In a world of IOR, the QTM still holds
Here’s Steve Waldman:
But the first-order effect of monetary policy is gone. Changes in the base used to engender straightforward imbalances between a direct opportunity cost and the convenience yield of holding money. A reduction of interest rates / expansion of the monetary base would lead to an increase in the direct cost of holding money rather than Treasuries, and put the economy in disequilibrium until NGDP or (too frequently) asset prices adjusted to increase the convenience yield attached to the monetary base. A contraction did the reverse. While we are stuck at zero we can argue over expectations or collateral chains, but the old, blunt, simple channel no longer functions.
And it will never function, as long as the Fed always pays interest comparable to Treasuries on base money. There is nothing special about zero, or 25 bps. What makes a liquidity trap is that the rate of interest paid on money is greater than or comparable to the rate of interest paid on Treasury bonds. So long as that is true, whatever the level of interest rates of interest, macroeconomic outcomes will be much less sensitive to changes in the quantity of money than in once-ordinary times. That is not to say, full-stop, that monetary policy is impotent. Those squishy expectations and institutional quirks may matter. But post-2008, we live in a world where insufficiently expansionary monetary policy has meant tripling the monetary base. Pre-2008, tiny changes in the quantity of base were sufficient to halt an expansion or risk an inflation. The relative impotence of changes in the monetary base is not a function of the zero-lower-bound. It is a function of the spread between base money and risk-free debt, a spread which may well be gone forever.
I can’t quite tell what he is arguing here. He clearly indicates that the quantity theory of money (QTM) no longer holds when you have interest on reserves (IOR.) That’s false but at least debatable. But he also seems to hint that monetary policy broadly defined is much weaker, which isn’t true at all. To the extent the central bank no longer influences the supply of the medium of account via open market operations, they are influencing the demand for the MoA via changes in IOR. Nick Rowe points out in the comment section that the Bank of Canada has kept inflation very close to 2% for 20 years by using both techniques (despite wild swings in fiscal policy.) So monetary policy is obviously still operative and powerful.
So lets look at the more debatable claim, is the QTM no longer true when you have IOR? No, Peter Ireland has a paper that shows it’s still true. Before discussing why, here’s a simple counterfactual to show the problem with Steve’s argument.
Assume the monetary base is 50% interest-bearing reserves and 50% non-interest-bearing cash. We are in normal times, with no liquidity trap. The government decides to embark on a Latin American-style monetary policy of 100%/year trend inflation. Because they read that OMOs don’t matter when you have zero IOR, they’ve decided to do so through changes in IOR, leaving the monetary base unchanged. So they lower IOR, which reduces the demand for reserves. Within a year or two the monetary base is almost 100% currency, which earns no interest. I.e. America circa 2007. And then they are stuck. They have no mechanism for causing prices to rise 100% per year, for year after year, if they are constrained by a decision to avoid OMOs. You need OMOs because money really does matter. The HPE works for currency, if nothing else. If you want even more inflation, there is no substitute for printing currency.
But that’s not the only problem with Steve’s argument, indeed the QTM holds much more than even the preceding example would suggest. Suppose the differential between the IOR and short-term market rates was always kept very small, say 1 basis point. At this spread the demand for reserves was very large, say 50% of GDP. Currency demand is 5% of GDP. What happens if the central bank permanently doubles the monetary base? Ireland showed that money is still 100% neutral. In the long run (once wages and prices have adjusted), all nominal interest rates and interest rate spreads remain unchanged. NGDP doubles, as does the stock of currency, bank reserves, M1, M2, the price level, the nominal output of the toaster industry, nominal ad revenues earned by Facebook, etc. Neutral is neutral, there’s no getting around that fact.
Don’t worry monetarists; the QTM is not going away.
PS. Commenters, I’m begging you to avoid bringing up the past 5 years. If you think they are relevant you have not understood a thing I’ve said here.
HT: Michael
Tags:
6. November 2013 at 06:56
The reason we haven’t tripled the price level by tripling the monetary base is because it’s not expected to be permanent. I think.
6. November 2013 at 07:11
Joshua, Yes, and/or a one-time shift from no IOR to market rates of IOR.
6. November 2013 at 07:25
Scott, IOR does make the base less volatile though, doesn’t it? (there was a paper to this effect published in 2010 or 2011). ?In other words, it helps the CB avoid printing money it will get rid of in future years. Or something to that effect.
6. November 2013 at 07:28
It’s odd for Waldman to argue that IOR means monetary policy is ineffective. Just reduce IOR back to zero and open a bigger spread between bonds and reserves. It is a really odd sort of self-induced zero bound argument. Because it’s more of an argument explaining how an increase in the monetary base hasn’t created substantial inflation due to other monetary policy changes cast as an argument for policy ineffectiveness.
6. November 2013 at 07:39
“He clearly indicates that the quantity theory of money (QTM) no longer holds when you have interest on reserves (IOR.) ”
Does he really you say that? Waldman states outright the opposite of what you claim. “Perhaps there are people in the world who think that paying 25 basis points of interest on reserves means that base money doesn’t matter, but I have not met any of them.“ I read Waldman as saying that it’s not the introduction of IOR that renders the QTM irrelevant but the removal of money’s convenience yield that kills it. Huge quantities of reserves have pushed their marginal convenience to 0, as demonstrated by short rates trading at IOR, so ensuing increases in the base have no effect.
6. November 2013 at 07:40
Scott,
So if you take the asset price/ expectation model, here is what happens.
1. Asset Prices – To the extent that OMP causes an increase in real asset prices, you get a marginal increase in the exchange of financial assets for goods/services, i.e. a bump in AD. However, if the OMP is sanitized by an equivalent increase in ER you get no effect. (Note though that the level of ER is entirely within the control of the Fed).
2. Expectations – To the extent that OMP signals looser monetary policy (and therefore higher NGDP) in the future, this will also cause a marginal increase in the exchange of assets for goods/services (essentially a shift in the financial asset to goods/services preference curve.) So with no change in asset prices and OMP entirely absorbed by ER, changed expectations will cause an AD increase.
6. November 2013 at 07:46
Scott,
You will notice the asset price/expectation model eliminates the need of a separate explanation for:
1. Time lags
2. Changes in k
3. The last 5 year
6. November 2013 at 08:00
“NGDP doubles, as does the stock of currency, bank reserves, M1, M2, the price level, the nominal output of the toaster industry, nominal ad revenues earned by Facebook, etc. Neutral is neutral, there’s no getting around that fact.”
Are you assuming stable velocity? Velocity is never stable and it depends on the transmision mechanism. Interacting with agents with higher MPC will increase velocity. If the central bank expanded 1 billion of base into bank’s accounts at fed or gave it to the unemployed the effects would be totally different.
6. November 2013 at 08:04
How much of the following would you agree with?
IOR and OMO are 2 alternative ways of influencing bank lending via interest rates policy.
If the CB lowers the interest rate target then banks will lend more (they will find more lending opportunities) and the public will want to hold more of their wealth in money at the new lower rates. The CB has to engage in OMO to provide the extra money needed to make these changes sustainable. It could avoid OMO if instead it was able to lower IOR. This will cause banks to lend more as with lowering the IR target, but no OMO is needed because the banks already have the money (in the form of excess reserves) needed to fund the additional cash (and bank acct) holdings that will result from the IOR change. Seems like there would a ZLB on both the IR target and IOR beyond which monetary policy would stop working (at least through the ways I have tried to describe).
In both cases there will be an equilibrium between interest rates (as charged by commercial banks) and money supply (as held in cash and bank accounts).
If you see the CB and the govt as one unified monetary authority then the base could also be expanded via deficits funded by new money.
6. November 2013 at 08:18
Think of IOR (or more precisely, the IOR-to-T-bill spread) as a soft reserve requirement. Back in the days when hard reserve requirements were binding constraints, the Fed could screw with the relationship between base money and NGDP by changing the reserve requirement. Now the Fed can do the same thing by changing the IOR-to-T-bill spread. The T-bill rate (or the natural T-bill rate) produces an upper bound on the velocity of base money, but there is no lower bound. By adjusting IOR, the Fed could make the monetary base arbitrarily large without affecting NGDP, but this is not really a result of IOR: the Fed could do the same time in the pre-IOR days by adjusting the reserve requirement. If it really wanted to do so, the Fed could break the relationship between base money and NGDP (away from the ZLB where it’s already broken) and make the correlation go in the wrong direction — but it’s hard to see how the Fed could ever have any reason to do so. And it could have done so in the pre-IOR days too.
6. November 2013 at 08:30
There is low money velocity. Where is the borrowing demand going to come from?
6. November 2013 at 08:43
Andy Harless,
Totally agree, except.
1. I think there is statutory limit on the rate of required reserves (with some exceptions for temporary increases).
2. I don’t know if there is actually any statutory requirement for the Fed to accept deposits from member institutions in excess of the reserve requirement.
6. November 2013 at 09:04
Scott, I agree with Waldman, and you might be reasoning from a price change. He does not address it, but there is also the demand side for lending and rates: when an economy already has heavy levels of debt (currently about 3.5x NGDP, a near-record in the US), then demand for borrowing is slack. This is what produces low interest rates (supply and demand for debt). Rates are a second order effect, not a primary.
IOR is neither a sufficient or necessary condition for reserves being locked up, but it helps. High leverage explains it quite well.
Also, you are very incorrectly positing a free market in reserves–>currency. There are extraordinary widespread limits on cash withdrawals in the US, and it’s a farce to assume otherwise — you are building that side of the argument on a false premise.
Try going to your bank and withdrawing $10,000 in currency. It’s by appointment, under protest, and many banks will just say to go pound sand (politely, of course). And if you withdraw $2,000 a day for a week, you could be considered to be “structuring” your funds illegally.
http://en.wikipedia.org/wiki/Bank_Secrecy_Act
Finally, Waldman is right, and I could have sworn that you’ve said as much: IOR is not a market mechanism for pricing reserves. Fed funds at least had the virtue that there was market feedback taking place, and that the Fed had to adjust supply conditions to meet demand to target price. When you assume a “sticky” and restricted market in currency, your argument appears to fall apart. IOR is by dictat.
Central planning of IOR means that there is no free market in reserves; and so supply versus demand for reserves does not matter (nearly as much — the risk/optionality of IOR also has an independent effect beyond the basis points they happen to be paying that day.)
IOR is icing on the cake, as already-high debt levels drive much of the slack demand for reserves, but IOR does eliminate free market pricing of reserves.
6. November 2013 at 09:32
The spread between risky assets and risk free assets is the banks compensation for taking on risk. When this spread is narrow, banks do not issue loans, the supply of credit falls, investment falls and demand falls.
IOR is risk-free money.
6. November 2013 at 09:51
jknarr – you still have market feedback, but the FED has always participated in OMO to manipulate that feedback to hit their target rate. IOR is just another tool, and the FFR still is fluctuating despite IOR, but the FED is working to keep it at 25bps. IOR is just another tool like adding and draining reserves with treasury purchases. Otherwise the FFR would probably go to “zero”.
I agree that borrowing demand is slack, and I don’t think IOR has anything to do with that. I cannot comment on the currency thing.
6. November 2013 at 10:45
Matt — the Fed Funds market is practically dead. Available reserves barely shift around.
http://research.stlouisfed.org/fred2/graph/?g=o6B
Your notion of how reserves are created to manage short term target rates is also a bit defunct for the moment: rates versus the base are extremely nonlinear approaching zero.
It takes trillions worth of balance sheet expansion to push rates through the final basis points to absolute target rate zero (ATRZ policy!).
Under “normal condition” higher interest rates, say 5% Fed Funds, day-to-day operations can be easily be accomplished in the millions range OMOs, and there is effective market push-and-pull participation.
Part of the point is that there is almost no market feedback at all approaching zero rates in any case: that’s the excess reserve surplus: it’s not traded.
Dictat IOR is simply institutionalized icing on the cake: excessive prior indebtedness is the root cause of near-zero rates: low yields are a symptom.
The IOR problem is not significant now, it becomes a huge issue in the future after a (natural) rate increase via base/NGDP declines: then, IOR will become the one and only central command interest rate (supplanting Fed Funds). The old system of Fed Funds market feedback is already dead, and not coming back — it will yield to an effectively command economy in reserves, with small and limited market participation around the target.
You work through the implications: (hint) private capital markets are rolled up into the public banking system.
A key point is that at zero rates, there are effectively no marginal creditworthy borrowers (at that given level of NGDP). Deleveraging needs to take place for reserves to be unlocked and rates to rise: i.e. a base/NGDP decline. Debt needs to be inflated or defaulted away in order for reserves come back to life. The vector for ZLB deleveraging is a swap between reserves and physical currency, but this is stymied by all-but-illegal notional sums.
So IOR does not change the present very much, because overindebtedness/slack-lending/bad balance sheets already kills off the interbank market (hence the demand for excess reserves as both liquidity and capital as well as reflecting zero price for borrowing — low demand and high supply). There was a one-off shift from cash-equivalent to interest-bearing with IOR, but this is mostly over.
IOR is a trojan horse for future re-intermediation. It does a bit on the margin in the here-and-now, along lines as Waldman suggests, but it’s not the big show.
6. November 2013 at 10:45
Doesn’t the Ireland paper have the ambiguity pointed out by John Cochrane and discussed in Nick Rowe’s post on two “neo-Wicksellian indeterminacies” (in particular, the second one)?
http://worthwhile.typepad.com/worthwhile_canadian_initi/2013/09/two-neo-wicksellian-indeterminacies.html
My take on it is of course somewhat heterodox 🙂
http://informationtransfereconomics.blogspot.com/2013/10/resolving-neo-wicksellian.html
6. November 2013 at 10:54
Joe, It depends.
mpowell, Good point.
JP, That’s what I was thinking, I should have been more specific. But even if the yield on reserves is close to the market rate (as in Canada over the past 20 years) the QTM still holds.
If reserves and T-bills ever became perfect substitutes, then T-bills obviously would be part of the base. That’s just a definitional change, with no policy implications.
I’d have to see how Steve responds to Nick’s point about Canada, before I would be comfortable claiming I fully understand his point. He seems to be making an argument about the likely future course of monetary policy, not just an abstract theoretical point.
dtoh, I agree that asset prices and expectations are part of the story.
Mike, I’m assuming that monetary policy affects V in the short run but not the long run. I am not assuming stable V in either the short or long run.
Fiscalist, You said;
“If you see the CB and the govt as one unified monetary authority then the base could also be expanded via deficits funded by new money.”
Yes, in exactly the same sense that if you consider the Fed and the potato chip industry to be one unified monetary authority, then Fed purchases of potato chips are expansionary. But I see no reason to focus on the issue of what the Fed buys, it’s the quantity of money (and IOR) that matter.
I also don’t see any obvious connection to banking. Loans don’t go up because money is printed, except in nominal terms. The real quantity of loans rise if there is a boom, and that will happen when monetary injections are greater than expected, regardless of whether the new money goes into the banking system or elsewhere. Banks can make more loans by selling bonds, they don’t need newly created money.
jknarr, You misunderstood my post, I never claimed IOR was a market mechanism. I agree that it is an administered price (actually subsidy.)
There’s over a trillion in cash held by the public, and the total is soaring higher by more than a billion dollars every week, so I doubt people have much trouble getting a hold of the cash they desire to hold.
6. November 2013 at 10:55
Andy, I agree, both IOR and RRs tend to raise the demand for the MoA, and hence are contractionary.
6. November 2013 at 11:00
To enlarge on the point Nick Rowe made with respect to Canada, that is that Canada has had interest on reserves since 1992, interest on reserves is currently in practice in several currency areas that comprise over half of global GDP.
In particular, to the best of my knowledge, Norway has had it since 1994, Sweden since 1994, Australia since 1997, the Euro Area since 1998, New Zealand since 1999, the UK since 2001 and Japan since 2008. With respect to explicit inflation rate targeting, New Zealand has practiced it since 1990, Canada since 1991, the UK since 1992, Sweden since 1993, Australia since 1994 and Norway since 2001.
Of these currency areas only the US, Japan and the UK pay an interest rate on reserves that exceeds the lowest rate on government bonds. In the rest of these currency areas the rate on reserves is typically on a little below the lowest rate on bonds. These three currency areas also happen to be the only ones currently clearly at the zero lower bound in policy interest rates, and which also currently have QE programs. I don’t think anyone would claim that any of the other currency areas are in a liquidity trap with the possible exception of the Euro Area, but I think a good argument can be made that the Euro Area’s monetary policy problems are extremely self inflicted.
And although Canada has a record of unsually low inflation variability, I don’t think any of the currency areas explicitly practicing inflation rate targeting have had much difficulty in hitting their targets, and that even includes the UK which has been above target for much of the time it has been at the zero lower bound in policy interest rates.
And as a final note, with the exception of the US, the Euro Area, Japan and the UK, the monetary base in each of these currency areas is currently only about 3-4% of GDP, which is actually pretty low by historical international standards.
6. November 2013 at 11:00
Jason, It may, but the general result holds in almost any plausible model (It’s been a long time since I read the Ireland paper.)
6. November 2013 at 11:02
Thanks Mark, I would assume that once rates rise above zero the Fed will set the IOR at slightly below T-bill yields. Does that sound right?
6. November 2013 at 11:03
Minor correction:
Norway has had interest on reserves since 1993.
6. November 2013 at 11:05
Scott,
“I would assume that once rates rise above zero the Fed will set the IOR at slightly below T-bill yields. Does that sound right?”
That’s how it typically ends up in practice.
6. November 2013 at 11:33
Scott, you said;
” I agree that asset prices and expectations are part of the story.”
My point was NOT that it was PART of the story, but rather that it is the FULL story. An asset price/expectation model is sufficient to fully explain all of monetary policy.
(To paraphrase a note from Compte to Mill)
While my repeated and lucid explanations have not yet shaken your point of view — which on my part I persist in firmly believing to be erroneous— permit me, nonetheless to maintain my hope that your perseverance in this error is not irrevocable and that it will later give way to the natural impact of your own reflections.
6. November 2013 at 11:34
Here’s another way to contrast your views with Steve’s. He’s making the case that we’re at, or close to, case 5b as described in your Hot Potato post. You’d probably say that we’re between 5b and 5c, but closer to 5c (I too think that we haven’t hit 5b yet). As such I’d argue that there are no large theoretical differences between you two, only empirical differences.
6. November 2013 at 12:30
Jknarr – it is my understand the FFR does fluctuate because not all banks are eligible for IOR.
Here is the Bloomberg FFR chart, which has FF down at 8-10 bps:
http://www.bloomberg.com/quote/FDFD:IND/chart
This is the explanation from the NY Fed:
“”While IOER has been effective at influencing the FFER, it has not served as a hard minimum rate at which all institutions are willing to lend funds. This is because some institutions are eligible to lend funds in the federal funds market, but are not eligible to earn IOER, such as the government-sponsored enterprises (GSEs). DIs, U.S. branches and agencies of foreign banks, Edge Act and agreement corporations, and trust companies are eligible to receive interest on reserves and excess reserve balances held at the Federal Reserve. Pass-through correspondents that are themselves eligible to receive interest may pass the interest they receive on balances that represent their respondents required reserve balances and excess balances, but they are not required to do so. By contrast, pass-through correspondents that are not themselves eligible to receive interest must pass back to their respondents the interest they receive on balances that represent their respondents required reserve balances. Other factors that influence the rates some institutions are willing to pay for fed funds in this environment include the impact on their balance sheet, which often determines a maximum rate a DI is willing to pay.”
Scott in a previous post responded to the NY Fed comment above with a question on why banks don’t borrow from the FF market at 10bps and park those reserves earning 25. Great question, and maybe that is exactly what is happening thus the reserve pileup from risk-less arbitrage. But not everyone is eligible to get IOR.
In the end it does not matter, because loan demand stinks. If people and business is ready to borrow the banks are in good capital shape to lend.
6. November 2013 at 12:39
I understand, I just think that QTM has intervening variables that can’t be assumed away — so:
1)money
2)?
3)price
The entire Krugman-Waldman thing was about step 2 between reserves and prices. Whether IOR is a market mechanism gets at that question (only a small bit admittedly): does permanent reserve creation matter to prices under IOR?
I really would not assume away the fascinating complications that are possible in step 2, or fall back on tautologies (such as velocity).
If IOR is moot, why does the Fed need it?
6. November 2013 at 13:04
dtoh: Yes, but what is money for? Assets are never just assets.
6. November 2013 at 13:07
http://ftalphaville.ft.com/2013/11/06/1688572/fed-wonk-special/
I’m not two pages into this (my midday reading at work is sadly minimal), but it directly addresses NGDP targeting and ultimately rejects it because it’s too easily misunderstandable (which is an understandable concern from the Fed in Nov 2013). I’m sure you’ll get to it—but please comment. Thanks.
6. November 2013 at 13:21
Lorenzo,
What is money for? MOE, MOA and sometimes a store of value (financial asset). Not sure what you’re getting at.
6. November 2013 at 13:37
Hi,
A few points:
1) I don’t claim that “monetary policy” becomes ineffective, if we include setting the rate of IOR as part of monetary policy. That remains a very powerful lever! I do claim, that in a system where reserves are locked into a paying very close to the Treasury rate QTM breaks, if M is defined as base money. This is the case Scott points to,in which the definition of money becomes base money + short-term Treasuries, and OMOs become mere swaps between these two terms of broader money.
2) As Mark points out, lots of countries have paid IOR for a while, operating so-called “channel systems” of monetary policy. Under channel systems, the quantity of base money continues to be a powerful lever, for the reason that mpowell suggests: The central bank has the power to set the IOR rate substantially below the Treasury rate if it wishes to.
But that is only possible because, under a channel system, base money is scarce, and nearly all of it is held as currency. With sufficient scarcity (qua JP Koning), there is no arbitrage between IOR and Treasury rates. Currency users accept even a very large opportunity cost for the convenience yield of holding base. (People kept money in their wallets even in the late 1970s/early 1980s, when the cost of doing so was in the high teens!)
My (factual) claim is that the US now operates a floor system, not a channel system, and my (speculative) claim is that it will continue to do so indefinitely. Under a floor system, where base money far exceeds quantities required by banks (to meet clearing and reserve requirements) and currency users, there is an arbitrage between IOR and T-bill rates. The IOR rate cannot be set below the Treasury rate, or more accurately, the Treasury rate cannot be held above the IOR rate, because excess reserves will purchase T-bills until IOR rates equal or exceed the bill rate. Under a floor system, Treasuries and base-money are near perfect substitutes, differing primarily because of institutional quirks, like the fact that (pretty much) only banks can directly hold reserves.
Putting aside these institutional quirks, reserves and Treasuries are currently near-perfect substitutes, “money” in the QTM sense is (short-term T bills + base money), and (I think) QE is primarily effective by virtue of 1) what it signals about future rate policy; and by 2) its effect on asset prices / reduction of term premia. (QE seems to effect financial markets much more directly and profoundly than economic activity, whether measured in nominal or in real terms!)
One way of maybe meshing QTM and my views is to point out that, if (in a floor system), money and “Treasuries” are perfect substitutes, that begs the question of how far out the yield curve you can go and still be “money”. “Money” by definition is an asset that is risk-free in the unit of account. 30-year Treasuries aren’t money: they are objectively volatile as yields fluctuate. Very short-term Treasuries are, in the floor system, money, as they carry negligible interest rate risk. QE and forward guidance have pushing “moneyness” (hi JP!) out the yield curve, by expanding the time-window over which Treasury debt is not subject to meaningful interest rate risk. So, in a generalized way, QE + forward guidance increases the “stock” of money, as it comes to include not only <1-month Ts, but also 6 month or longer bills and notes. Where one draws the line will always be arbitrary, but I think it reasonable to claim that forward guidance that promises to peg short-rates over a long period, made credible by continued QE, has increased the moneyness of a range of assets, and central banks retain the ability to adjust how far out the yield curve moneyness goes. But I don't think that confers a great deal of power, because these assets that become money-like in being near perfect substitutes for base money still carry no opportunity cost to hold. And I don't think that the price level is proportionate to or pinned down by this broader construction of money. So, much as I love syntheses, I am sure I will leave Scott unsatisfied!
6. November 2013 at 15:04
“If you want even more inflation, there is no substitute for printing currency.”
A fall in the demand to hold currency could happen for a number of reasons (although it’s implausible) and that would also cause prices to rise.
6. November 2013 at 15:18
“I also don’t see any obvious connection to banking. Loans don’t go up because money is printed, except in nominal terms.”
Here is my logic.
The fed wants to create inflation by expanding the monetary base.
If it does so by OMO then as everyone is already holding all the bonds they want at current prices they have to lower interest rates to induce less bond holding and more money holding. Financial wealth has not changed, but some of it has moved from bonds to money. Its hard to see how this actually induces any real changes in spending. But lower interest rates will likely induce more demand for loans – and this is the channel via which the new money will flow into the economy.
If it does it via lowering IOR then again the main effect is via lending. If a bank is getting 2% IOR and this is reduced to 1% then some loans that were previously not profitable will now become so. How else will changing IOR affect the economy ?
It seems to be all about increased lending and scarcely at all about HPE. Is my logic wrong ? Does the mere act of holding more money and less bonds create a HPE ? Why ?
On “But I see no reason to focus on the issue of what the Fed buys, it’s the quantity of money (and IOR) that matter.”
I don’t get this. If the fed buys chips with new money then total financial wealth has increased and so likely will induce more spending (there will be a HPE).
If the fed swaps bonds for money then financial wealth stays the same and I see no reason why spending will change (No HPE.
6. November 2013 at 15:24
“To the extent the central bank no longer influences the supply of the medium of account via open market operations, they are influencing the demand for the MoA via changes in IOR.”
To the extent that money as a MoA is influenced, it will be by way of influence on money as a MoE. MoA is derivative to MoE.
We’re using dollars as a medium of account because dollar accounts form by way of dollar exchanges. Dollars are exchanged “universally”, hence people use dollars as a medium of account.
6. November 2013 at 15:38
Prof. Sumner,
This is a phenomenal new post by Ashok Rao:
http://ashokarao.com/2013/11/06/inflation-not-just-now-but-forever
“A lot of people argue for more inflation now but are broadly okay with a low, two percent long-run target. This hurts America’s most vulnerable workers without really benefitting anyone in return…….
A rising tide lifts all boats, but equity necessarily has a little bit of zero sum to it. A liquidity trap hurts everyone but, in the process, creditors get more than they deserve, and debtors less. A two percent inflation target may have been sufficient protection of the Great Moderation, but it is absurd that most policymakers are unwilling to acknowledge important structural shifts – the safe asset shortage, really – since. A two percent inflation target is undesirable today for the same reasons perfect price stability would have been an idiotic decision in the 1990s. Yet most policymakers cheerfully accept the former but look at the latter as they would the gold standard.
The benefits are not restricted to aggregate demand……”
6. November 2013 at 18:24
“NGDP doubles, as does the stock of currency, bank reserves, M1, M2, the price level, the nominal output of the toaster industry, nominal ad revenues earned by Facebook, etc. Neutral is neutral, there’s no getting around that fact.”
“Mike, I’m assuming that monetary policy affects V in the short run but not the long run. I am not assuming stable V in either the short or long run.”
How can ngdp, the stock of currency, m1 and m2 all double without velocity being stable? It seems you are assuming constant velocity.
6. November 2013 at 18:28
Seems to me the Fed should cut IOR by one basis point every month…maye they should done this way back but late is better than never…
6. November 2013 at 18:58
Steve Waldman,
I think you’ve got this more or less right. To expound and clarify a little bit.
The correct way to view QTM is as an equilibrium condition between the MOE and NGDP. However for this work, you need to subtract out that portion of the base not used as an MOE, specifically (at least within the context of recent history), you need to subtract out ER.
(As you correctly noted, from the perspective of policy efficacy, ER presents no issue since the Fed can easily control IOR and by extension ER.)
I would suggest in a floor system, a better way to think of things is not that Tbills are a money substitute, but rather that money becomes a financial asset substitute. Effectively money becomes dual purpose. It can be used both as an MOE and as a store of value, i.e. a financial asset. For the QTM equilibrium to hold you need to focus solely on the quantity of money which is being used as an MOE and to subtract out any increase in the base which results in an increased use of money as an SOV (either ER or currency held by the private sector for non-transactional purposes).
As I have repeatedly emphasized to Scott, QTM works as an accurate description of an equilibrium condition, but a) it ties you up in knots when you try to use it to describe the mechanism by which monetary policy works and b) it requires (as you have demonstrated) a lengthy and detailed elucidation to address the large recent increase in the base.
As you have alluded, a much better model is one which relies solely on a) expectations (partly encompassed by signals of future rate increases) and b) real prices of financial assets. Under this model all that is required to prove policy efficacy and to explain the mechanism is to show that Fed action results in a marginal increase by the non-banking sector in the exchange of financial assets for goods/services. (And by financial assets I mean all financial assets including lines of credit, commercial paper, home equity loans, credit card balances, Tbills held in portfolio or money being used as an SOV.)
This marginal increase in the exchange of assets is caused by either a) a change in the real price of financial assets prices (relative to goods and services) and/or b) a change in expectations of future NGDP.
I would quibble a bit with:
1. Your assertion that QE affects financial assets more profoundly than real economic activity – This can’t be the case unless you reject the most basic of economic tenets…. downward sloping convex indifference curves which exist as equally for financial assets as they do for apples.
2. Your highlighting of term premia as an important aspect of the effect of QE – You can cover this under the more general description of the effect of financial asset prices if you accurately define price as 1/(1 – the real risk adjusted after tax annualized IRR).
6. November 2013 at 19:28
Errata – Last sentence paragraph 4 (either ER or currency held by the private sector for non-transactional purposes).
“Private sector” should read “non-banking sector”.
7. November 2013 at 01:30
dtoh
“I would quibble a bit with:
1. Your assertion that QE affects financial assets more profoundly than real economic activity – This can’t be the case unless you reject the most basic of economic tenets…. downward sloping convex indifference curves which exist as equally for financial assets as they do for apples.”
So Bill Gates spends the same proportion of his income on food as does some unemployed person in the Bronx?
Wealthy people and poor people have different indifference curves.
That is why monetary policy is so innefective. Significant increases in asset prices as a result of QE arent causing significant increases in demand becuase assets arent widely held meaning the wealthy with a low MPC experience most of the wealth effect while the poorer with a high MPC dont experience much wealth effect. The top 20% own over 90% of stocks while the bottom 60% own around 2.5% of stocks.
7. November 2013 at 02:45
Am sure someone has already linked to this, or discussed it, but top left chart on Fig 1.1 at the back of the paper looks very familiar.
http://www.imf.org/external/np/res/seminars/2013/arc/pdf/wilcox.pdf
And pages 15-26 rediscovering the effects of downwardly sticky nominal wages are very promising. And liberal reading of Svensson (page 58) too.
Goldman’s top economist says this paper and the English one suggest much more fed easing.
http://www.businessinsider.com/hatzius-fed-will-lower-unemployment-threshold-to-6-percent-2013-11
Other brokers don’t like it.
http://www.businessinsider.com/ny-feds-fg-paper-not-a-policy-signal-2013-11
But a good debate, moving things the MM way.
7. November 2013 at 02:56
Papers above being presented at this:
http://www.imf.org/external/np/res/seminars/2013/arc/index1.htm
Are you going, Scott? You should be!
7. November 2013 at 03:39
Mike,
People don’t need to “hold” assets in order to “exchange” them for goods and services, and it’s not just people, it’s firms as well. Two examples of exchanging financial assets for goods/services would be;
A company sells Treasury securities held in portfolio in order to build a new line in a factory.
And individual charges a pack of cigarettes to their credit card.
The distribution of financial assets doesn’t really matter. We know from the most basic of economic theories that indifference curves are downward sloping so that if the real price of financial assets rise relative to the price of goods and services, then there will a marginal increase in the exchange of financial assets for goods and services. It doesn’t matter who holds the assets. All we care about is the aggregate indifference curve.
7. November 2013 at 04:38
dtoh
“The distribution of financial assets doesn’t really matter. We know from the most basic of economic theories that indifference curves are downward sloping so that if the real price of financial assets rise relative to the price of goods and services, then there will a marginal increase in the exchange of financial assets for goods and services. It doesn’t matter who holds the assets. All we care about is the aggregate indifference curve.”
The curve can slope sideways. For example Warren Buffet will get no more utility from adquiring more food when compared to acquiring financial assets.
If most wealth is held in few hands then the aggregate indifference curve will have a less downward slope or more horizontal slope which means the increase in demand of goods and services from an increase in price of financial assets will be smaller. Therefore monetary policy becomes less efective at stimulating demand.
People of high net worth dont change their demand for goods and services much in response to an increase in financial asset prices when compared to middle class and lower classes.
“People don’t need to “hold” assets in order to “exchange” them for goods and services, and it’s not just people, it’s firms as well.”
Money is an asset. How are you gonna make an exchange if you dont provide an asset? Unless you are gifted the goods maybe. I know its firms as well.
7. November 2013 at 06:43
dtoh, I still think the HPE is important.
JP. I think he also overlooks the importance of zero interest on currency, as I indicated in the post.
jknarr, The Fed does not need IOR. And I never rely on velocity tautologies.
Steve, This is the comment I was responding to:
“That is not to say, full-stop, that monetary policy is impotent. Those squishy expectations and institutional quirks may matter.”
I now see that I misinterpreted this, as did Nick Rowe I believe.
The most plausible explanation of why QE matters is that markets believe reserves and T-bills will not be perfect substitutes in the future, and that some of the QE is going to be permanent.
I find it very implausible that the Fed would adopt a policy in the future that is both very different from current channel-type systems in other countries, and also makes monetary policy considerably more difficult to undertake. If you control policy via a monopoly on asset A, why make asset A a perfect substitute for an asset B that you don’t control? But if they want to shoot themselves in the foot, I guess I can’t stop them. Also note that currency remains interest-free, so the quantity of money will still matter along that dimension as long as we still have paper money.
John, That’s true.
Fiscalist, You said;
“Financial wealth has not changed, but some of it has moved from bonds to money. Its hard to see how this actually induces any real changes in spending.”
You lost me here. The monetary injections raise asset prices, as you indicated. So real financial wealth does increase if prices don’t increase.
Travis, Thanks for the link. Good post.
Mike, It’s a ceteris paribus claim. The effect of the policy is to double everything. But other things might also happen, so the actual change may not be a doubling.
The income distribution has nothing to do with the effectiveness of monetary policy, rather the key is to have a credible target, and make sure enough of the monetary injections are permanent to hit the target. If policy is not very effective at boosting NGDP, it will ipso facto be not very effective at boosting asset prices.
James, Thanks for the links. I’m already doing two DC trips. I did a paper on November 1st in DC, and have another on the 14th. Only so much time . . . .
7. November 2013 at 07:06
“You lost me here. The monetary injections raise asset prices, as you indicated. So real financial wealth does increase if prices don’t increase.”
Clarifying question: I can see that lower interest rates will cause bond prices to rise. Perhaps also people will take some of the new money and buy other assets (such as stocks) with it and this will drive up these asset prices until people are happy with the new mix of money and assets prices. In your model is it this increase in wealth that eventually drives increased spending on things other than financial assets or is there something else at play ?
7. November 2013 at 07:25
You will be the elephant in the room. Or rather NGDP forecast targeting will be.
7. November 2013 at 07:42
“NGDP doubles, as does the stock of currency, bank reserves, M1, M2, the price level, the nominal output of the toaster industry, nominal ad revenues earned by Facebook, etc. Neutral is neutral, there’s no getting around that fact.”
“Mike, It’s a ceteris paribus claim. The effect of the policy is to double everything. But other things might also happen, so the actual change may not be a doubling.”
But by assuming ceteris paribus you take away the vital aspect of the issue. You cant assume something that never happens (stable velocity) and which is the central aspect of why monetary policy is failing.
“The income distribution has nothing to do with the effectiveness of monetary policy, rather the key is to have a credible target, and make sure enough of the monetary injections are permanent to hit the target. ”
The target cant be credible if the mechanism is innefective. The credit channel is innefective becuase people wish to develerage as a result of historically high debt to gdp levels therefore lending cant expand sufficiently. Portfolio rebalancing from QE isnt increasing spending enough from the wealth effect of increased asset prices becuase assets arent very widely held. No capacity means no credibity. New mechanisms need to be put in place to work outside the credit channel and interact with the broader public which has a higher average MPC than just the current set of asset holders.
“If policy is not very effective at boosting NGDP, it will ipso facto be not very effective at boosting asset prices.”
Stocks are at all time highs while ngdp growth is below par.
7. November 2013 at 12:55
“The most plausible explanation of why QE matters is that markets believe reserves and T-bills will not be perfect substitutes in the future, and that some of the QE is going to be permanent.”
Scott – this seems contradictory. If some of the excess base will be permanent, then you can’t have a spread between the rate earned on reserves and the rate earned on T-bills, for the reasons Steve described. Even if you get rid of IOR, that means the T-bill rate will remain at zero, until either the central bank removes the excess reserves (which means they’re not permanent) or until deposit growth reaches the point where all the excess reserves have become required reserves (but that could take a very long time).
This is the key passage in Steve’s prior post: “where base money far exceeds quantities required by banks (to meet clearing and reserve requirements) and currency users, there is an arbitrage between IOR and T-bill rates. The IOR rate cannot be set below the Treasury rate, or more accurately, the Treasury rate cannot be held above the IOR rate, because excess reserves will purchase T-bills until IOR rates equal or exceed the bill rate.”
7. November 2013 at 18:57
Fiscalist, You said;
“Perhaps also people will take some of the new money and buy other assets (such as stocks) with it and this will drive up these asset prices until people are happy with the new mix of money and assets prices. In your model is it this increase in wealth that eventually drives increased spending on things other than financial assets or is there something else at play ?”
In my model stock prices do not rise because people take money and buy stocks, they rise on new information, expectations of higher future NGDP. If the policy is announced at midnight, stocks open higher the next morning before the very first trade. And I believe that higher future NGDP is caused by expectations of a future hot potato effect, which occurs once we exit the zero bound. And those expectations of faster NGDP growth tend to boost current spending—what Keynes called “confidence.”
James, I may be the skunk in the room on Nov. 14th.
Mike, I’m not sure you understand how economists use ceteris paribus. Let’s say a monetary policy is capable of producing a 100% boost in NGDP. Then I say the effect of the policy, in and of itself, is to raise NGDP by 100%. Now suppose other factors depress NGDP by 13%. Then it may actually rise by 87%. But I’d still say monetary policy caused a 100% rise, and other factors caused a 13% fall and the net effect is 87%. In other words, we try to isolate the causal impact of each separate factor. That’s equally true of fiscal policy, supply side reforms, or anything else.
None of this has anything to do with velocity being constant.
Regarding stocks, how would they have done if the Fed had followed ECB policy and produce a double dip recession? Stocks would have done much worse. And so would the economy.
Jared, The question of ERs is entirely separate. In my view rates will rise when we get robust NGDP growth. Without IOR, (or a IOR lower than T-bill yields) this means that when the economy is robust, the Fed would have two choices:
1. Pull out the ERs.
2. Hyperinflation.
And #2 wouldn’t take long at all, if you’ve studied Latin American history. We regard these scenarios as far-fetched, and in a way they are, but only because the Fed is not that reckless. I’ve seen MMTers claim that if you were in normal times, say 1995, and you doubled the base, interest rates would fall to zero. In fact, if we doubled the base in 1995 we’d have had hyperinflation within months.
Now technically you could say if we had hyperinflation all the ERs would immediately go out as currency, and I’d agree. But even Steve admits the QTM applies to currency when rates are positive.
I think Steve’s mistake is to write a post as if he is describing the future of monetary policy in America, whereas he is actually describing a situation that is unlikely to occur when the economy recovers. On the other hand I’m completely on board to the idea that we’ll hit the zero bound again in the next few recessions, and thus that the zero bound situation deserves careful consideration. But that’s very different from a permanent zero bound, which is like the black hole in physics, you throw out the textbooks and start over.
7. November 2013 at 20:54
Mike, I’m not sure you understand how economists use ceteris paribus. Let’s say a monetary policy is capable of producing a 100% boost in NGDP. Then I say the effect of the policy, in and of itself, is to raise NGDP by 100%. Now suppose other factors depress NGDP by 13%. Then it may actually rise by 87%. But I’d still say monetary policy caused a 100% rise, and other factors caused a 13% fall and the net effect is 87%. In other words, we try to isolate the causal impact of each separate factor. That’s equally true of fiscal policy, supply side reforms, or anything else.
So under ceteris paribus what happens to velocity in your example? I thought ceteris paribus means velocity stays constant.
“None of this has anything to do with velocity being constant.”
But if you make a claim like “M1 and ngdp move up by the same amount” velocity must play a role right?
7. November 2013 at 21:16
Scott, you said;
In my model stock prices do not rise because people take money and buy stocks, they rise on new information, expectations of higher future NGDP. If the policy is announced at midnight, stocks open higher the next morning before the very first trade. And I believe that higher future NGDP is caused by expectations of a future hot potato effect, which occurs once we exit the zero bound. And those expectations of faster NGDP growth tend to boost current spending””what Keynes called “confidence.”
Scott, can I assume that you will agree i) that a rise in stock prices (or the price of any other financial asset) will in and of itself cause an increased exchange of financial assets for goods and services, and ii) that such an increase is, by definition, an increase in NGDP.
Can I also assume that you will agree that an increase in demand for something (e.g. Fed buying financial assets) will in general tend to cause the price of that something (e.g. financial assets) to rise.
7. November 2013 at 23:09
Scott
“In my model stock prices do not rise because people take money and buy stocks, they rise on new information, expectations of higher future NGDP.”
If expectations stay the same asset prices will still increase through portfolio rebalancing if interest on reserves is lowered by the fed and safe assets are taken off the market through QE. Do you agree?
We have had a significant increase in stocks while income expectations have remained low. Therefore it seems stocks havent risen as a consequence of expectations but portfolio rebalancing.
8. November 2013 at 11:24
Mike, I mean you hold V constant when considering the impact of M-policy alone, but not when considering the impact of all shocks hitting the economy combined. I think language is confusing things here, as the words don’t convey the distinction very well. Maybe someone else can help.
dtoh, You said:
“Scott, can I assume that you will agree i) that a rise in stock prices (or the price of any other financial asset) will in and of itself cause an increased exchange of financial assets for goods and services, and ii) that such an increase is, by definition, an increase in NGDP.
Can I also assume that you will agree that an increase in demand for something (e.g. Fed buying financial assets) will in general tend to cause the price of that something (e.g. financial assets) to rise.”
No to both. Never reason from a price change on the first point. It depends on why stocks rise. And people rarely spend stocks on goods, they sell stocks for money. Money that someone else gives up.
On the second, the Fed bought massive quantities of Treasury debt in the 1970s, and this caused T-bond prices to fall sharply.
lxdr, Stocks have risen due to future profits being discounted back to the present at a lower discount rate. That doesn’t really require any rebalancing, although it might occur. Profits are also rising.
8. November 2013 at 11:33
Scott,
Let me rephrase the question. If the price of financial assets rises relative to the price of goods and services, and there is no change in expectations, will that result in an increased exchange of financial assets for goods/services?
8. November 2013 at 12:31
“Mike, I mean you hold V constant when considering the impact of M-policy alone, but not when considering the impact of all shocks hitting the economy combined. I think language is confusing things here, as the words don’t convey the distinction very well. Maybe someone else can help.”
I’ll try to help, and please let me know if I botch it:
Holding V constant just means V is whatever it would have been otherwise. If the Fed permanently doubles the monetary base and makes no other policy changes (Reserve requirements or interest on reserves), then in the long run V will be whatever it would have been anyway. So you can assume NGDP will be double what it would have been if the base was held constant. It doesn’t mean V will never change.
This assumes that permanently doubling the base has no long term effect on V. (I believe that’s part of the long term neutrality of money.)
8. November 2013 at 20:16
“Mike, I mean you hold V constant when considering the impact of M-policy alone, but not when considering the impact of all shocks hitting the economy combined. I think language is confusing things here, as the words don’t convey the distinction very well. Maybe someone else can help.”
Yes, I understand what your saying so far.
“None of this has anything to do with velocity being constant.
NGDP doubles, as does the stock of currency, bank reserves, M1, M2, the price level, the nominal output of the toaster industry, nominal ad revenues earned by Facebook, etc. Neutral is neutral, there’s no getting around that fact.”
You said that you hold v constant and that v isnt relevant.
I guess what you were trying to say was that the reduction in velocity is not because of MP but a shock from somewhere else?
If you do think MP is not responsible for decline in velocity I disagree MP certainly affects velocity. Imagine all the excess reserves that are sitting in reserve accounts at the fed were placed into the accounts of all people evenly instead? Velocity would pick up significantly.
9. November 2013 at 08:34
dtoh, People exchange money for goods. I suppose money is a sort of financial asset, but I’d prefer you were more specific. People don’t exchange stocks and bonds for goods.
Negation, That’s right.
Mike, You said;
If you do think MP is not responsible for decline in velocity I disagree MP certainly affects velocity.”
I certainly agree it affects V in the short run, and said so. But not in the long run.
10. November 2013 at 00:17
ssumner
“I certainly agree it affects V in the short run, and said so. But not in the long run.”
Why would MP only affect V in short but not long run? I think MP must always affect V to some degree.
Do you think the large decline in V since 2008 is mostly as a result of MP or other factors? What other factors could cause such a massive decline in V?
10. November 2013 at 19:28
Are the effects of $2.2t of excess reserves different from the effects of $2.2t of Reverse Repurchase Transactions? Aren’t they just two separate ways to drain $2.2t of reserves?
10. November 2013 at 21:27
Negation of Ideology
I understand and I think your explanation of V is correct. But the thing is the fed never just does one off MP adjustments. It constantly updates daily the amount of base or QE in order to attain its targets so therefore the short term effects on V are always happening. Therefore it is vital to consider the mechanism it employs and its effect on V.
14. April 2017 at 04:31
[…] “Don’t worry monetarists; the QTM is not going away.” http://www.themoneyillusion.com/?p=24597 He then warns us to disregard the last 5 years as having no bearing on his […]