Monetary policy is not about banking
When I advocate something like QE or negative interest on reserves, I often get people complaining that this will not boost bank lending, or that we shouldn’t even be trying to boost bank lending. It almost makes me want to tear out my hair. What in the world does banking have to do with monetary policy? Yes, it may or may not boost bank lending, but it doesn’t matter, as monetary policy is about the hot potato effect. And yes, the Fed should not be trying to boost lending, any more than it should try to boost sales of microwave ovens. NGDP is what matters.
Jim Glass directed me to a new study by the Bank of England, which confirms that monetary policy is about the hot potato effect (aka portfolio rebalancing channels) not bank lending. Here is the abstract.
We test whether quantitative easing (QE) provided a boost to bank lending in the United Kingdom, in addition to the effects on asset prices, demand and inflation focused on in most other studies. Using a data set available to researchers at the Bank, we use two alternative approaches to identify the effects of variation in deposits on individual banks’ balance sheets and test whether this variation in deposits boosted lending. We find no evidence to suggest that QE operated via a traditional bank lending channel (BLC) in the spirit of the model due to Kashyap and Stein. We show in a simple BLC framework that if QE gives rise to deposits that are likely to be short-lived in a given bank (‘flighty’ deposits), then the traditional BLC is diminished. Our analysis suggests that QE operating through a portfolio rebalancing channel gave rise to such flighty deposits and that this is a potential reason that we find no evidence of a BLC. Our evidence is consistent with other studies which suggest that QE boosted aggregate demand and inflation via portfolio rebalancing channels.
PS. I have a new post over at Econlog that is far more important than this post. Read the whole thing.
Tags:
12. April 2015 at 07:45
Granted I haven’t been paying much attention lately, but I think this is the first time I’ve seen you state in so many words that the hot potato effect is simply ye olde portfolio rebalancing channel. Well and good, but doesn’t that mean that you are basically a disciple of James Tobin? Nothing wrong with that AFAIAC, but it doesn’t seem as if that’s how you see yourself.
12. April 2015 at 08:01
I’ve always pointed this out, but I’ve also pointed out that the portfolio-re balancing effect is minor, especially when you’re buying low yielding assets. Bank lending is a much more important determinant of the broad money supply, which is why in normal times monetary policy is all about banks, because they control how much the banking sector extends credit by putting a floor on the return they’re willing to earn from the private sector by setting the interest rate, as the BoE explains: http://www.bankofengland.co.uk/publications/Documents/quarterlybulletin/2014/qb14q1prereleasemoneycreation.pdf
It seems like a massive QE program could easily be offset by only a relative minor shock to bank lending, in terms of the effect on the broad money supply, which is why many critique QE as weak.
Also this ignores fiscal policy, if governments spend money into the economy, is this extra money necessarily being offset by the debt they incur and the resulting promise of fiscal consolidation in the future, or are the markets widely expecting the government and central bank to strategically monetize the debt? As e.g. some like Adair Turner is expecting Japan to do: http://www.project-syndicate.org/commentary/japan-monetization-government-debt-by-adair-turner-2015-03
If that’s the case, then government deficit spending along with QE can be considered a major source of new broad money in the economy.
12. April 2015 at 08:16
Six channels of monetary transmission spelled out by the NY Fed:
http://www.newyorkfed.org/education/pdf/2012/McCarthy_money_transmission_mechanism.pdf
Visualization from Mishkin:
http://i.imgur.com/PKICWP2.png
[These are from a debate running over at reddit started by MMTers claiming “money policy works only by moving the interest rate and hoping something happens”. So as a free extra bonus we get Nick Rowe’s nifty presentation of the MMT model in IS-LM terms, which I’d never seen before.]
12. April 2015 at 08:38
Well, this is actually a very important post. When you have people such as John Cochrane, Tyler Cowen, and Scott Grannis contending that QE was nothing more than a swap of reserves for bonds, then I think this post is extremely important. When the Fed conducts QE it creates reserves, and it also gives an equal amount of cash to bond sellers
The QE program was large enough to influence the economy through the hot potato effect. And remember, some people who sold bonds to the Fed spent their money, boosting aggregate demand.
12. April 2015 at 08:50
Robert L. Hetzel:
“Nominal GDP: Target or Benchmark?”
http://www.richmondfed.org/publications/research/economic_brief/2015/pdf/eb_15-04.pdf
12. April 2015 at 08:52
“When the Fed conducts QE it creates reserves, and it also gives an equal amount of cash to bond sellers”
Reserves are created to pay for the bonds.
“The QE program was large enough to influence the economy through the hot potato effect. And remember, some people who sold bonds to the Fed spent their money, boosting aggregate demand.”
At the ZLB, people are indifferent between holding money (reserves) or bonds. Why would then spend more?
12. April 2015 at 09:42
Scott,
maybe there is no real disagreement here, but my framing is slightly different. The purpose of QE is to boost jobs and lending, but the purpose of QE is not to boost jobs and lending at a particular bank that receives deposit inflows as a result of QE.
12. April 2015 at 09:47
M,
They’ll spend more because their stock of money has risen relative to their normal annual expenditure. I.e. their stock of money will presumably have risen above their preferred “stock of money: annual spending” ratio. But I still don’t accept that market monetarism is the best solution, because when the state buys assets, that messes up the ratio of the price of assets to the price of current consumption stuff. I’ll expand on that.
If a fall in AD is in any way caused by a decline in capital spending (as opposed to current spending) that decline will be either rational or irrational. To the extent that it is RATIONAL, the state should not interfere with the relative prices of capital goods or assets relative to the price of current spending stuff, and that interference takes place under market monetarism when the state buys assets / capital goods (as mentioned above).
But to the extent that the decline in capital spending is IRRATIONAL, the situation won’t be helped by the state buying up assets, and thus raising the price of assets. In fact that might even CUT spending on capital goods / assets. Ergo in neither case (rational and irrational) does market monetarism make sense.
In contrast, if the state simply prints fresh base money and spends it on a variety of stuff (and/or cuts taxes) that increases spending (private and/or public). If the initial reduction in capital spending was rational and stays rational, then spending on capital assets will automatically rise along with the rise in current spending. Assuming markets are working properly, that’s an ideal solution. In contrast, if there was an initial and IRRATIONAL cut in capital spending, that irrationality may well remain. But firms failing to invest will get driven out of business. That’s free markets. To the extent that firms continue to under-invest and for irrational reasons, simply printing and spending fresh base money does not solve that problem, but neither does market monetarism.
Conclusion: the hot potato effect works better where the state simply spends more on a wide variety of stuff (tax cuts included) rather than specifically on assets.
12. April 2015 at 10:53
@Ralph Musgrave: “the state buying up assets, and thus raising the price of assets”
But the primary effect of the Fed buying Treasury bonds, is not to raise the price of Treasury bonds. It’s not even a reliable effect. So this asset mispricing that you seem concerned about, doesn’t seem to be a significant factor.
12. April 2015 at 11:14
Sumner wrote:
“When I advocate something like QE or negative interest on reserves, I often get people complaining that this will not boost bank lending, or that we shouldn’t even be trying to boost bank lending. It almost makes me want to tear out my hair. What in the world does banking have to do with monetary policy?”
You’d save some hair if you actually made an effort to understand how the general public is able to spend more such that NGDP rises, and such that prices rise.
I’ve explained this many times on this blog.
At the moment in time after the Fed expands the monetary base via OMOs, at that point the general public’s cash balances have not yet increased. Sure, the owners of the member banks who have property rights over those reserves can be said to have had their cash balances increased, but as far as the general public is concerned, their total cash balances have not increased. If the belief is that more money and spending “stimulates” employment and output, then the general public’s economic activities are not yet “stimulated”.
The actual mechanism by which the general public’s money balances rise is via bank credit expansion. Lending takes place without a prior increase in money saving, and once that new bank credit is spent and redeposited into the banking system, that is what causes the aggregate money supply in the general public to increase.
If the banks did not increase their lending, then all of that QE would not lead to more money deposits being created. It will have little to no effect on either NHDP or aggregate prices. The “hot potato effect” on you personally, or any other individual personally, only takes place if there is actually an increase in money spending taking place by other people such that if you wait, you will lose in real terms. But that means that the hot potato effect presupposes bank credit expansion, for that is, again, how the general public acquires increasing money balances.
You cannot keep increasing your spending year after year unless you keep receiving more and more money year after year. You don’t get sent checks by the Fed. No, your money balances increase by way of a previous bank credit expansion which through the lending and redepositing increases the money balances of whoever is spending money on you (e.g. your employer).
“Yes, it may or may not boost bank lending, but it doesn’t matter, as monetary policy is about the hot potato effect.”
The hot potato effect presupposes the existence of increasing bank lending which is having the effect of increasing money balances and money spending in the general pubkic such that if you hold a sum of cash, it will decline in purchasing power.
While it is true that member banks receiving new reserves can in theory increase not their lending but the bank owner’s own dividends and personal expenditures, which would also release new money into the general public’s property, but if that was the ONLY way new money leaves the banking system, the size of the increase in bank reserves would have to be gigantic, and the general public would have to be willing to live with their money balances increasing only if bank owners live a more luxurious lifestyle.
Bank lending is the main channel by which most people earn more money today than people did say 50 years ago. It is not direct. In many cases new bank credit which leads to increased deposits has to be spent and respent a number of times by a number of people before your or my money balance can increase above what it otherwise would have been. Bernanke explained this in his blog post when he spoke about monetary inflation requiring 2 or 3 quarters beforw it has a sufficient impact on the general economy so as to raise prices and spending. Of course he’s wrong about it actually succeeding in raising output permanently.
12. April 2015 at 11:45
All,
Let’s not forget that the last 8 years of data have been collected in a banking environment where leverage has been forcibly reduced (Dodd Frank, Basle Accord, etc) and therefore the banking channel has not worked as in previous periods. Can we disentangle the exogenous regulatory effect from the simple view “banking channel does not work at the zero lower bound” ?
12. April 2015 at 13:26
Scott, I know you’ve commented on this (http://pubs.aeaweb.org/doi/pdfplus/10.1257/jep.24.4.45) Ohanian article before about how the Recession was caused by disequilibrium in the labour market, but do you actually think it has any explanatory power?
I agree with you that the Recession was mainly a function of poor monetary policy, and Milton Friedman identified the Great Contraction, however Ohanian presented evidence that suggested problems with the labour market even before the Great Contraction and hinted at ways certain tax policies harmed the labour market around 2007-8.
Would it be fair to say that the business cycle is, at its *core*, a function of neoclassical economics. But it’s when you combine failures in monetary policy that neoclassical fluctuations because Depression-esque downturns?
12. April 2015 at 15:24
Kevin, I’m a disciple of all good macroeconomists. As for the hot potato model, there are many things that one can spend excess money on, including goods, services and assets. The key is not to obsess with just one asset class.
Britonomist, It’s all about base supply, base demand, and NGDP. Everything else is a footnote.
Jim, There are still MMTers around?
And thanks for the links.
M. People are rarely completely indifferent, and in any case they won’t be indifferent forever. It’s actually the expected hot potato effect that really matters.
Vaidas, The goal should be to boost NGDP, not lending or jobs.
Ashton, In the US the business cycle is mostly monetary, say roughly 80%. In other countries real factors are relatively more important.
I do think that things like the minimum wage increase and the extended unemployment benefits played a small role, but it was mostly tight money.
12. April 2015 at 15:28
@Jim Glass
By purchasing MBSs, the FED is effectively lending money directly (because that is not a risk free asset), don’t you think? That channel does not seem to be in your two documents. That would be a 7th channel… Or Am I missing something here ?
12. April 2015 at 15:34
“Britonomist, It’s all about base supply, base demand, and NGDP. ”
And what happens when the central bank policy causes ‘base supply’ to be endogenous to demand, by always providing reserves when needed? Which is what many economists suggest actually happens.
12. April 2015 at 16:13
As near as I can tell, it’s another in a lengthening list of studies that back up the process by which I’ve said the QE plus reinforcing stimulus of the Chicago Plan of 1933 works since 2008.
12. April 2015 at 17:59
“What in the world does banking have to do with monetary policy?” I don’t know what the answer is with our present system of fiat money and a government monopoly on the issuance of currency (alas, I am not an economist), but when there was a combination of free banking and a gold standard banks issued banknotes, which were money–that is, media of exchange. That in itself is enough to make it appear to some observers that banks collectively were practicing “monetary policy”; but perhaps you will say that only *adjusting the unit of account* counts as “monetary policy.” (Is that the same thing as *adjusting the supply of base money relative to demand*?)
12. April 2015 at 18:26
“which confirms that monetary policy is about the hot potato effect (aka portfolio rebalancing channels) not bank lending.”
Shouldn’t the HPE be more oriented around people “rebalancing” from money into goods and services instead of companies rebalancing into riskier assets? Excessive portfolio rebalancing increases the size of financial sector undermining real gdp and also contributes to instability.
Heli’s of central bank emoney should be superior in terms of generating less financial excess and instability.
Studies of money neutrality dont take into account emoney heli’s because they are unprecedented.
12. April 2015 at 18:36
“And what happens when the central bank policy causes ‘base supply’ to be endogenous to demand, by always providing reserves when needed?”.
When needed for what? To hit an inflation target, change unemployment, change an exchange rate … please specify.
12. April 2015 at 18:41
Sumner wrote:
“As for the hot potato model, there are many things that one can spend excess money on, including goods, services and assets. The key is not to obsess with just one asset class.”
But how does the general public come to own an “excess” supply of money, given that the general public does not deal directly with the central bank, I.e. they do not receive any money from the central bank directly?
12. April 2015 at 18:41
Monetary policy is very much about banking.
12. April 2015 at 21:41
As MF and others point out, a very confusing point where for the first time Sumner defines “hot potato”, which most people would assume is money velocity, as something else (parenthetically) ‘portfolio rebalancing channels’ (sic, it even fails Spell-Check!).
Let me be more clear, since I have no vested interest in obfuscation. The Bank of England paper is bad for Sumner’s MM-is-good thesis. It is saying this: consumers don’t want loans, and any money a central bank prints and sells for bank paper (bonds, mortgages) is used for speculation by Wall Street and/or not lent by banks (but banks do collect interest on it from the Fed, in the USA, which enriches the banks). Consequently, QE does not stimulate the real economy on Main Street (only at best Wall Street firms and banks). Why is this bad for Sumner’s NGDPLT? Because it undercuts the argument: ‘print money and they–Main Street–will spend it and GDP will recover’.
12. April 2015 at 21:43
Re Sumner’s Econlog post, which has nothing to do with this post, I’ve been banned there for doing nothing more than I do here. FYI for those of you that think the internet is open:
Econlog: “Commenter ban in place. See your email for more information. ”
My comment: There are several concepts that people are bandying about in the comments. First, sticky prices have been shown to not be so sticky by Bils et al (2004), but this does not go to sticky wages. Not-so-sticky prices have been shown to still have an effect in Neo-Keynesian models by Virgiliu et al (2010), and finally Bils (2014) say sticky wages have an effect in a certain theoretical model. However, none of this goes to the question of GDP. It’s old, archaic thinking only concerning employment back when employment and GDP were linked. Simple example (which corresponds to reality): a corporation has robots that produce widgets. The robots need at best one very skilled and costly human to supervise them, but the corporation, to decrease their search costs to replace a human if said human dies, hires 10 such workers, nine of which stand around and only one does real work (typical btw in some construction projects too). During a recession, the corporation fires 3 out of 10 humans, leaving 7 to do the supervising work that really only needs one human. The three humans go on welfare. How is GDP effected? Not that much so long as demand for widgets recovers after the recession (assume widgets are consumed by the top 1% of society). That’s the modern economy, as opposed to the economy of your granddad.
12. April 2015 at 22:26
OT- Exercise: identify the numerous fallacies behind this paragraph (bonus points if you use the phrase ‘we owe it to ourselves’). It also goes to the heart of the fallacy behind Sumner’s NGDPLT. From http://www.project-syndicate.org/commentary/japan-monetization-government-debt-by-adair-turner-2015-03 : “Japanese government debt now stands at more than 230% of GDP, and at about 140% even after deducting holdings by various government-related entities … And it is debt that will never be ‘repaid’ in the normal sense of the word. … For Japan to pay down its net debt even to 80% of GDP by 2030, it would have to turn a 6%-of-GDP primary budget deficit (before interest payments on existing debt) in 2014 into a 5.6%-of-GDP surplus by 2020, and maintain that surplus throughout the 2020s. If this was attempted, Japan would be condemned to sustained deflation and recession. … Instead of being repaid, the government’s debt is being bought by the BOJ, whose purchases of ¥80 trillion per year now exceed the government’s new debt issues of about ¥50 trillion. Total debt, net of BOJ holdings, is therefore falling slowly. Indeed, if current trends persist, the debt held neither by the BOJ nor other government-related entities could be down to 65% of GDP by 2017 And because the government owns the BOJ, which returns the interest it receives on government bonds to the government, it is only the declining net figure that represents a real liability for future Japanese taxpayers.” Wow, how wrong can you be?
12. April 2015 at 22:29
@Scott
1. What does the “B” in FRB stand for?
2. For the millionth time, it’s not HPE. It’s a marginal increase in the exchange of financial assets for real goods and services by the non-financial sector with money as the intermediate medium of exchange.
3. Bank lending is only one mechanism for exchanging financial assets for goods and services.
4. QE works not just through the actual purchase of assets but also (and probably primarily) through expectations of future NGDP growth.
5. If you believe NGDP expectations don’t impact bank lending decisions, you’ve never done business with a bank.
13. April 2015 at 00:58
I think you should do a post on the hot potato effect — how it works, examples of the effect in action, evidence that it is the key mechanism
13. April 2015 at 06:13
Scott:
“The goal should be to boost NGDP, not lending or jobs.”
Yes, but why do we want to stabilize NGDP? There are two answers – jobs and lending.
13. April 2015 at 07:09
Britonomist, You asked:
“And what happens when the central bank policy causes ‘base supply’ to be endogenous to demand, by always providing reserves when needed? Which is what many economists suggest actually happens.”
I would certainly hope that is what happens, as any sensible target (inflation, NGDP, etc) makes the base endogenous.
Philo, In that example banks can impact the demand for gold, and hence the price level, But so can all other firms. I’d call “monetary policy” changes in the demand for gold done by government policymakers, such as a central bank.
I do think that banks play a more important role under the gold standard than they do under a fiat regime, if that’s your point.
CMA, Monetary policy that stabilizes NGDP does not cause financial instability. I have dozens of posts showing that helicopter drops are a horrible idea, a solution to a problem that does not exist.
Ray, And now you’re on the ZMP worker bandwagon? Dear God, what next?
dtoh,
1. I don’t know.
2. Why do oil prices drop after a big new oil discovery, if not the hot potato effect?
3. That’s right.
4. But of course expectations of future NGDP growth depend on expectations of future growth in the monetary base.
5. I agree that NGDP affects bank lending, just as all microeconomic variables are impacted by macro variables. NGDP also impacts the microwave oven industry.
Jeremy, I have lots of such posts, take a look at my short course on money.
Vaidas, Not because we want to target jobs or lending, but rather because we want to avoid artificially destabilizing jobs or lending. Hence our QE has no specific jobs or lending target, just the goal of creating a stable macro environment.
13. April 2015 at 07:25
Dr. Sumner,
You may not believe that the Quantitative Easing Policy (QEP) has anything to do with banks and bank lending but the Federal Reserve Bank (FRB) certainly does.
“The QEP consisted of three key elements: (1) The [Bank of Japan] changed its main operating target from the uncollateralized overnight call rate to the outstanding current account balances (CABs) held by financial institutions at the BOJ (i.e., bank reserves), and ultimately boosted the CAB well in excess of
required reserves. (2) The BOJ boosted its purchases of government bonds,
including long-term JGBs, and some other assets, in order to help achieve the targeted increases “Deprived of their traditional tool for policy, the four major central banks have begun adopting unconventional monetary policy, including forward rate guidance, asset purchases and programs to directly support bank lending.”[1]
“Over recent years, short-term nominal interest rates in many countries have effectively been driven to the zero lower bound.5 Deprived of their traditional tool for policy, the four major central banks have begun adopting unconventional monetary policy, including forward rate guidance, asset purchases and programs to directly support bank lending.”[2]
“What sort of monetary policy would change expected inflation, the expected path of short term rates, and/or the term of premium? First, a central bank can commit to zero beyond the period that their reaction function can normally call for, for what Eggerttson (2006) called ‘committing to be irresponsible’. Such a strategy – often termed “signaling” – is time inconsistent however: The central bank will want to renege on its commitment and return to normal policy when conditions improve. The second and third methods – outright asset -purchases and *bank lending* [Quantitative Easing] – can help resolve the apparent time inconsistency of a commitment to an announced policy path by changing the central banks incentives through its balance sheet.”[3]
Perhaps the FRB and the BOJ were wrong to put some much focus on bank lending and perhaps it was ineffective but it seems that bank lending was an intended element of QEP.
[1] http://1.usa.gov/1yqZXwq
[2] http://1.usa.gov/1FFMXLa
[3] http://bit.ly/1yByZBM
13. April 2015 at 07:26
Scott,
I agree.
13. April 2015 at 07:35
@Scott
2. Why do we have to keep going over this. OMP are different than an oil discovery, helicopter drop, or bumper apple crop. OMP are an exchange, the latter are not. They are fundamentally different.
4. “Expectations of future NGDP growth depend on expectations of future growth in the monetary base.” Yes, but that’s like saying, “Expectations of flooding depend on expectations that people will be using umbrellas.”
As I keep saying, HPE while predicatively accurate has causation reversed. People don’t buy more stuff because they have more cash; rather they acquire more cash because they intend to buy more stuff.
13. April 2015 at 08:37
@Ray Lopez: “for the first time Sumner defines “hot potato””
It never came up, in his thousands of previous posts? Not in the short course on money? Well, thanks for doing that extensive and detailed research for us.
“I have no vested interest in obfuscation”
Of course not. Obfuscation requires that you understand, first, before you try to conceal. You never make it through the first step.
“The Bank of England paper is bad”
Ah! Yes. Finally, something we agree on.
“I’ve been banned [at EconLog] for doing nothing more than I do here.”
Congratulations! And, in your case, very well deserved!
13. April 2015 at 09:15
Easy to confuse the mechanism with the policy. Certainly CBs can alter lender behavior by targeting the rates in a fractional reserve lending system, but they do so in order to target other variables like unemployment and inflation, not because they want more lending.
This is really just the “Fed policy is reflected in nominal interest rates” mistake in another guise.
Complaining a Fed policy won’t increase lending is a bit like a farmer saying “hey, why are you adding fertilizer? that won’t increase irrigation!”
13. April 2015 at 09:22
dtoh,
2. Suppose there is no such thing as money. What happens to the barter value of oil after an oil discovery? Why would HPE apply to oil, but not money?
4. NGDP is driven by the monetary base, the monetary base is not driven by NGDP (at least, not nearly as much).
To say HPE has causation reversed doesn’t really make sense. That would suggest that consumers print money, not the Fed.
13. April 2015 at 09:24
OMG. I agree with Major Freedom — pretty much without reservation.
However, I would also add that the study’s “findings” are logically obvious and simultaneously part of the short comings of QE. Given that QE lowers long term interest rates, and that means lowering the profitability of loans, we should not be surprised that QE actually hinders banking activity. This ends up ultimately muting the “hot potato” effect brought on by the increase in bank reserves.
This is precisely why English NGDP grew better AFTER QE stopped than during QE. It will also be the reason the US NGDP will grow better after QE is stopped and rates rise here.
I like QE because it primes the pump (monetary base) to prepare for money supply expansion. However, as a single golden arrow to save an economy… well, an understanding of the monetary base, money supply, and the banking role in that process just doesn’t support that belief.
13. April 2015 at 13:02
Jim Glass:
“When needed for what? To hit an inflation target, change unemployment, change an exchange rate … please specify.”
Needed for when banks are short of reserves to satisfy liquidity requirements, apparently. If that’s the case, then any arbitrary increase in the base is utterly irrelevant as it doesn’t affect bank behavior at all (since they could have got that extra base anyway if they decided to extend more credit). While the portfolio re-balancing effect is a broad money story, not base money story. It’s people using their own deposits at banks they got from selling bonds to repurchase assets, not banks using their own reserves at the central bank. It’s broad money that matters.
13. April 2015 at 14:00
@ Britonomist
Needed for when banks are short of reserves to satisfy liquidity requirements, apparently. If that’s the case, then any arbitrary increase in the base is utterly irrelevant…
“Apparently”? I don’t understand the uncertainty in that. Are you saying that the central bank can’t control “base supply” to hit, say, an inflation or exchange rate target? Because money supply always must meet ‘liquidity requirements’? I’m not sure what that means either.
But it seems to suggest the idea floating around in some quarters that the supply line for money is horizontal, so that this model applies…
http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/04/reverse-engineering-the-mmt-model.html
Is that your suggestion?
13. April 2015 at 14:35
Jim, I say ‘apparently’ because this is what various economists tell me. Since I don’t directly work at an executive position in a central bank I cannot directly confirm this is the case. Again, this is going back to this paper: http://www.bankofengland.co.uk/publications/Documents/quarterlybulletin/2014/qb14q1prereleasemoneycreation.pdf
“As discussed earlier, the higher stock of deposits may mean that banks want, or are required, to hold more central bank money in order to meet withdrawals by the public or make payments to other banks. And reserves are, in normal times, supplied ‘on demand’ by the Bank of England to commercial banks in exchange for other assets on their balance sheets. In no way does the aggregate quantity of reserves directly constrain the amount of bank lending or deposit creation.”
This has nothing to do with MMT btw.
13. April 2015 at 16:29
Scott H:
Be sure you keep in mind that the lowering of interest rates is not a cause for less lending. Banks lend according to real returns, not only nominal returns. A nominal interest rate of 2% in a world where prices are falling say 3%, is about as lucrative a proposition as a nominal interest rate of 5% in a world of 0% price inflation.
Lower interest rates do not cause less or more lending. They are in fact determined by the going rate of profit in the economy, with an adjustment for risk.
————–
Sumner asked:
“Why do oil prices drop after a big new oil discovery, if not the hot potato effect?”
How do oil and oil based products get from the oil companies to the general public?
They sell it.
For what? In this case for money, or a claim to a future sum of money, I.e. a debt instrument.
How does money get from the member banks to the general public?
They sell it.
For what? In this case they sell it primarily for a claim on a future sum of money, I.e. a debt instrument.
Once the money is sold by the member banks, it is spent and redeposited back into the banking system. THAT is how most money is created and how greater sums of money become the property of the general public.
Now ask what would happen with this process if the member banks sold very little money. The hot potato effect would only become as pronounced to the extent to which the banks are selling money. If the banks are not selling much money, the hot potato effect is not as pronounced. This is the case no matter how many billions or trillions of dollars they have come to own reserves because of QE. It is only to the extent that the banks are willing to lend, will QE transform into more money in the general public’s property, along with higher aggregate prices and spending.
13. April 2015 at 16:43
There is no mechanistic, direct correlation between reserves and total spending. Increasing reserves by say X% does necessarily lead to any constant f(X)% NGDP.
If banks are not willing to lend more and more over time, then NGDP will almost certainly collapse no matter how much reserves increase.
If you are faced with two alternative projects, one that is expected to earn a return of negative 1% and the other that is expected to earn a return of positive 1%, and the project managers tell you that the maximum interest payments they can make is $1000 per year, then if you do the math, any sum of money you have stored in your basement, which you store for the purposes of making investments, that is greater than $100,000 will just sit there. If someone sold you $300,000 more in exchange for your furniture, that won’t affect your investment opportunities returns or your ability to lend.
Now you can of course change the assumptions of the above and say “Pshaw, I will just spend everything above $100,000 on vacations and sports cars and mansions for myself, that is how all those reserves in my basement can make their way into the broader economy.”
Sure, you can pretend that is how reserves could potentially translate into more spending given you are not lending more than $100,000, but then I will just change the assumptions and say sorry, you have shareholders to answer to. They want you to earn them money. They don’t want you to declare dividends. And even if you did, that will just result in them depositing that money right back with you, and hold it as saved money, or they tell you “Invest this in interest earning projects”.
Do you see the importance of bank lending as it pertains to monetary policy now? Some people on this board get it.
13. April 2015 at 17:22
MF — I’m not talking about a generic “interest rates going down”. I’m talking about the spread between long term and short term interest rates going down. That is the bank’s profit margin — they borrow short and lend long. QE brings down long term interest rates.
13. April 2015 at 18:45
Scott H:
Oh OK, I get that.
If long term rates go down because of the Fed introducing a higher demand for long term bonds via QE, then if long term profits are assumed unchanged, then there will be less of an incentive to lend long, and more of an incentive to borrow and go long equity.
Bernanke did mention more than once that QE was intended to bring about the so-called “wealth effect”.
14. April 2015 at 02:06
OT: Reading Calomiris book “Fragile By Design” and he makes a point in several passages about the “inflation tax”. It’s not as high as I thought, but still high. For example in Brazil, from 1947-87 (during the populist strongman era) the inflation tax (of which government got 75%, the rest went to private banks) was: “figure 13.2 we present estimates of the magnitude of the Brazilian inflation tax, as it is generally defined, from 1947 to 1986.34 During Vargas’s term in power from 1951 to 1954, inflation began to accelerate as he increased spending in order to satisfy his constituents without a concomitant increase in taxation. The inflation rate during these four years (1951-54) averaged 17 percent (compared to 7 percent for the four previous years), which generated an inflation tax that averaged 3.3 percent of GDP”
The inflation tax went as high as 8% of GDP in 1961-63 but generally was about 4%/yr during this 40 yr period. Not as bad as I thought, though 4% of GDP is a hefty drag on the economy (and for the USA would mean, with its low growth rate, causing a net negative growth rate). This inflation tax also no doubt is calculated conservatively to exclude ‘menu costs’ and other indirect costs of inflation. If anybody has further insight please post. B. Cole is dancing a victory lap now…inflation is not so bad, lol. But Sumner’s plan will bring ruinous hyperinflation if we go past the tipping point, that’s for sure.
14. April 2015 at 02:07
QE has not been about bank lending and that’s an important point. However, the paper’s contention that this absence of lending has something to do with “flighty deposits” seems absurd from a banking perspective.
As to the broader generalization of what monetary policy is not … perhaps
14. April 2015 at 03:39
@MF, Scott H
You do realize that there is no empirical support for the claim “QE brings down long term interest rates.” ? This blog and many other blogs have shown that many times over
@Ray
I have never seem a quantitative assessment of the inflation tax in Brazil, but when inflation went down in 1994, virtually all banks went down, the government had to save all the state banks and some of the private banks as well. That cost 20% of GDP in new public debt. You have to consider that we had 30% a month inflation. For the amusement of the readers here, whenever I talk about this, I tell people the first 6 nominal salaries I made in my professional life (the currency was called Cruzeiro) :
First month: Cr$ 30.000.000,00
second: Cr$ 37.000.000,00
third: Cr$ 46.500.000,00
fourth: Cr$ 56.000.000,00 (then they cut 3 zeros and called it Novo Cruzeiro)
fifth: NCz$ 68.000,00
sixth: NCz$ 82.000,00
When it was finally converted to something stable, all these numbers were roughly equivalent to 554 dollars …
14. April 2015 at 07:07
@Tall Dave
You get recessions because wages and prices are sticky whereas the price of financial assets is not sticky. If this were not the case, monetary policy would be both unnecessary and ineffective.
OMP works because it induces the non-financial sector to exchange more financial assets for real goods and services, thus pushing up aggregate demand (NGDP).
OMP is not like an oil discovery or helicopter drop where some party simply receives mores goods or more money. It requires that a counter-party is induced to enter into the trade. This doesn’t happen because someone want to hold more money. It happens because someone wants to exchange financial assets for real goods and services. Money is only the intermediate medium of exchange for effecting this exchange.
Scott thinks that people buy more goods and services because OMP has magically increased the amount of money they hold (like an oil discovery or bumper potato crop), but this is not the way it works. It’s the other way around. People increase the money they hold because they intend to buy more goods and services.
14. April 2015 at 07:34
In the modern economy, most money takes the form of bank
deposits. But how those bank deposits are created is often
misunderstood: the principal way is through commercial
banks making loans. Whenever a bank makes a loan, it
simultaneously creates a matching deposit in the
borrower’s bank account, thereby creating new money.
http://www.bankofengland.co.uk/publications/Documents/quarterlybulletin/2014/qb14q1prereleasemoneycreation.pdf
14. April 2015 at 09:38
dtoh,
Money is just another asset, an asset whose available quantity the Fed can affect through OMO. Yes, OMOs are exactly like an oil discovery: there is now more money. It’s not “magic,” just fiat money.
“People increase the money they hold because they intend to buy more goods and services.”
And people buy oil products to drive their vehicles, and buy more oil products when intend to drive more. Doesn’t matter what they do with it, consumers can’t produce more oil — or more money.
http://www.themoneyillusion.com/?p=23314
14. April 2015 at 09:47
Also the counterparty does not make OMOs special, the counterparty is just as necessary for oil discoveries to affect oil prices.
For instance, it was recently estimated there are a thousand tons of diamond a year being created in Saturn’s deep atmosphere. This has no effect on diamond prices because no one is or conceivably ever will be selling that diamond, at least within investor timeframes.
14. April 2015 at 10:16
Scott H.
“Given that QE lowers long term interest rates, and that means lowering the profitability of loans, we should not be surprised that QE actually hinders banking activity. This ends up ultimately muting the “hot potato” effect brought on by the increase in bank reserves.”
Thank you!
In about 2010, the Fed undertook “Operation Twist” with the intention of driving the yield curve flatter. At the time, I was arguing that this was exactly the wrong prescription. The level of rates doesn’t drive money growth. The change in rates may have some influence. But, the biggie is the shape of the yield curve and the magnitude of spreads.
14. April 2015 at 10:52
Scott,
When the person says “It’s not about the money” everyone knows that in fact it is about the money. For by the sake of saying it the person has revealed that money is a consideration, either for what money the person is getting or what money the person is forsaking.
So it is with you. For you want so desperately to believe that monetary policy can exist independent and neutral of the profit seeking, creative destructionists that prowl the capitalistic landscape. You want to believe this because recognition of this reality makes monetary policymakers players in the game. I understand why you don’t want that responsibility but you need to man up and accept that if you own the policy you own the consequences of it, including how it is gamed by profit seeking agents.
Monetary policy that looks the other way as credit expands and then rushes to rescue lenders from bad lending decisions is morally and ethically bankrupt. So is with monetary theory that pretends lenders or borrowers should never face a margin call or otherwise face the consequences of their own greed or stupidity.
If you wish to believe that there is no bubble in credit expansion then you must also accept that there is no bubble in credit contraction. Yet from what you argue you do not want to allow credit contractions. Credit contractions cause recessions and NGPLT exists to not allow recessions!
If NGDPLT is about sustaining credit growth and credit growth originates from banks, how can monetary theory not be about banking?
14. April 2015 at 11:47
dtoh, So the people who produce oil don’t sell it? Of course they do. How is that not an exchange?
Umbrellas have no causal impact on rain, base growth has a causal impact on NGDP.
Scott, I agree that some people expected far too much from the QE programs. There are much better tools.
JKH, I’m not qualified to comment on the flighty deposits.
Dan, Obviously you are new here. I’d suggest you look at some of my older posts to find out my views. I certainly do not hold the views you claim I do, indeed I agree those views are absurd.
You said:
“Yet from what you argue you do not want to allow credit contractions. ”
Oh yes I do!
14. April 2015 at 16:28
Jose:
“You do realize that there is no empirical support for the claim “QE brings down long term interest rates.” ? This blog and many other blogs have shown that many times over”
Impossible, because you cannot observe the counterfactual world where QE did not take place.
What you really mean by “empirical support” is a particular theory that you are utilizing to understand empirical data. Empirical data does not on its own tell us anything. We need a theory that is integrated with historical data in order to come to understanding of history.
Long term rates can in theory be brought down counterfactually by way of an exogenous counterfactual increase in the supply of long term loans. It is possible for QE to encourage the bankers to increase the rate at which they extend long term loans and at the same time accept the consequences of receiving a lower rate. I don’t mean lower in the sense of a temporal day to day or month to month decrease in rates over time. I mean the rates are lower than they otherwise would have been had QE not taken place.
This of course includes long term rates actually rising over time. They are rising over time, but because of QE, they are rising less than they otherwise would have risen.
What “empirical evidence” really is, is an excuse for people not to think like an economist, not think very much at all, and rest content on their laurels falsely believing that the theory they had in mind all along has been ex post vindicated, when in reality it is that theory that they used to understand the historical data and make claims that the historical data shows their theory is correct.
Economics is not a positivist science. It is not empirical. The reason it is not an empirical science is because there are no constant causal relations in the field of human action. This differs from the hard sciences, where the subject matter does behave according to constant causal relations between variables.
When you tell me that this blog has shown that there is no empirical evidence that QE brings down long term rates, all I read, and all I should read, is that you are saying this blog does not promote the a priori theory that QE brings down long term rates, but rather the a priori theory that QE does something else in a constant causal manner.
I see no good reason to deny the possibility that a central bank buying long term bonds will never have the effect of making long term rates lower than they otherwise would have been.
15. April 2015 at 05:57
@scott
“Umbrellas have no causal impact on rain”
Exactly my point. Base alone has no causal effect on real growth. (And NO one cares about nominal growth other than for it’s usefulness as a policy target in order to achieve the actual goal of higher real growth.) Base is merely correlated to real growth because it is used as a medium of exchange to facilitate the exchange of financial assets for real goods and services. Monetary policy works because it induces an increase in that exchange of financial assets for real goods and services. The increase in base is merely a by-product of that transmission mechanism.
If you assume, that like money (in a positive nominal rate environment), the demand for oil is inelastic, the an oil discovery will only result in the price of oil dropping relative to the price of everything else. Similarly merely increasing the base, e.g. via a helicopter drop, will only result in the price of money changing relative to everything else, i.e you’ll get only nominal growth.
Maybe you’re arguing in a helicopter drop scenario that HPE is the cause of an increase in NGDP….that’s correct, but I don’t think that’s your argument. I think your contention is that HPE is the transmission mechanism by which higher real growth is achieved through actual real world monetary policy.
If you base your reasoning on oil discoveries or bumper potato crops though, I can see why you could be misled into that conclusion, especially since, as I have said before, HPE is perfectly fine as a predicative hypothesis even if it confuses causality.
15. April 2015 at 08:28
@dtoh: “Base alone has no causal effect on real growth.”
Surely you’ve been here long enough to know that the argument has two distinct parts: (1) growth in the monetary base can target NGDP; and (2) nominal shocks can have real effects. You seem to agree on the first one, so are you actually denying that the nominal economy can affect the real economy?
“merely increasing the base … you’ll get only nominal growth.”
Everybody is already aware that you cannot continue to increase the rate of money printing in order to get higher real growth. This experiment was tried in the 1970’s, and, as you say, the extra money merely translates into price inflation.
But you’re leaving out the interesting case, which is the effect of money printing when inflation is very low and/or the economy is depressed.
“HPE is the transmission mechanism by which higher real growth is achieved through actual real world monetary policy.”
Yes — but specifically in the case of an economy that is suffering from a negative nominal shock. The nominal shock, when combined with sticky wages and debts, causes real GDP to be below potential. In that special case, increased nominal growth will allow the real economy to return to potential. HPE is the mechanism by which increases in the monetary base cause increases in NGDP. A depressed economy is the context in which increased NGDP results in increased RGDP instead of inflation.
15. April 2015 at 09:51
“The nominal shock, when combined with sticky wages and debts, causes real GDP to be below potential. In that special case, increased nominal growth will allow the real economy to return to potential. HPE is the mechanism by which increases in the monetary base cause increases in NGDP. A depressed economy is the context in which increased NGDP results in increased RGDP instead of inflation.”
Is a simpler way of saying this “A lack of adequate liquidity (monetary base) can have impacts on RGDP”?
15. April 2015 at 10:09
Good points, Don.
“A depressed economy is the context in which increased NGDP results in increased RGDP instead of inflation.”
I might quibble with the inflation part though — I think we’d prefer to say “you get more inflation, but the beneficial tradeoffs in employment and RGDP are preferable at some levels of inflation that are higher than in a non-depressed economy (or higher than the ECB is currently allowing).”
That seems to be the point at which people fall off the market monetarist wagon, it’s that tricky question of “how do we define the ideal monetary policy?” Scott’s classic example of what “what happens to monetary policy in an oil shock?” is an easy case to understand at the extreme, but ECB 2015 policy is murkier.
Still, it’s pretty clear that ECB policy is tighter than that which would optimize the combination of RGDP/employment/price stability, rather than sacrificing the first two on the altar of the latter as they seem wont to do.
15. April 2015 at 12:07
@Don Geddis
To be clear, I don’t agree with the contention that HPE is the transmission mechanism.
1. Fed action through higher financial asset prices and/or changed expectations of growth induces the non-financial sector to exchange more financial assets for real goods and services, i.e. that is the increase in RGDP.
2. If the non-financial sector were not so induced, there would be no willing counter-party to OMP (outside of the financial sector), and OMP would only be possible to the extent that the financial sector exchanged assets for Fed reserves (which would have no impact on the economy.)
3. To the extent that the increase in RGDP changes the supply and demand balance for real goods and services (or expectations thereof), it may cause some degree of inflation which would be additive to NGDP beyond the real growth.
4. So to reiterate, the increase in the base (or money supply) is not what drives increased RGDP or NGDP, but rather it is the action of the non-financial sector to exchange more financial assets for real goods and services. The non-financial sector acquires money to effect that exchange. They do not make the decision to purchase goods and services because they have the money. It’s the other way around.
You do raise an interesting point as to whether in a depressed economy higher NGDP would engender an increase in real growth. Since higher NGDP is a result of higher RGDP under the way monetary policy is currently conducted, that situation is unlikely to occur, but it’s could be tested if the primary monetary tool were helicopter drops rather than OMP. In that case, I suspect it would work once or twice, but once it became known and expected, I think that not only nominal but also real wages and prices would become sticky as well.
16. April 2015 at 05:48
dtoh, You said;
“Monetary policy works because it induces an increase in that exchange of financial assets for real goods and services.”
I don’t agree. More base money causes a rise in NGDP. And if there are sticky wages then a rise of NGDP causes a rise in output. And a rise in output causes a rise in purchases.
16. April 2015 at 07:23
Scott,
Then tell me why the counter-party to OMP enters into the trade.
17. April 2015 at 04:09
dtoh — Same reason that counterparties buy oil from a new oil strike. There is a market for oil, there is a market for financial instruments. People buy and sell either all the time without the Fed “inducing” anyone.
The Fed is just another market participant, the only thing special about them is that they can print money. That’s why they’re called “open” operations.
it is the action of the non-financial sector to exchange more financial assets for real goods and services.
Yes, in the sense that cash is a “financial instrument.” More cash is buying less real goods and services.
17. April 2015 at 06:46
dtoh, Because they are offered a good price.
17. April 2015 at 10:11
Scott,
And what is a good price? I’ll answer that.. one which induces them to exchange financial assets for good and services.
()In a positive nominal rate environment, there is no price which is good enough induce them to exchange assets for money.)
17. April 2015 at 10:20
Or I should say to exchange assets for money unless they intend to buy goods and services with the money.
17. April 2015 at 11:07
Again, that only makes sense if you ignore the fact the Fed is conducting transactions at market prices and doesn’t offer any “inducements” that other participants are not also offering.
You haven’t offered any explanation or evidence as to why Fed operations are magically different than other transactions, such that they induce people to go out and buy goods and services.
17. April 2015 at 11:14
Anyways, it’s trivially easy to prove your theory doesn’t work — suppose the CB simply sends every person $10,000 in cash. According to your theory, that can’t be inflationary because the monetary base doesn’t drive inflation, and we didn’t exchange any financial instruments for goods and services.
17. April 2015 at 14:20
TallDave:
“Again, that only makes sense if you ignore the fact the Fed is conducting transactions at market prices and doesn’t offer any “inducements” that other participants are not also offering.”
That is not a “fact.” That is a theory, and it is false.
There cannot even be market prices in a world with central banking. The very fact that the Fed is in existence makes market pricing an impossibility. This is why we have the business cycle. There is a perpetual and constant hampering of real resource allocation because people do not have access to the needed information and be constrained to all individual preferencea that only a market process can provide.
To have a non-market, indeed an anti-market institution like centralized printer of money, for the purposes of buying treasury X, makes a market for treasury X an impossibility.
A market price is a price that prevails in a free market. We do not live in a world with a free market. No, it is not the case that just because a free market does not exist, that you can steal the term anyway and claim equality between this world and an actual free market world, simply so that you can throw the term around as if you are entitled to it.
18. April 2015 at 00:58
@Tall Dave
Please reread my comments. I’ve been very clear that if helicopter drops (i.e. sending every person $10,000 in cash) were the tool used for monetary policy, then HPE would be the transmission mechanism for higher NGDP. In other words, the increase in the base would cause the increase in NGDP. I totally agree with that.
What I’ve said is that when OMP (not helicopter drops) are used as the tool, then you have a different transmission mechanism and different causality, i.e. a marginal increase in the exchange of financial assets for goods and service by the non-financial sector.
18. April 2015 at 10:25
dtoh, You said:
“And what is a good price? I’ll answer that.. one which induces them to exchange financial assets for good and services.”
I don’t follow your logic. How does this relate to OMPs? And exchange of assets for goods is not an OPMP.
18. April 2015 at 13:56
Scott,
Scott,
OMP is not an involuntary exchange with the counter-party. The counter-party has to have a reason for the exchange. Does the counter-party enter into the trade in order to hold more money. Or..does the counter-party enter into the trade with the intention to use the money to buy goods and services.
And to be clear, I’m not talking about financial intermediaries who just broker the trade. I’m talking about the ultimate counter-party in the non-financial sector.
18. April 2015 at 22:19
What I’ve said is that when OMP (not helicopter drops) are used as the tool, then you have a different transmission mechanism and different causality,
I’m not sure you see why that’s problematic, though. In both cases, the monetary base has increased. So in both cases, the HPE for an increased monetary base arises to explain the inflation. Your magical Fed explanation is superfluous.
Again, you haven’t identified anything magical about Fed actions. People trade bonds all day voluntarily.
I’m talking about the ultimate counter-party in the non-financial sector.
They’re not a party to anything. They just have some money that is now worth a little less, just like if it were gold and someone produced more gold.
19. April 2015 at 08:26
@Tall Dave
I’m not sure you see why that’s problematic, though. In both cases, the monetary base has increased. So in both cases, the HPE for an increased monetary base arises to explain the inflation.
In one case people and firms sell financial assets (and I use this term in the broad sense) specifically for the purpose of buying goods and services. This translates directly into increased real aggregate demand (Scott may nitpick that there is no such thing as real AD… so RGDP if that makes him happier). Helicopter drops, I suspect, if used as a regular policy tool, will simply result in money becoming super-neutral in the short term so all of the increase in NGDP will be nominal.
Again, you haven’t identified anything magical about Fed actions. People trade bonds all day voluntarily.
Yes, but when you and I trade bonds, it’s a wash. There is no change in the aggregate holding of financial assets by the non-financial sector (e.g. no increase in indebtedness) and generally no impact on aggregate demand. When the non-financial sector sells financial assets to the financial sector (e.g. the Fed or private financial institutions), it generally results in an increase in AD as the purpose of the exchange is to acquire money for the purchase of goods and services. (It’s actually a little more complicated in the case of private financial institutions (e.g. banks) and depends in part on their reserve position and whether they reduce other asset holdings.)
They’re not a party to anything. They just have some money that is now worth a little less, just like if it were gold and someone produced more gold.
If there is no ultimate counter-party in the non-financial sector, then OMP simply end up increasing ER with no effect on the economy.
It may be helpful to think of financial assets as being fungible. The transmission mechanism is not limited to the simple case of the non-financial sector selling Treasuries out of a portfolio directly through a primary dealer to the Fed. It is just as likely to be effected by a bank selling Treasuries to the Fed and replacing them with other assets (e.g. an increased draw down on a line of credit by a corporate customer or increased credit card balances by individual customers.) In either case, it results in a reduction of the net holdings of financial assets (or increased indebtedness if you want to think about it that way) by the non-financial sector specifically for the purpose of buying real goods and services.
30. April 2015 at 07:15
dtoh,
Nonresponsive. You just can’t have HPE working sometimes when the monetary base increases and sometimes not. The money buys “real goods and services” either way. The fixation on “financial assets” doesn’t make any sense, wage increases could have the same effect.
If there is no ultimate counter-party in the non-financial sector, then OMP simply end up increasing ER with no effect on the economy.
Again, this “ultimate counterparty” nonsense is like claiming people don’t change the price of gold when more gold is discovered.
I think your entire misunderstanding can be explained with one statement: Prices do not require transactions to change.