Goodhart on NGDP targeting
Here’s Charles Goodhart in the Financial Times:
But observations of policy-making over the years raise doubts that an ad hoc entry into a new policy regime will be followed by a nimble exit when the appropriate time comes. The fear is that, once the sell-by date of these initiatives passes, central bankers will be acting contrary to everything learnt, painfully, in the 1970s. They will be relating monetary management to real variables on a longer-term basis. In the end, any short-term benefit will be dwarfed by the long-run pain as they push inflation higher in the vain pursuit of a real economic objective.
NGDP does not target real variables; it targets a 100% nominal variable. And it is not an expedient—the goal is to permanently replace the discredited inflation targeting regime, which forces central banks to lie about what they are really trying to do.
While there may now be a case for some further temporary monetary expansion, this can be done within the context of the present flexible inflation target.
Central bankers would be better employed by improving unconventional instruments of monetary policy. The UK’s funding-for-lending scheme is a good start, as it offers a route to stimulating aggregate demand that bypasses the clogged arteries of conventional stimulus. The BoJ already has a significant portfolio of loans on its books, and the Fed would be wise to follow if the pace of the US expansion remains tepid.
It’s a big mistake for central banks to get in the business of influencing the amount of credit in the economy. Much better to focus on producing a stable growth path for nominal income and let the markets decided how much of that income gets allocated into borrowing and lending. Instead of ad hoc “unconventional instruments” that are unlikely to work in an emergency, why not use a policy target that reflects the central bank’s actual goals. Then you won’t ever need to use unconventional instruments to make up for the failure of inflation targeting.
Adopting a nominal income (NGDP) target is viewed as innovative only by those unfamiliar with the debate on the design of monetary policy of the past few decades. No one has yet designed a way to make it workable given the lags in the transmission of monetary policy and the publication of national income and product.
I had thought that “targeting the forecast” was one of the cutting edge ideas in monetary economics. If so, then data lags are not a problem. It’s a shame that people like Michael Woodford have not kept up with research in monetary economics over the “past few decades.”
Rather, a NGDP target would be perceived as a thinly disguised way of aiming for higher inflation. As such, it would unloose the anchor to inflation expectations, which could raise, not lower, interest rates by elevating uncertainty about the central bank’s reaction function.
NGDP targeting is not intended to raise the rate of inflation. For any long run inflation target, there is a NGDP target path that produces the same expected rate of inflation. Ex post inflation might be higher, but it’s equally likely that it would be lower. And the inflation uncertainty would not raise interest rates, as the debt markets care much more about NGDP than inflation. In the long run nominal interest rates are more closely correlated with the long run trend rate of NGDP growth than inflation.
We do not know, and cannot predict, what will be the sustainable rate of real growth in our economies. Let’s hope it is well above the relative stagnation observed in recent years in the UK, US, and Japan. But it would be over-optimistic to believe our economies can permanently revert to prior faster growth. In the short run, excess monetary expansion might temporarily lead to a burst of growth. But the likely implications of a dash for growth and the abandonment of an inflation target would at some point unhinge the government debt market.
Again, Goodhart is assuming debt markets care about inflation. They don’t. They care about NGDP growth. As long as NGDP growth is around 4%, long term nominal rates will remain relatively low. That’s the case regardless of whether the 4% NGDP growth is associated with o% RGDP growth and 4% inflation, 2% RGDP growth and 2% inflation, or 4% RGDP growth and 0% inflation.
Furthermore, Goodhart seem to believe that we need to estimate the long term RGDP growth rate in order to choose a NGDP target path. But one of the strongest arguments for NGDP targeting is that it avoids the need to estimate trend real GDP growth, or output gaps, or the natural rate of unemployment.
HT: Nicolas Goetzmann
Tags:
30. January 2013 at 13:26
I assume that if supply factors (productivity growth) were able to create 10% RGDP growth, then you would not advocate a 5% NGDP growth path. Can you explain how you choose the annual NGDP growth rate target if it is not related to RGDP growth? Or are you just saying that fluctuations in RGDP growth will be small, and a little more or a little less inflation here and there is not a big deal?
30. January 2013 at 13:33
Goodhart is only rehashing his Davos speech:
http://thefaintofheart.wordpress.com/2013/01/22/the-big-guns-come-out-against-ngdp-targeting/
30. January 2013 at 14:13
Ssumner,
“Much better to focus on producing a stable growth path for nominal income and let the markets decided how much of that income gets allocated into borrowing and lending. ”
Basics of banking: banks don’t lend out anyone’s money, they don’t lend out base money, or anything really, they create deposits against loans, so income cannot be “allocated to lending”. Money has worked like this since the earliest historical records, there is no allocation of anything, just creation of credit.
OTOH borrowing (really creating credit money by banks) does impact spending and incomes. And yes, you are right that Fed has a weak impact on credit, so it has weak impact on income and NGDP.
30. January 2013 at 14:31
Goodhardt’s critique of an “ad hoc entry into a new policy” would have more bite if his preferred policy wasn’t, you know, an ad hoc entry into a new policy. (“The BoJ already has a significant portfolio of loans on its books, and the Fed would be wise to follow…”)
Instead of trying to make unconventional policy (which hasn’t worked) work better, maybe a better idea is to make conventional policy work better?
30. January 2013 at 14:40
I am pleasantly surprised, but I get the sense that the enthusiasm for NGDP targeting has been receding a bit in the UK recently – also these articles:
Chris Giles, FT: http://www.ft.com/cms/s/0/726213c6-6878-11e2-9a3f-00144feab49a.html#axzz2JUnEwEAY
Stephanie Flanders (BBC, housetrained and therefore probably quite well connected): http://www.bbc.co.uk/news/business-21246665
Martin Weale (MPC member): http://blogs.independent.co.uk/2013/01/30/martin-weale-ngdp-target-is-playing-with-fire/
I do agree with you about targeting the forecast though, Scott. Actually, the slow emergence of the final NGDP figures is an advantage for a central bank in some ways, because it allows plausible deniability to go on longer (which, is by the way, why I think that an official inflation – or NGDP – forecast should be done independently of the central bank – say by the Office for Budget Responsibilty in the UK).
30. January 2013 at 14:43
@Marcus Nunes, or his VoxEU column: http://www.voxeu.org/article/monetary-targetry-might-carney-make-difference
30. January 2013 at 14:43
J, You said;
“I assume that if supply factors (productivity growth) were able to create 10% RGDP growth, then you would not advocate a 5% NGDP growth path.”
You assume wrong. If you are referring to the US, I would favor 5% NGDP growth regardless of the RGDP trend line.
Marcus, I do recall that now.
OhMy, You said;
“Basics of banking: banks don’t lend out anyone’s money”
Yes they do. Banks are financial intermediaries.
Max, Well put.
30. January 2013 at 14:47
See also this critique of NGDP targeting from New Zealand economist Matt Nolan: http://www.tvhe.co.nz/2013/01/31/inflation-stickiness-demand-and-judging-the-success-of-monetary-policy/
30. January 2013 at 14:57
Ssumner,
“Basics of banking: banks don’t lend out anyone’s money”
Yes they do. Banks are financial intermediaries.
Um, no. Amazing that you would go on record with such nonsense. Krugman is equally lost, that is why he lost the debate with Keen although in the end he at least conceded on banking.
If you sign a loan, and the bank credits your account, where did the loan come from? Where did the numbers in your ledger come from? Nowhere. Please read up on the history of money since Babylonia, money has been a debt record since… always. And debt is created the moment you sign a loan. Equally, a bank that credits your account against the loan you just signed is not debiting anyone else’s account, so no, you are not borrowing anyone’s money.
Read comments and the post itself:
http://www.interfluidity.com/v2/3402.html
Read history of money and what is banking:
http://moslereconomics.com/mandatory-readings/what-is-money/
30. January 2013 at 15:08
Dr. Sumner:
“why not use a policy target that reflects the central bank’s actual goals.”
I feel stupid for asking this, because I am sure it’s been answered a billion times before, but don’t the central banks “actual goals” include 2% consumer price inflation target? The statement you made almost sounds as if the Fed’s actual goal is to stabilize NGDP and ignore consumer price inflation.
“Again, Goodhart is assuming debt markets care about inflation. They don’t. They care about NGDP growth.”
I disagree. Economists around the world, many who have done research into debt markets, are teaching students how debt claims are priced, and they include price inflation premiums, not NGDP. Debt markets don’t care about NGDP growth. They care about inflation.
30. January 2013 at 15:13
OhMy:
Holy crap, you said banks DON’T lend out anyone’s money. Dr. Sumner correctly replied that yes, they do lend out people’s money by acting as financial intermediaries. And your reply is to say, in effect, “That’s nonsense, because banks can expand credit”?
Dude, has it ever occurred to you that banks can do both? That they can lend money deposited with them, AND expand loans not backed by deposits?
Identifying that banks can expand credit is not a refutation of the claim that banks are financial intermediaries who can also lend money deposited with them.
Keen is taking you for a long ride if you got that from him.
Warren Mosler? Isn’t he a communistic, modestly sized business owner who is masquerading as a monetary economist?
30. January 2013 at 15:16
“See also this critique of NGDP targeting from New Zealand economist Matt Nolan: http://www.tvhe.co.nz/2013/01/31/inflation-stickiness-demand-and-judging-the-success-of-monetary-policy/”
Just as a side note – that isn’t so much a critique of NGDP targeting, as a criticism of the misinterpretation of what flexible inflation targeting is. IMO, Goodhart is misinterpreting both in his comments – and I agree with Scott.
Both NGDP targeting and flexible inflation targeting are very very similar … and the only reason I’m still in the FIT camp is because of my prior assumptions that: 1) it is variability, and certainty, about price growth that matters rather than income levels 2) growth targeting is superior to level targeting. These are priors I’m willing to change if the weight of evidence was to turn.
30. January 2013 at 16:22
OhMy:
What happens to all the money I or anyone else deposits in my financial institution? It certainly does not simply sit in the actual bank. Methought that is why they have reserve requirements. And, why would they pay me for them providing the service of holding my money? Free security! This is the best deal ever!
I guess all those bank holidays implemented during the “great contraction” was just so bankers could have some extra days off.
30. January 2013 at 17:45
“J, You said;
“I assume that if supply factors (productivity growth) were able to create 10% RGDP growth, then you would not advocate a 5% NGDP growth path.”
You assume wrong. If you are referring to the US, I would favor 5% NGDP growth regardless of the RGDP trend line.”
Would you advocate 5% NGDP growth for China? Why would you favor 5% NGDP growth regardless of the RGDP trend line? Wouldn’t that prevent a booming economy from booming?
30. January 2013 at 17:53
Ohmy,
Great post. The common veiw of banking is a view of banks as they operated in the 1930’s.
30. January 2013 at 18:10
OhMy,
Thinking on this a little bit more. When the bank makes a loan, it creates an asset on its balance sheet, and a liability on the borrowers balance sheet, and credits cash to the borrower. And until that cash is spent, it is also still an asset for the bank…
The bank does not have unlimited room to create assets in this way. There is a limited amount of space on the banks’ balance sheet. Enter securitization — the bank creates a trust, sells shares in the trust. The trust issues bonds, and uses the proceeds of the bond sale to purchase the bank’s assets.
The function of a bank is not to be an allocator of money, but rather to be an allocator of risk.
30. January 2013 at 19:25
@OhMy It is *true* that banks can issue loans without having backing reserves… today. But banks need to eventually settle up reserves. Mosler thinks the CB will be forced to provide reserves “as needed,” but his view is incompatible with the CB’s ability to target nominal variables, such as inflation. That seems wrong to me. Isn’t there decent evidence that the CB *can* target these variables? I remember some Canadian inflation graph with inflation hovering between 1 and 3 percent: exactly the BoC’s inflation target.
30. January 2013 at 19:42
OhMy, The issue of whether banks need reserves to make loans obviously has no bearing on my claim. Banks do make loans, and they do act as financial intermediaries.
30. January 2013 at 19:43
J, No, why would it prevent an economy from booming?
30. January 2013 at 21:01
Oh, Charles Goodhart, blah, blah,blah and P.U.
BTW, I bet if an economy has nice period of sustained economy growth, it INCREASES the potential for rapid growth in the future.
Bouyed by sales, businesses invest in new plant and equipment. The latest R&D becomes effective. People move into jobs they like, start entering training programs. Tax revenues mean governments can spend on infrastructure. A virtuous cycle—did not this happen in the 1990s? Very good increases in productivity?
Now, the defeatists and gloomsters are having a field day. Productivity is permanently crippled. We have to accept cruddy growth, because you know, something changed on a dime in 2008.
Meanwhile, on debt, Goodheart is even worse. We have to be careful how we might effect debt markets. Really?
Has he looked at Ireland? With a combo of austerity and a damaged GDP, debt has skyrocketed in Ireland. Japan?
The great way to reduce public debt to GDP is print (digitize) a lot of money,and run balanced budgets. That will calm nerves in debt markets.
Are the Goodharts and Taylors of the world hired by bondholders? People who want perma-zero bound as they own bonds?
I just don’t get it.
31. January 2013 at 01:25
“NGDP does not target real variables; it targets a 100% nominal variable.”
Unfortunately, the problem is that we have done a terrible job of communicating this point. I still think the root cause of all this is that people have been taught poor macroeconomic intuitions for generations. Most economists cannot conceive of NGDP as anything but a misbegotten hybrid of inflation and real output, they have no other way of conceptualizing it. (Yes, along with the callous conflation of levels and growth rates.)
They can’t think of monetary policy in terms of M * V making a full-fledged nominal variable of its own. They can only think in terms of a mysterious “aggregate demand” which is somewhat and yet not really related to regular demand, which is based on a complicated equilibrium between the bond market and the propensity to consume, and which produces inflation by regularly “overheating” the economy, or else through inertia. But in the long run only prices get affected because there are limits to productive capacity at a given time. Monetary policy has to twiddle this mysterious AD force in such a way as to stabilize this “inflation” thing, because that’s the only thing it can truly handle and any other ambitions are reckless irresponsibilities, don’t you remember the 70’s?
It took me ages to realize that most economists didn’t see the economy the way that I did, that they all still essentially thought like this: http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/04/living-in-a-demand-side-world.html
and that hence a lot of what I was being taught was confusing because it was nonsense.
The stuff that John Papola is complaining about (http://www.forbes.com/sites/beltway/2013/01/30/think-consumption-is-the-engine-of-our-economy-think-again/print/) is just a byproduct of this underlying deeper confusion. Irving Fisher wasn’t confused, but 1929 destroyed his credibility, and then Keynes took over, despite Fisher’s own compensated dollar plan being essentially what ultimately saved the economy. Even Scott’s approach of viewing NGDP as being the inverse market value of the medium of account is enough to dispel these illusions, though I think Nick Rowe and Bill Woolsey see an even clearer picture. I realize we can’t get everyone to read the posts Scott has linked to on the sidebar (not even if Scott raises the font size) but maybe at least we could provide a little more sophistication in our popular articles? Starting with abandoning the approach of “selling” NGDPLT by calling it a hybrid of real growth and inflation? Most people then get the same impression as Krugman – NGDPLT is just a clever trick to lower real interest rates and boost growth vis a vis the status quo.
31. January 2013 at 01:27
Also, I found this old blog post on Forbes.com from December ’08 advocating monetary easing through devaluation, not sure if anyone’s interested: http://www.forbes.com/2008/12/09/dollar-devaluation-gold-pf-ii-in_fb_1209soapbox_inl.html
31. January 2013 at 01:47
To be clear, I was saying that: monetary policy is about controlling the flow of spending. Spending is the exchange of money for output. So controlling spending is controlling the flow of money. The money behind spending, and which accounts for nominal variables, is created by the central bank. The total flow of that money is NGDP. When spending is stable and production shifts, you may get a change in the mix between RGDP and prices. But NGDP, the total level of spending, the total amount of money flow (the inverse value of money, according to Scott), stays stable. So we are not mixing real and nominal variables.
There are two sides to every exchange, real goods on one side and money on the other. If we want to talk about the “aggregate” of all output, as a kind of abstraction, then we must remember that the “spending” on that output is coming from money on the other side of the exchange, and the volume of its flow. (Which you can approximate by pretending that money’s value is determined by supply and demand on a single market, even though it actually flows through and between all markets. And calling its value the inverse of NGDP is mixing prices and quantities in a supply and demand model…).
So seeing the job of monetary policy as controlling M*V (as Lars is forever pointing out) is in fact getting to the core of what monetary policy does, and how nominal variables are determined – i.e. through supply and demand exchanges of goods for money on billions of markets, and that the value of that money is what central banks must stabilize. And then you can see that, although on the goods side of the exchange we measure “the real part” Y (the same stuff being made) times P (their prices jumping in exchange with money), and that is a hybrid – the goal of central banks is to stabilize the other side of the exchange, M*V, the “money part”. It is more important to keep the “money part” stable (M*V) than the “nominal part” P (the exchange rate between money and goods”, as since wages are the stickiest variable a fluctuation in P which merely offsets a change in production Y doesn’t matter, whereas P fluctuations caused by changes in MV really do matter. And central banks shouldn’t meddle with the rest.
31. January 2013 at 02:05
We don’t need to put banks into macroeconomic models, we just need to put money in.
31. January 2013 at 05:29
[…] Sumner addressed this specific point in his post on Charles Goodhart: […] Goodhart is assuming debt markets care about inflation. They don’t. They care about […]
31. January 2013 at 05:43
Saturos, Well put.
1. February 2013 at 00:36
Scott, I am curious as to why you think that debt markets do not care about inflation. All investors care about inflation because it eats in to their returns. Debt investors tend to ignore inflation when it is benign, but this can’t be assumed to be a given surely.
1. February 2013 at 01:24
“It’s a shame that people like Michael Woodford have not kept up with research in monetary economics over the “past few decades.””
Whoa. This is pathetic. Woodford says something you disagree with, and suddenly he has no idea what he is talking about?
Let me quote yourself back to you when Woodford said something you like:
http://www.themoneyillusion.com/?p=8202
“I’ve pointed out that cutting edge research in macro (i.e. Woodford) suggests that the most powerful determinant of current movements in AD is future expected movements in AD. ”
So then he still did “cutting-edge research”? And now he is decades out of touch?
For some reason, you also seemed to value his opinions:
“Seriously, five years ago I would have been thrilled to hear a Washington Post reporter ask Woodford if he was influenced by me. And I still am.”
http://www.themoneyillusion.com/?p=16264
For some reason, you were downright giddy, when Woodford seemed to endorse NGDP targeting:
http://www.themoneyillusion.com/?p=15973
May I ask why, if the guy has “not kept up with research in monetary economics over the “past few decades.””?
By all means, criticize his arguments, but don’t imply he has no idea what he is talking about. You have no leg to stand on.
Pathetic.
1. February 2013 at 01:43
“It’s a big mistake for central banks to get in the business of influencing the amount of credit in the economy.”
I cannot see that central banks are NOT in that business. No matter how else you are looking at monetary policy (and I know that your view is quite different from mine), it *does* set the basis for the interest rate structure throughout the banking system that in turn sets the supply curve for loans.
I agree with OhMy that the influence that the Fed exerts on credit markets is only weak, but this also means that the effect of Fed policy on economic activity is much weaker and not as direct as you assume.
Money is of course a kind of credit, so changing the interest rates for credit has a significant influence on the supply of commercial bank money.
1. February 2013 at 05:53
“And the inflation uncertainty would not raise interest rates, as the debt markets care much more about NGDP than inflation. In the long run nominal interest rates are more closely correlated with the long run trend rate of NGDP growth than inflation.”
Do you have anything formal on this? (I think it makes sense, of course.)
I think nominal GDP targeting would tend to raise interest rates. I do tend to be simplistic, I know. But I think that the higher inflation uncertainty would cause lenders to supply less, the greater nominal GDP certainty would cause borrowers to borrow more. That looks like a reason to expect higher interest rates.
Inflation targeting provides greater certainty for lenders (at the expense of everyone else, including borrowers.) Nominal GDP targeting provides less uncertainty for borrowers (at the “expense” of lenders no longer being shielded from uncertainty.) Why not higher real and nominal interest rates?
I think a fancy argument would show that the demand would rise more than the supply would increase, and more credit would be provided at the higher interest rates. Eagle might have already shown this. I guess I should check.
What is the counter argument? I don’t see it. Why would lenders want to lend more because prices are high when their other incomes are low?
Is it avoiding default risk?
1. February 2013 at 06:04
I find it very difficult that a bank would turn down a profit opportunity if it doesn’t have the reserves to back up the loan they are issuing. Banks issue when there is a profit opportunity for them; this has nothing to do with how much savings there are. Otherwise, how would we be able to swipe a credit card and essentially take out a loan any time we want. Keynes even talks about this issue in his paper The Ex-Ante Theory of the Rate of Interest.
Now, banks do take in savings and they do issue credit; however, I find it very difficult to believe that the amount of savings they have affects the amount of credit they issue. I don’t know why a bank would turn down a profit opportunity just because they don’t have the savings required–to me, that seems absurd. I also think it’s a myth to believe pumping in bank reserves would increase lending because it doesn’t increase the amount of profit opportunities for lending that exist. The previous point seems to be vindicated in the past 4-5 years.
Here is the paper by Keynes that I’m referring to:
http://esepuba.files.wordpress.com/2011/10/keynes-the-ex-ante-theory-of-the-rate-of-interest.pdf
1. February 2013 at 06:15
I’d also like to point out that changing the interest rate has a huge impact on the amount of credit issued. Lower interest rates make credit cheaper for banks and for everyone else as well. Lower interest rates reduce the costs of servicing the debt and make debt far more enticing. This can have the propensity to cause unsustainable investment booms–like the housing boom. Just think about it from a balance sheet perspective.
What would a 3% mortgage rate versus a 7% mortgage rate make on your monthly mortgage payments? It makes a massive impact, especially considering compound interest. It can entice people to take on more debt, especially the lowering of interest rates causing asset prices to rise via the present value effect. If people see rising asset prices with falling interest rates and you get people borrowing money to buy asset prices because they are enticed by capital gains, you could create a positive feedback loop between rising asset prices and increased debt.
Keynes actually talks about such investment booms and talks about how they are unsustainable. He says this is the case because an investment boom creates more economic activity. As economic activity increases, the rate of interest usually tends to increase. This is what usually causes the boom to go to a bust in his view. Now, a central bank could come in and lower interest rates to continue the boom for a longer period of time, but in the end, the boom is simply not sustainable because it’s based on unrealistic expectations for asset prices(ex. a 9-10% continued rise in house prices is simply not sustainable when NGDP is growing at 4-5%).
Keynes talks about this effect in his paper written in 1937 called The General Theory of Employment. Here is the link to the paper:
http://membres.multimania.fr/yannickperez/site/Keynes%201937.PDF
1. February 2013 at 06:16
I’d also like to make one more point. It’s funny how the “Keynesian” Paul Krugman didn’t even understand the role that finance and a banking system played in the economy yet continues to call himself “Keynesian”. I’ve said this before and I’ll say it again, these “New Keynesians” need to call themselves Classical Hicksians. Especially considering that Keynes was “radically opposed” to the view and model put forth by Hicks(IS/LM), yet those “Keynesians” keep pointing to IS/LM as Keynes’ model.
1. February 2013 at 08:21
Suvy,
Right now local community banks seem adverse to risk in any form. It used to be greater risk carried a higher interest rate, but now it seems they will only make loans to nearly risk free customers.
1. February 2013 at 08:50
Chuck E,
The reason you are seeing that is for several reasons. For one, it’s because the financial crisis shifted the tolerance for risk of both the borrowers and lenders. In other words, their expectations have shifted. No one is expecting a 10% increase in house prices or any sort of an investment boom. The second is because the tolerance of debt of both the borrowers and lenders has shifted. Borrowers are more careful about taking on debt because the last time they took on massive amounts of debt caused their balance sheets to be underwater. The lenders also got in trouble because they were too easy with the credit they issued. The third part is that there is a shortfall in aggregate demand. When there is not enough demand, investment is less than it otherwise would be because investment is primarily driven by demand.
Everything that is happening right now is exactly what Minsky talks about. Whether it is the boom or the bust. The boom wasted resources and capital on useless items. We’re suffering the consequences in the bust. We had a debt fueled investment boom over the past 30 years that was based on unsustainable expectations driven by unsustainable credit expansion.
1. February 2013 at 12:13
[…] …and here is Scott Sumner's reply themoneyillusion.com/?p=19119 …and I 100% agree with Scott on this one. […]
1. February 2013 at 12:34
“NGDP does not target real variables; it targets a 100% nominal variable.”
I am reminded of “How the World Sees” Market Monetarists according to Noah Smith… http://noahpinionblog.blogspot.com/2012/09/econotrolls-illustrated-bestiary.html
1. February 2013 at 14:31
The view from a major fund management house in London.
http://www.bondvigilantes.com/blog/2013/01/23/nominal-gdp-targeting-for-the-uk-coming-sometime-maybe/
At least it is being debated, even if the author like most “big guns” still see it NGDP as merely RGDP+inflation. And at least it publicizes the charts.
I think your discussions recently are very helpful explaining why it is not just RGDP+inflation.
1. February 2013 at 15:50
ssumner,
I’m with shining raven.
“It’s a big mistake for central banks to get in the business of influencing the amount of credit in the economy.”
I am the kind of econo-nerd who stays home Saturday nights to read David Marsh and Allan Meltzer and for 20 yrs I have been laboring under the apparently wrong impression that the one and only thing that CB’s try to do is influence the amount of credit in the economy.
Please correct my reasoning.
1. The fed fears inflationary growth
2. The FOMC votes to raise rates
3. The FOMC desk at the NY fed does repos with the primary dealers to lower reserve levels thus causing scarcity thus raising price of overnight reserves the CP market reacts instantly etc…
4. The curve is raised as long rates reflect market expectation of future short rates
5. The rising curve causes my senior loan officer to charge
higher rates on the loan for my expansion plan
6. I cannot afford these higher rates and choose not to borrow
No credit has been created, thus the Fed has influenced the amount of credit in the economy.
Other subject:
With all of this mishegas about bank loans above OhMy et al.
I am reminded of Jamie Dimon’s recent congressional testimony regarding the London whale losses. He very clearly said that all the big banks suffer not from the issue of finding money or reserves to fund loans but figuring out what to do with all the deposits flooding in because there are nowhere near enough good lending opportunities out there…anyway they wind up prop trading the excess instead of lending.
2. February 2013 at 08:42
Tom, They care about NGDP, not inflation. If inflation is high then NGDP growth will generally be high, and lenders will demand a higher interest rate to offset the fact that future dollars will purchase a smaller share of NGDP. But it’s higher NGDP growth that causes highe rinflaiton. If NGDP growth is constant but inflation rises from 2% to 5%, nominal rates will barely budge.
Shining Raven, That was sarcasm, obviously.
On your send point. Yes, monetary policy can impact lots of variables in the short run, including employment, steel output, toaster output, and credit markets. My point is that the Fed shouldn’t target any of those 4 variables, they should target NGDP.
You said;
“I agree with OhMy that the influence that the Fed exerts on credit markets is only weak, but this also means that the effect of Fed policy on economic activity is much weaker and not as direct as you assume.”
You are mixing up two completely unrelated issues; can the Fed influence NGDP, and will more NGDP lead to more RGDP?
Bill, I’d look at George Selgin’s stuff on how NGDP targeting actually reduces risk for lenders and borrowers.
Reductio ad absurdum: Imagine 0% inflation and 10% RGDP growth. You are thinking about lending money for 30 years. What sort of interest rate will make it worth while for you to do so? Obviously you’d need to be offered a very high rate, as the country will be 20 times richer in 30 years. Money will be worth a lot less in the sense that a $20 bill is worth far less to you or me that it is to a peasant in Bangladesh (even if the countries had the same cost of living.)
Suvy, You said;
I find it very difficult that a bank would turn down a profit opportunity if it doesn’t have the reserves to back up the loan they are issuing.”
Yes, but this has no bearing on my claim. You could say the same about retailers. They sell items even if there is no immediate restocking from factories. Yes, banks can sometimes lend out money owned by the owner of the bank, just as retailing can occasionally sell goods that they themselves manufacture.
You said;
“I’d also like to point out that changing the interest rate has a huge impact on the amount of credit issued.”
Never reason from a price change. Lower rates are usually associated with less lending.
James of London, The person who wrote that doesn’t seem to get the idea. There is no need to estimate trend RGDP growth.
Paul Einzig, Raising short rates often reduces long term rates.
Are you THE Paul Einzig?
2. February 2013 at 13:55
Shining Raven, That was sarcasm, obviously.
Sorry, that is not how I understood it. Then I take it back.
2. February 2013 at 14:45
Yes, monetary policy can impact lots of variables in the short run, including employment, steel output, toaster output, and credit markets.
One of these things is not like the others with regard to monetary policy. Money is credit, so you can hardly say that the impact of monetary policy on credit markets is incidental.
You keep saying that you are not interested in credit markets, or banking, or loan activity, and only in monetary policy, but in my view this makes no sense. The credit markets determine the amount of “broad money” that we actually use in buying and selling and settling transactions. How can you possibly think about monetary policy and completely neglect this? In my opinion, you can’t.
You are mixing up two completely unrelated issues; can the Fed influence NGDP, and will more NGDP lead to more RGDP?
No, I’m not. Perhaps did not phrase this precisely enough. I understand that NGDP can rise both through inflation or an increase in RGDP. I was thinking only about a possible increase in broad money that could lead to inflation – but even there my point is that the influence of the Fed on the broader “money” supply is weak.
And of course I think one should really care about RGDP, and unemployment, although of course it is hard for the CB to really do something about this.
2. February 2013 at 16:09
Hi Scott,
Honestly, like Shining Raven, I didn’t get the sarcasm…..Sorry, I’m new ’round here!
As I’m sure you know Einzig died in 1973 when I was 6. I am an FX spot trader with Transylvanian Jewish roots, so Einzig is a hero of mine, I collect his books.(of which there are many)
2. February 2013 at 20:31
Saturos –
“monetary policy is about controlling the flow of spending. Spending is the exchange of money for output. So controlling spending is controlling the flow of money. The money behind spending, and which accounts for nominal variables, is created by the central bank. The total flow of that money is NGDP.”
This is 100% correct and a very good explanation. But I prefer to think of the inverse. We want to define the dollar as a share of GDP. Instead of saying we want the flow of money(GDP) to be 16 Trillion, we should say we want the dollar to be valued at 1/16 Trilionth of GDP. Then if the dollar rises above that value, we devalue; if it falls below that value, we revalue.
I realize it’s the same thing, but framing is important. If it’s thought of as stabilizing the dollar rather than stabilizing spending, it sounds less like an intervention into the economy.
3. February 2013 at 02:06
“Never reason from a price change. Lower rates are usually associated with less lending.”
My point was simply this: the cost of a loan is the interest rate you pay on it. I do agree that there are feedbacks where more lending could result in a higher interest rate and less lending could result in a lower interest rate.
One of the essential points of Keynes’ theory about investment booms is that the increased borrowing during an investment boom and the increase in investment actually drives up the interest rate. This is why, in his theory, investment booms “carry within them the seeds of their own destruction”.
“Yes, but this has no bearing on my claim.”
I never said it did. In fact, I agree that debt markets have to do a lot more with the amount of activity in an economy than they do with inflation. In my eyes, the key factor in determining interest rates is aggregate demand, not inflation. The reason you see high interest rates with high inflation is primarily because of high levels of aggregate demand that leads to the high inflation. This is why I think NGDP growth are much better correlated to interest rates than inflation.
3. February 2013 at 10:42
Shining Raven, No, money is not credit.
Paul, I’m also a fan of Einzig.
Suvy, You said;
“My point was simply this: the cost of a loan is the interest rate you pay on it. I do agree that there are feedbacks where more lending could result in a higher interest rate and less lending could result in a lower interest rate.”
My point had nothing to do with “feedbacks.” Just basic S&D theory. Changing a price has no impact on equilibrium quantity–that’s S&D 101
3. February 2013 at 11:36
“My point had nothing to do with “feedbacks.” Just basic S&D theory. Changing a price has no impact on equilibrium quantity-that’s S&D 101”
I reject the supply and demand theory when it comes to credit issued. The supply of funds has very little to do with the amount of credit actually issued. The credit is issued when the bank can simply make money. If they see a situation where they can get 5% and borrow at 3%, they make the loan(accounting for default risk and other risks). The amount of bank reserves/savings has very little to no effect on the amount of debt issued. I’m looking at it from a perspective of individual balance sheets. If there is an opportunity to make money for a bank by issuing a loan, they will do so.
3. February 2013 at 11:44
The big difference between finance and every other thing that is bought and sold is that the amount of finance available cannot be modeled by a closed system. Banks create money when they issue credit; therefore, the traditional supply and demand analysis for other goods doesn’t work when it comes to the credit issued.
For example, in a market for bananas, the supply of bananas at any given point in time is fixed. In finance, the banking system can never run out of credit it can issue; it stops issuing credit whenever there are no profit opportunities that arise from issuing loans.
4. February 2013 at 00:28
[…] Sumner’s response to this point emphasises a fundamentally different perspective about what debt markets really care about. He stated that, “Goodhart is assuming debt markets care about inflation. They don’t. They care about NGDP growth.” […]
4. February 2013 at 01:04
Scott, Thanks for the response. You claim that higher NGDP growth causes higher inflation, hence bond investors therefore care more about the growth in the level of NGDP. This chart however shows an inverse correlation between NGDP and inflation during the 1990’s, although a strong positive correlation during the 2000’s.
http://www.creditcapitaladvisory.com/2013/02/04/why-bond-investors-care-inflation-ngdp/
So bond investors who focused only on NGDP changes in the 1990s would have missed the fall in inflation and the subsequent fall in interest rates (and the bond bull market). Investors who followed this rule in the 2000’s would have fared well as you imply. I think the issue here is that the assumption of the endogeneity of inflation is problematic. Ed Nelson (St Louis Fed) wrote an excellent short piece on this in 2008 highlighting the similarity of Friedman and Taylor on the endogeneity of inflation and comparing Bill Philips’ dataset which generated data anomalies due to exogenous shocks. In the 1990’s, the integration of China and India into the global supply chain was such an exogenous shock highlighting that NGDP growth does not always lead to rising inflation.
4. February 2013 at 01:47
Scott: Shining Raven, No, money is not credit.
I knew you would say this. But I do not understand it. Of course money is credit. What else would it be? It is a claim on future consumption!
How would you define money? Please don’t say “it’s the medium of account, the medium of exchange, and a store of value.” All true, but in my opinion not helpful.
Money is certainly not the same as currency (“pictures of dead presidents”). And it is not a commodity. So what is it in your opinion, if not credit?
I can easily imagine our economy running pretty much as it is without any currency at all. Most payments are already made electronically from deposit balances, with checking cards or credit cards or – in banking systems more advanced than the American one – by direct electronic deposit (actually, I have to admit that I don’t know how many checks Americans are still writing, and how much is done electronically via online banking – when I left the US 10 years ago, there was only phone banking at my credit union…).
In my opinion it is pretty clear that we are running something that is almost a pure credit economy, with currency as a quaint afterthought, used for convenience in small-scale payments. So in this case, “money” is a balance on a demand deposit, or a line of credit that I draw on. Every time I pay by credit card, money is created as credit.
I completely agree with Suvy here:
Banks create money when they issue credit; therefore, the traditional supply and demand analysis for other goods doesn’t work when it comes to the credit issued.
I reject the supply and demand theory when it comes to credit issued. The supply of funds has very little to do with the amount of credit actually issued. The credit is issued when the bank can simply make money.
I think this is correct, and I cannot see how money can be anything else *but* credit. I mean, it is always somebody’s liability, the bank’s or the central bank’s, and if you pay it back, it is gone, simply extinguished! How is this not credit?
4. February 2013 at 03:13
@OhMy/Geoff/Churchman
Let me sketch a super simple example, and tell me where I’ve gone wrong: Say we have persons x & y, and banks A & B and everybody’s balance sheets are empty… and there are no reserve requirements (we’re in Canada) or capital requirements (for simplicity).
x wants to buy y’s house for $100k, and get’s a loan from A to do so. A writes in their liabilities ledger “owe x $100k” and under assets “loan to x for $100k.” Then x writes y a check for $100k. In the process of clearing the check (a day or two later), A’s CB reserve account is overdrawn and B’s CB reserve account credited: both by $100k. A borrows B’s $100k of reserves to repay the CB by the end of the day. A makes money on the spread between interest it collects from x and the interest it pays to B. B makes money on the spread between what it collects from A and what it pays to y (y now has a deposit at B, i.e. B has on it’s liabilities ledger “owe $100k to y”). The CB had to TEMPORARILY create $100k to credit B, but was repaid by A. At the end of the day the CB still has an empty balance sheet, but all the other players don’t. How has A or B acted as an intermediary for anybody else’s deposit, since they all started off with empty balance sheets?
It’s easy to add in reserve requirements, say 10%. The only difference is that at the end of the day B must keep $10k in it’s CB reserve account because it holds a $100k IOU (deposit) to y on its books, which means the Fed had to create that $10k. But it’s A that originally obtained it from the CB as a loan (in an OMO) by pledging the “loan to x” asset on it’s BS as collateral. Why? because originally A had to meet the reserve requirement. Thus when A is overdrawn, it’s only overdrawn by $90k, the other $10k it had already borrowed from the CB. Thus it only needs to borrow back $90k from B to repay the overdraft, leaving B with its $10k reserve requirement CB deposit. Thus the CB’s balance sheet is no longer empty at the end of the day: it has a $10k loan on its books to A to support $100k of money spent and deposited in the real world.
4. February 2013 at 04:44
Tom Brown:
I’m not in the same boat as OhMy. I don’t agree with what he said about banks not being intermediaries.
I don’t really know what the motivation for your latest post is, so all I will say is that your initially assumed price of $100k itself depends in part on the extent and height of credit issuance, as credit issuance from banks add to the total money supply and spending, which is one half (nominal demand) of price formation (the other being supply). Not sure how that fits into what you’re saying though.
4. February 2013 at 06:39
Tom, You misunderstood me. I didn’t claim NGDP predicts inflation. I said bond investors don’t care about inflation. So if following NGDP causes you to miss changes in inflation, who cares?
Shining Raven, You said;
“I knew you would say this. But I do not understand it. Of course money is credit. What else would it be? It is a claim on future consumption!”
By this definition a $20 gold coin was “credit” back in 1929.
I can also envision an economy with very little base money. But the base would still be the MOA, and hence the all important factor determining NGDP. Even if it was .00001% of NGDP.
4. February 2013 at 07:05
@Geoff, My motivation is to lean people’s views. I’m confused as to which two parties the bank is an intermediary between in my example. I thought perhaps you meant by that it was an intermediary between private non-bank savers and borrowers, and I don’t see that in my example. Perhaps you meant something else?
4. February 2013 at 07:20
“learn” not “lean”
4. February 2013 at 08:18
Scott: By this definition a $20 gold coin was “credit” back in 1929.
Are you kidding now? Do you want to argue about a gold standard system, or are we talking about the monetary system as it exists today?
Don’t you think there is a difference? What is your point in bringing this up?
Anyway, I’ll bite: Yes, a gold coin issued by a bank is “bank credit” in a way that unstamped bullion is not. One has to be accepted by the bank in payments of debt to the bank, the other only if the bank feels like it or trades gold.
(And incidentally, did we not have a fractional reserve system in 1929 as well? You are not saying there was no paper money that was not fully backed by gold?)
We *do have* an economy with very little base money (and what do you mean by this? Reserves + currency?). Nonetheless, I cannot understand that apparently you think this is *all* we have to care about, and money creation by banks can somehow be disregarded.
4. February 2013 at 18:46
Tom Brown:
“I’m confused as to which two parties the bank is an intermediary between in my example. I thought perhaps you meant by that it was an intermediary between private non-bank savers and borrowers, and I don’t see that in my example. Perhaps you meant something else?”
I can’t really follow your example.
To me an intermediary is just a party that is lent money for the purposes of investment, and the intermediary uses his or her expertise and knowledge in judging and researching worthwhile investment opportunities that the lender hired the intermediary to find.
How that fits into your example I am not sure.
All I know is that I agree with Dr. Sumner that banks really do engage in the above, and that makes them intermediaries (in addition to whatever other activity they engage in, such as new credit issuer apart from being an intermediary).
I think you’re making things a little too complicated.
4. February 2013 at 20:52
Geoff: “too complicated” … perhaps! The point of my example is that a loan is made, and a purchase made with the funds, but the process didn’t involve the bank acting as an intermediary between a saver and a borrower. I’m guessing you might agree because I don’t think I’m using “intermediary” in the same way you are. In my example the bank is simply a “new credit issuer” having presumably done their homework on the prospective borrower and determined that he was a good credit risk. They weren’t doing that on behalf of anyone else who “loaned them money” but they do do their homework well because they want the loan they made to be seen as collateral worthy by others (other banks or the Fed).
5. February 2013 at 11:46
Tom Brown:
OK, well, if in your example you aren’t presuming an intermediary, then I guess you’re not presuming an intermediary.
Would you agree that banks not only issue new credit, but accept deposits from clients, for the purposes of investment, whereby the client hopes to earn interest/dividends/capital gains that they could not get on their own because they lack the requisite expertise and knowledge? If so, then we’re all in agreement.
5. February 2013 at 17:23
Geoff: yes, I agree that they accept deposits from clients. I’m thinking of clients like me when I speak of clients: They can put my deposit in their Fed reserve account, collect 0.25% a year on it (for clients like me, that should really be all the investment expertise they need), send me a tiny fraction of that ($3 they sent last year), and provide me with access to my funds on demand. And they make loans.
I realize banks are into a WHOLE lot more than that, but that simple model is what I’m referring to when I talk about banks. That model doesn’t exclude the possibility that they try to maximize the spread between their rate of return on their assets vs what they pay out for their liabilities… while meeting regulatory requirements (reserve/CAMELS etc)… but I doubt customers like me benefit monetarily from much of that. I’m a SUPER cheap liability for them, and that’s about is for “client” status. I benefit from the convenience they provide and that’s about it.
I’m not talking about brokerage services, or investment services, etc. They pay clients like me as little as they can get away with… and I’m supposing their shareholders benefit from all the fancy spread optimization they do.
What I was really objecting too with the use of “intermediary” was I thought perhaps you were using it like I’ve seen Krugman use it: As if the bank were nothing more than a store of gold coins, and it had to wait to receive some in it’s coffers from depositors before it could loan them out to borrowers. Clearly you’re not using “intermediary” in that fashion!
6. February 2013 at 09:40
Shining Raven,
I think you might be confusing fractional-reserve “money creation” with base money creation. I know this is a common confusion.
http://en.wikipedia.org/wiki/Fractional_reserve_banking#Money_creation_process
7. February 2013 at 19:54
Tom Brown:
“They can put my deposit in their Fed reserve account, collect 0.25% a year on it (for clients like me, that should really be all the investment expertise they need), send me a tiny fraction of that ($3 they sent last year), and provide me with access to my funds on demand. And they make loans.”
You don’t consider them loaning those deposits to borrowers in addition to the Fed?
“What I was really objecting too with the use of “intermediary” was I thought perhaps you were using it like I’ve seen Krugman use it: As if the bank were nothing more than a store of gold coins, and it had to wait to receive some in it’s coffers from depositors before it could loan them out to borrowers. Clearly you’re not using “intermediary” in that fashion!”
I think we understand each other.