target, or price level target, or whatever).
Why, exactly, wouldn’t such a central bank work? At some points maybe r=0 for extended periods, so what they are doing is pointless in the short term, but since their policy would, in your model, very definitely determine the path of interest rates, why wouldn’t it be able to do all the things that the current interest rate-targeting central bank can do?
(sorry, I cross-posted before)
@K:
Actually, that part of what you said doesn’t make sense. The existence of $1Bn of reserves has no impact on the behaviour of any bank. They simply don’t care. It costs them nothing, and it earns nothing. So they don’t have to figure out anything. They know the CB will buy it back in time to start raising rates again. (Or the CB will institute IOR=FF (a floor system) and the banks will never have to care about the quantity of excess reserves.)
Eventually won’t someone at the bank say “we’ve got a billion dollars sitting here earning us nothing. Why don’t we buy some stocks with it?” ?