Why current AD depends on expected future AD: Scott Sumner in ISLM

Scott Sumner argues that nominal interest rates are near zero because monetary policy (specifically expected future monetary policy) is too tight. He argues that tight (expected future) monetary policy makes expected inflation low, which makes nominal rates low. He also argues that tight (expected future) monetary policy makes real rates low as well.  I want to re-state Scott Sumner's argument in terms of a standard ISLM model. I know Scott doesn't like the ISLM model, but I do. And more economists understand the standard ISLM model than perhaps understand Scott's verbal reasoning.

Scott talks about monetary policy affecting nominal GDP. I am going to talk about monetary policy affecting the position of the Aggregate Demand curve. My way of talking is both more conventional and more accurate. (Because the effect of a shift in AD on NGDP depends on the slope of the Aggregate Supply curve, unless the AD curve is unit elastic, which it probably isn't, and because price times quantity is conceptually distinct from price times quantity demanded).

If a loosening of expected future monetary policy caused the expected future Aggregate Demand curve to shift to the right, this would have two effects on the position of the current AD curve.

First, if the expected future AD curve shifted right, this would cause the expected future price level to increase, which would increase the current expected rate of inflation, which would increase the wedge between current nominal and real interest rates, which would increase current AD. There are three ways of showing this in an ISLM diagram, all equivalent. If you put the nominal interest rate on the vertical axis, a 1% (i.e. 1 percentage point) increase in expected inflation shifts the IS curve vertically up by 1%. If you put the real interest rate on the vertical axis, a 1% increase in expected inflation shifts the LM curve vertically down by 1%. If you put both nominal and real interest rates on the vertical axis, then expected inflation creates a vertical wedge between the IS and LM curves, on the right of their intersection, and a 1% increase in expected inflation increases the size of the wedge by 1%. This point is usually well-understood by any student of second-year macro, so I won't dwell on it more.

Second, if the expected future AD curve shifts right, this would cause the expected future level of real output to increase (unless the future AS curve were vertical). There are (at least) two reasons why an increase in expected future real output (real income) would shift the current IS curve right. First, an increase in expected future real output and income would increase current demand for consumption (reduce current saving), which would shift the current IS curve right. Second, an increase in expected future real output (and real output demanded) would increase the expected profitability of current investment, which would shift the current IS curve right. (A third effect would be if an increase in expected future real output and income caused current asset prices to increase, which might increase current investment and consumption.)

One way to define the current "natural rate of interest" in the ISLM model is the real rate of interest at which the IS curve intersects the natural rate of output (or "full employment", if you prefer). An increase in expected inflation reduces the real rate of interest relative to the natural rate of interest, even if the nominal rate of interest is stuck at zero. Nearly everybody understands that first point. But an increase in expected future real output, by shifting the current IS curve right, will raise the current natural rate of interest relative to any given actual real rate of interest. Fewer people understand that second point.

And it's to help people understand that second point that I have written this post.

If expected future monetary policy is too tight, and so expected future AD is too low, that lowers the nominal rate of interest relative to the real rate of interest, but it lowers the natural real rate of interest as well. That is two reasons why an expected future monetary policy that is too tight can force nominal rates down to zero, yet still leave the rate of interest above the natural rate.

The standard ISLM model can be used to exposit the conventional view that monetary policy is powerless to increase AD when nominal interest rates are zero. But the standard ISLM model can also be used to exposit Scott Sumner's "unconventional" view that the power of a misguided (expected future) monetary policy is precisely what makes (current) monetary policy appear to be powerless.

Comments

[JKH]

Nick,

This may be a dumb question, and maybe I’m completely missing the point here, but how do you identify expected future monetary policy?

I assume that if it’s identifiable, it must be observable in financial market pricing.

But in observing what you observe, how do you distinguish between expected future monetary policy versus the expected future impact of current monetary policy?

Does my question even make any sense?

[JKH]

I guess its the implied forwards for the future fed funds or Bank of Canada rate. Or would you see it as something different than that?

[Nick Rowe]

JKH: I'm not sure how I would identify it. It depends on how you think of monetary policy. The most conventional ISLM model thinks of monetary policy as the stock of money supplied. Scott thinks of monetary policy as targeting the time path of expected Nominal GDP. So Scott would look at indicators of expected future NGDP.

If you think of monetary policy as the overnight rate target, you hit a conundrum. Because high current nominal interest rates can also be the result of loose past and expected future monetary policy (a past and expected future practice of setting interest rates lower than they need to be to keep inflation from rising). That's one of Scott's main points, and he is absolutely right on that one. Suppose you start with 2% inflation, and 5% interest rates, and inflation is steady. Suppose the Bank of Canada cuts interest rates to 4% and keeps them there. Inflation eventually rises to (say) 12% before the BoC decides to act. It's now not enough to raise the interest rate to 5%; it needs to raise it to over 15% to stop inflation rising further.

Nominal interest rates are a very misleading indicator of the stance of monetary policy. That in part explains why I so strongly resist people who insist that what the BoC does with monetary policy is set an interest rate. Even though I know full-well that in some "realistic" descriptive sense that is exactly what the BoC does, in another, and equally important sense, that is a very misleading way to think about what the Bank of Canada does.

This is not easy to explain.

[Nick Rowe]

Put it another way. You can define a tight or loose monetary policy as setting an interest rate above or below some equilibrium rate. But that equilibrium rate in turn depends on whether past and future monetary policy is tight or loose. The looser is past and future monetary policy, the higher is the current equilibrium rate that defines tight or loose monetary policy. That means you cannot define the "stance" (tightness or looseness) of monetary policy as a high or low interest rate. You can perhaps define it as an interest rate reaction function. Monetary policy cannot be captured in a number. It is instead defined as a way of rule that determines how you set one number in response to the information you have.

[Nick Rowe]

Suppose you are riding a bike in a straight line. Then you turn the steering to the left and hold it there. This does not in fact cause the bike to turn left; it causes the bike to fall over to the right. If you want to turn left, this is what you do: you turn the steering to the right, so you start to lean left, and now you turn the steering left.

You cannot describe which way the rider is turning by describing the current position of the steering. It depends on the past position of the steering as well. The analogy to thinking of monetary policy as setting an interest rate is exact; except it depends not just on the past but on the expected future position of the steering as well. It's not stable. It can only be dynamically stable as a reaction function with feedback, and then only maybe.

On the other hand, you CAN define monetary policy in terms of the time path of the money supply, without a feedback loop. It's stable.

[Too Much Fed]

What if lower interest rates don't get excess savers to spend and excess debtors can't go further into currency denominated debt (they already have too much)?

Would that cause AD in the present to be below AS in the present?

What if AD and AS have both been influenced by currency denominated debt?

[Too Much Fed]

Can near zero nominal interest rates mean the "fungible" money supply is skewed towards too much currency denominated debt?

[Gary Marshall]

Hello Mr. Rowe,

The reason that interest rates are so low is because there is little demand for funds in the economy at this time. Firms and people have borrowed enough and they have wisely decided to sit back and absorb their debts.

The only way to raise rates is by increasing the demand for funds, perhaps through a government fiscal operation. McGuinty seems on board.

Regards,
Gary Marshall

[reason]

No I think the problem is to be found in foreign exchange markets. There is a lack of good investment in the USA at current and expected exchange rates.

[reason]

And further I think is more due to a disfunctional international financial system, rather than monetary policy errors.

[reason]

Sounds like I'm an ally of Soros.

[Adam P]

Nick, if we're being really precise then actually the correct thing to say is not that current monetary policy is too tight but that expected future policy is too tight.

This nicely ties in with discussion before about what interest rate the CB can and can't control. The nominal short term inter-bank rate is low, that means money is currently loose pretty much by definition. Banks can get as much money as they want for virtually zero cost.

However, longer nominal rates are low which does imply low expected inflation. Thus, the future expected path of monetary policy is not loose enough. This is not Bernanke's fault really, he's been trying to promise loose policy will last a while but then guys like Plosser come along and bang the drum for the bond vigilantes. (Someone should really tell Plosser to sit down and shut up).

Also, I wish you'd stop writing as if this argument is original to Sumner. The idea goes back, at least, to Keynes (he just had a somewhat unhelpful view of expectation formation).

[Adam P]

Actually, I see you've already incorporated my comment. I started typing after your first sentance... sorry for the lack of faith!!!

[Adam P]

Nick, "A third effect would be if an increase in expected future real output and income caused current asset prices to increase, which might increase current investment and consumption."

this is not a third effect, it's the price mechanism for the other two.

[JKH]

I supposed TIPs implied inflation premiums embedded in bonds yields would be an indicator of expected monetary policy. Inflation premiums above the explicit/implicit inflation target implies expected monetary policy too loose.

It’s conceivable in one scenario that such a premium curve could be corrected by the central bank tightening tomorrow, returning all inflation premiums to target. Monetary policy in that case changes expectations overnight, not over a period of time. The original expectation was wrong. But because the correction only took one day, you could just as easily rationalize ex post that current monetary policy had been too tight as you could prove that expected monetary policy had been too tight – or both - a bit of a logical conundrum in expectations? Or not?

[JKH]

Sorry, I meant current and expected monetary policy too loose in my 2nd paragraph example above.

[bill woolsey]

Great presentation, Nick.

I favor a stable growth path for nominal expenditure. Isn't that conceptually price times quantity demanded? Scott does favor a stable growth path for nominal output. As a practical matter, I favor total final sales, he favors (nominal) GDP. The difference is inventory investment. Conceptually, I am trying to get at price times quantity demanded, but that is really the graph, and not the goal.

[Kevin Donoghue]

Nick,

This is a helpful post and if that is what Scott Sumner is saying then he is right I think. The important question is, how should expectations of future monetary policy be steered? The current central bank governor cannot bind future governors to a particular course of action. Presumably it's up to the legislature to define the powers and duties of the central bank in such a way as to ensure that only a mild inflationist would want the job of governor. Krugman (1998) didn't address that question. To me that paper seems like a purely theoretical exercise. He is just showing that he can play a very Keynesian tune on an instrument designed for freshwater melodies. Full marks for virtuosity but, if he actually meant it as a policy proposal, he should have gone into more detail about how the BOJ needs to change.

I'm a bit shaken by the news that there are people who put both the nominal and real interest rates on the vertical axis of a single IS-LM diagram. Should we not put them in the stocks?

[JKH]

"Nominal interest rates are a very misleading indicator of the stance of monetary policy. That in part explains why I so strongly resist people who insist that what the BoC does with monetary policy is set an interest rate. Even though I know full-well that in some "realistic" descriptive sense that is exactly what the BoC does, in another, and equally important sense, that is a very misleading way to think about what the Bank of Canada does."

Does this sort of debate come up here?:

http://www.cdhowe.org/display.cfm?page=monetarySynopsis

[scott sumner]

JKH's first few questions raise a very good point. There is more that one path of monetary policy consistent with any given future price level target. Thus one policy might be to set the 5 year forward money supply at X, and then have a 2% expected inflation rate 5 years from now. Another policy would be to set the 5 year forward money suplly at (0.9)*X, and then adopt a 10% expected inflation target 5 years from now. The higher inflation target will boost velocity (or lower Md) to insure that the same price level target can be hit with a smaller money supply. In principle, there are an infinite number of monetary policy paths for any given future price level targets. This is known as the indeterminacy problem, although to aviod confusion with another indeterminacy problem, McCallum argues it should be called the "path multiplicity problem." He argues that this isn't much of a practical problem for policymakers. I'm not sure why, perhaps because if you tie down the expected inflation rate then there is only one money supply path consistent with your inflation target path. I confess I am not an expert on this issue.

Adam's right that these aren't my ideas. I don't think Keynes spoke much about interest rates moving perversely in response to monetary policy. I recall reading where he denied this could occur, but then if you read enough Keynes you can find almost any view expressed somewhere. Keynes certainly had no awareness that an expansionary monetary policy could raise real rates, as he generally ignored real rates in his analysis. I got the idea from Robert King (the JEP from 1993) where he describes it using a forward-looking IS-LM model, so Nick is definitely right that these ideas can be shown using IS-LM.

I have a new post that discusses my views in more detail.

[Patrick]

Adam P: "Banks can get as much money as they want for virtually zero cost. ... However, longer nominal rates are low which does imply low expected inflation. Thus, the future expected path of monetary policy is not loose enough"

I think this worries me. All that free money can't find a productive home - it's déjà vu all over again. What if the problem is not so much the liquidity trap but rather a no-grow trap (insert Nouriel Roubini list here). Maybe the CB can't raise inflation expectation because nobody really believes there is going to be any real growth regardless of what monetary policy does. Without real growth, any increase in nominal GDP would just be pure inflation, and nobody believes that the CB of an advanced economy is going to allow high levels of pure inflation for very long. And even if the CB did allow high levels of inflation, money would just pour into inflation hedges like oil or gold and not investment to create jobs, which is what the US really needs to recover.

[Too Much Fed]

scott sumner said: "Thus one policy might be to set the 5 year forward money supply at X, and then have a 2% expected inflation rate 5 years from now. Another policy would be to set the 5 year forward money suplly at (0.9)*X, and then adopt a 10% expected inflation target 5 years from now."

What is your definition of money supply?

[winterspeak]

Scott:

1) Do you believe that banks lend our reserves, or that the quantity of bank reserves has any impact on how much banks lend?

2) Do you believe a Bernanke helicopter drop of dollars is monetary or fiscal policy? If you believe it is monetary, can you please explain how.

Thanks

[Nick Rowe]

Adam P. "Nick, 'A third effect would be if an increase in expected future real output and income caused current asset prices to increase, which might increase current investment and consumption.'
this is not a third effect, it's the price mechanism for the other two."

Yes. What I should have said is "Apart from the standard Euler equation effects on C and I, there may be additional effects from changes in asset prices and balance sheets relaxing borrowing and liquidity constraints on households and firms, as well as wealth effects in an OLG framework".

More accurate, and you would have understood it better, but perhaps not as accessible to most readers.

JKH: I was really puzzled by your comment...till I read your correction immediately below! Yes, various measures of expected inflation would be one important indicator. Expectations of future real GDP growth would be another. (Scott wants to add them both together into expected NGDP growth, but there's nothing in my ISLM analysis that says the weights on inflation and real GDP growth should be the same --though this is a picky detail.)

I think the expectational conundrum you point to is a direct consequence of central banks framing monetary policy in terms of interest rate targeting. I feel a post on that topic welling up inside me....

Kevin: thanks! Partly it's the time consistency problem (wanting to bind the actions of future governors). But mostly I think it's a "framing" problem. Interest rate targeting is a socially constructed reality. (The next post is really starting to well up inside me....)

But: "I'm a bit shaken by the news that there are people who put both the nominal and real interest rates on the vertical axis of a single IS-LM diagram. Should we not put them in the stocks?"

I'M ONE OF THOSE PEOPLE! There's nothing wrong with putting both nominal and real interest rates on the same axis, as long as you ignore the point where IS and LM intersect, and focus instead on the two points where the expected inflation wedge between them just fits. It's just like the tax wedge in micro, where we put both buyer's price and seller's price on the same axis, and put a tax wedge between S and D curves. Much easier than shifting the S or D curve.

JKH @8.59. Unfortunately it's very hard to bring that sort of debate to the CD Howe Monetary Policy Council. I tried to once, but the whole agenda is so ruled by the socially constructed reality of interest rate targeting, it would be like treating Holy Communion as a wine and crackers party in church. David Laidler (for one) totally understands the point though (I basically learned it from him originally, anyway.)

Scott: I'm glad to see you didn't totally reject my interpretation of you. And your answer to JKH is very much in line with my bicycle analogy.

Scott's strength in this topic is to take ideas from all over the place, put them together into a coherent whole, and apply them to the current recession. None of the bits are new, AFAIK, but the whole package is.

Patrick: trying to understand your argument: I think you have expected future real GDP affecting the SRAS curve?

[Nick Rowe]

Bill: Missed your comment, sorry> Thanks!

"I favor a stable growth path for nominal expenditure. Isn't that conceptually price times quantity demanded?"

I put on my schoolmarm voice, and reply: "you mean *desired* nominal expenditure?"

I see your point on inventories, and agree. Conceptually though, Qd=Q only if inventories are never out of stock, and there's never a line-up for services. (In Cuba, of course, the difference between desired expenditure and actual expenditure is massive, and has nothing to do with inventory accumulation/decumulation. But the empirical difference matters less in market economies where downward-sloping demand curves are the norm, and firms almost always want to sell more at the current price.)

[JKH]

“(NICK) On the other hand, you CAN define monetary policy in terms of the time path of the money supply, without a feedback loop. It's stable ...

(SCOTT) perhaps because if you tie down the expected inflation rate then there is only one money supply path consistent with your inflation target path...

(NICK) I think the expectational conundrum you point to is a direct consequence of central banks framing monetary policy in terms of interest rate targeting. I feel a post on that topic welling up inside me ....”

Seems like an oblique way of tackling the both the expectation question and the interest rate targeting issue; i.e. jointly - but could be interesting. Not immediately obvious to me why the money supply route should be less ambiguous than my interest rate example, but what do I know?

[JKH]

Nick,

Winterspeak has asked a couple of good questions of Scott above.

I have a question that seems to me to flow naturally in line. This would be for you or Scott. It’s has to do with something I raised on his blog some time ago, which I think is fundamental to the question of “money supply targeting”. This might also have something to do with an “instruments, indicators, etc.” type discussion that both of you may have had some time ago.

The point that I’ve raised previously but that I find monetarists tend to sweep under the rug (in my view) has to do with differentiating the components of the monetary base – or more precisely failing to differentiate them.

However one feels about how the Fed has handled the issue of excess reserve expansion (e.g. Scott hates payment of interest on reserves), there’s absolutely no way that the Fed could have elicited the same magnitude of response through currency that it has through bank reserves. That’s because, as I maintained in your last post, the Fed’s supply of currency is non-discretionary in response to the public’s discretionary demand. Conversely, the Fed’s supply of bank reserves is discretionary (particularly if they pay interest on them to maintain lower bound interest rate control), while the banks’ acceptance of those reserves in aggregate is non discretionary.

My question is:

Why is it that monetarists in discussions seem reflexively to ignore this fundamental distinction in the nature of the components of the (expanding) monetary base?

I.e. why do you always talk about “the base” as “the base”, when the decomposition of the base and its behaviour is potentially so revealing?

E.g.

Excess reserves have increased about 10,000 per cent.
Currency has increased about 10 per cent over the same period.
Doesn’t this difference matter to your analysis?

[Nick Rowe]

JKH @6.45: "Not immediately obvious to me why the money supply route should be less ambiguous than my interest rate example, but what do I know?"
For any equilibrium time path, there exists a second equilibrium time path in which all prices and money supplies are doubled, but all interest rates are the same as in the first. (And a third equilibrium time path in which all prices and money supplies are trebled, and so on.) The reason is that people care only about the ratios of variables measured in monetary units. That means that if you pin down the time path of the money supply (whichever money supply you choose) that pins down the time path of equilibrium prices. But if you pin down the time path of any interest rate, you do not pin down the time path of prices.

@7.20: Few macroeconomists have detailed knowledge about precise definitions of different monetary aggregates and the relations between them. We used to know more about this in the 1970's, when monetarists used to argue about precisely which monetary aggregate we wanted to grow at some fixed k% per annum. Monetarists won the war in the 1970's, but we lost that particular battle over the k% money supply growth rule. We were defeated badly, and sort of gave up. Time passed, banking technology and practices changed, and central bank operating procedures changed. We weren't convinced that interest rate targeting could work in theory, but it seemed to work in practice, and as long as it could deliver the monetarist goods of keeping inflation on target, we didn't see the need to revisit that battleground of our defeat.

Now we are slowly waking up to the stunned realisation that we might have been right after all when we said that interest rate targeting couldn't work. Right now, it seems to have failed as badly, or worse, than money supply targeting. But then money supply targeting wasn't that great either, we have to admit. So we are sort of lost. We have a mental framework in which some sort of monetary targeting could in principle deliver the goods (in a way that we know interest rate targeting cannot, for the reasons I gave above), but we have next to no idea how to implement it in practice. And the currency/reserves discontinuity is one symptom of that. Even if we found a way to implement money targeting well in practice, who knows if it would still work when banking technology changes again?

Monetary aggregates have recovered their old hold on our theoretical thinking, given the failure of the alternative, but we just don't want to make the investment into figuring the details out to make them work in practice. Because we think it's a lost cause.

In that context, what were once weird ideas (sorry Scott) like targeting NGDP expectations start to look good. And NGDP, like any measure of the money supply, has $ units, whereas interest rates don't have $ units. So we know NGDP targeting can work in principle, just as money supply targets can, and interest rate targets cannot.

[JKH]

Interesting, Nick. Thanks.

[Too Much Fed]

JKH said: "My question is:

Why is it that monetarists in discussions seem reflexively to ignore this fundamental distinction in the nature of the components of the (expanding) monetary base?

I.e. why do you always talk about “the base” as “the base”, when the decomposition of the base and its behaviour is potentially so revealing?

E.g.

Excess reserves have increased about 10,000 per cent.
Currency has increased about 10 per cent over the same period.
Doesn’t this difference matter to your analysis?"

There you go!!! That is my kind of question. How about these?

What exactly is a bank reserve?
Can I use a bank reserve to make an interest payment?
If there are so many excess reserves, I would like to go to the bank and withdraw some. Can I?

Lastly and similar to above, why do you always talk about “the [fungible] money supply” as “the [fungible] money supply”, when the decomposition of the [fungible] money supply and its behaviour is potentially so revealing?

E.g.

Has total currency denominated debt grown faster than total currency?

Doesn’t this difference matter to your analysis?"

[Too Much Fed]

Nick's post said: "Now we are slowly waking up to the stunned realisation that we might have been right after all when we said that interest rate targeting couldn't work. Right now, it seems to have failed as badly, or worse, than money supply targeting. But then money supply targeting wasn't that great either, we have to admit. So we are sort of lost."

IMO, most of the big financial crises in history have been caused by too much currency denominated debt. So, why not target currency denominated DEBT LEVELS???

[Too Much Fed]

Nick's post said: "We used to know more about this in the 1970's, when monetarists used to argue about precisely which monetary aggregate we wanted to grow at some fixed k% per annum. Monetarists won the war in the 1970's, but we lost that particular battle over the k% money supply growth rule. We were defeated badly, and sort of gave up."

Is it possible that banks create currency denominated debt first and look for reserves later?

[Too Much Fed]

Nick's post said: "We weren't convinced that interest rate targeting could work in theory, but it seemed to work in practice, and as long as it could deliver the monetarist goods of keeping [price] inflation on target, we didn't see the need to revisit that battleground of our defeat."

IMO, the only reason(s) it worked was because the fed got congress and the executive branch to implement their cheap labor policies of outsourcing, legal immigration, and illegal immigration along with positive productivity growth to produce price deflation. While claiming to be fighting price inflation, the fed gave people positive price inflation expectations along with bad assumptions about future wage income growth to SUCKER the lower and middle class further and further into currency denominated debt to fight the price deflation.

Yes, lower interest rates are needed to get the lower and middle class further and further into currency denominated debt.

No, lower interest rates are not about getting the excess savers to spend.

[Patrick]

Nick: I'm actually still trying to sort out my own thoughts. Are expectations about unemployment real or nominal? I'm thinking they're real. And if they're real and very negative, what does that mean (if anything)?

[Nick Rowe]

Patrick: Join the club ;). Unemployment is real, as opposed to nominal (it doesn't have $ in the units), so expectations of unemployment are real too.

If people expected higher future unemployment, that's roughly equivalent to expecting lower future real output. IS curve shifts left.

[winterspeak]

Nick: Maybe I should have asked you, and not Scott, because I'm really struggling with these two basic questions.

Seriously -- by what mechanism does the quantity of bank reserves impact anything else? Is it because you believe reserves enable more bank lending?

Also, if Ben Bernanke drops dollars out of a helicopter, is that monetary of fiscal policy? If it's monetary -- how? If Geither threw out the dollars instead, would that make it fiscal?

The only sense I can make out of any of the "expectations" stuff in this post, is that if people saved less, there would be more aggregate demand. To me, this leads to a discussion about why the are "spending less" and since I am not a monetarist I can easily think about reasons that have nothing to do with inflation expectations and everything to do with making your monthly nut expectations. Still, help me understand the core mechanisms of what you are talking about.

[Nick Rowe]

Hi Winterspeak: Yes, in normal times, when banks are reserve constrained, and/or earning less (or no) interest on reserves than on loans, and increase in reserves should increase bank lending. Not working well at present, presumably because banks are capital constrained? But I am definitely NOT the expert on this.

Helicopter money is equivalent to a money-financed transfer payment, which is equivalent to a money-financed tax cut. It's both monetary and fiscal. Monetarists normally emphasise the monetary aspect, even though they acknowledge it's both. That's because the fiscal change is temporary, but the extra money stays in circulation permanently.

If Bernanke does it, it's helicopter money. If Geithner does it, he has to get the money from the public by selling bonds, so the net effect is that the public has the same amount of money, but more bonds. So it's really helicopter bonds.

Helicopter money is slightly different from an open market operation. An OMO is helicopter money plus vacuum cleaner bonds (sucking bonds out of people's pockets). But if you believe Ricardian Equivalence, so bonds are not net wealth, and people discount the future tax liabilities implied by the bonds, vacuum cleaner bonds has no effect on consumption, so an OMO is equivalent to helicopter money. And helicopter bonds has no effect.

[JKH]

winterspeak,

see my comment two posts forward:

http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/11/interest-rate-targeting-as-a-social-construction.html

Nov 10, 8:25 a.m.

which includes:

"Winterspeak’s final question on the previous post is relevant here. Banks are not reserve constrained. They are capital constrained. This is ALWAYS the case, not just sometimes. But that doesn’t mean they aren’t reserve motivated. Chartalists get this slightly wrong. If banks are recipients of huge excess reserves provided by the central bank, they will attempt to find alternatives (to receiving interest reserves) that DON’T USE UP CAPITAL – i.e. risk free or near risk free assets such as treasury bills or extremely high quality short term credit (very difficult to find these days). Moreover, ARBITRAGE ensures rates adjust very quickly and any balance sheet expansion due to excess reserves is limited by the speed of the market arbitrage rate adjustment. E.g. if the Fed eliminated the 25 basis point reserve interest compensation, rates on treasury bills and near risk free assets would move so quickly that the banking system balance sheet effect and money supply effect would be quite contained. I’m afraid Scott Sumner gets this point wrong. Any tactical advantage of excess reserve deployment is only temporary due to risk free rate arbitrage; this plays a back seat role to the strategic constraint of capital requirements for any risky lending by banks. This is why the Treasury injected TARP capital into the banks, and why excess reserves are primarily a Fed liability tool for its own balance sheet expansion rather than an intended commercial bank lending motivator."

[Adam P]

Nick: "Yes. What I should have said is "Apart from the standard Euler equation effects on C and I, there may be additional effects from changes in asset prices and balance sheets relaxing borrowing and liquidity constraints on households and firms, as well as wealth effects in an OLG framework"."

Yep, agreed. Sorry to nit pick.

[Kevin Donoghue]

It's just like the tax wedge in micro, where we put both buyer's price and seller's price on the same axis, and put a tax wedge between S and D curves.

Fair enough, I haven’t seen it done so I shouldn’t assume it confuses things. And now I come to look I see you have explained this previously.

[Winterspeak]

Thanks for the details Nick. I now know, with confidence, that you are wrong. Your explanation was clear though

jkh: interesting nuance. I know you've been giving thought to bank behavior in the face massive excess reserves and I'm interested to read your conclusions.

[Nick Rowe]

Adam P: It was a good nitpick. It got my head clearer.

Kevin: I find it helps prevent confusion.

Winterspeak: Yep, I'm much happier to be clear and wrong, than too fuzzy even to be wrong ;)

Weird! 3 replies all about clarity!

[Too Much Fed]

Nick said: "If people expected higher future unemployment, that's roughly equivalent to expecting lower future real output. IS curve shifts left."

What if productivity increases so there is higher future unemployment (less hours) and real output stays the same or rises?

[Too Much Fed]

Nick's post said: "Hi Winterspeak: Yes, in normal times, when banks are reserve constrained, and/or earning less (or no) interest on reserves than on loans, and increase in reserves should increase bank lending. Not working well at present, presumably because banks are capital constrained? But I am definitely NOT the expert on this."

I think you need to look at the financial condition of the borrowers. In general, are there any goods in short supply besides cheap oil and maybe food (no need to borrow to expand capacity)?

How about the lower and middle class? I'd say there are at least three(3) classes. They would be those that don't like to borrow for hardly anything, those that are smart enough to think it is not the time to be borrowing, and those (maybe the majority) who don't qualify to be borrowing (too much debt already, suffering negative real earnings growth and/or even negative nominal earnings growth).

I find it almost impossible to convince most economists that negative real earnings growth is a problem.

[Too Much Fed]

Nick said: "If Bernanke does it, it's helicopter money. If Geithner does it, he has to get the money from the public by selling bonds, so the net effect is that the public has the same amount of money, but more bonds. So it's really helicopter bonds."

I believe you should dump the phrase "helicopter money" and be specific.

Here is the way I look at it. If the fed can't get more private debt created (preferably on the lower and middle class) to solve their problems, bernanke can drop currency or drop excess reserves.

If the fed can't get more private debt created (preferably on the lower and middle class) to solve their problems, geithner can drop currency or drop gov't debt.

It seems to me there is/are excess reserves to bail out the banks and currency to bail out the foreigners. Where is the bailout for the lower and middle class?

Here is one last thing. If the treasury and the fed are "joined at the hip", are gov't debt and excess reserves the same thing???

[winterspeak]

Too Much Fed: The assertion that Bernanke's "helicopter drop" amounts to monetary policy is ludicrous, the very last refuge of scoundrels and monetarists.

If the Govt writes you a cheque for building a stealth bomber, it's fiscal policy. If the Govt writes you a cheque for being old, it's fiscal policy. If the Govt drops a cheque on your head from a height it's... monetary policy?!? Complete nonsense!

You take that cheque and you deposit it in a bank, and *that* is what creates additional reserves. The Treasury has the option of swapping these reserves for Treasury bills, but the reserves have to be there FIRST. You cannot buy a t-bill until you have the money to do so. The government creates the initial money, and the t-bill purchase changes its term structure. This is really obvious, and it's fascinating how completely impossible it is for some people to see.

This should also make it obvious why simply giving reserves to banks has no impact on anything. But the banks were not bailed out by excess reserves, that was as impotent as anything out of the Obama administration's financial castratos. Banks were bailed out by the Treasury -- TARP's $800B was booked as deficit spending, it was (again) fiscal, not monetary. There's an interesting point here about capital forbearance, but let's move on.

Obama has made it clear what he thinks of the lower and middle classes (and even the non-banker upper classes). A payroll tax holiday would have gone a long way to recapitalizing the private sector and improving loan quality at banks. But instead we have Lloyd Blankfein feating on caviar and babies, served by Obama et al.

[Scott Sumner]

Too much Fed, I usually define money as the base. But in the example you cite M1 or M2 would also work, but the numbers would be different.

Winterspeak, In normal times reserves do influence bank lending, but now that banks can earn money on reserves, there isn't much relationship. Expectations of whether the reserve injection is temporary also play a role.

I regard a helicopter drop as monetary policy, because it doesn't affect the size of the national debt. But I understand that many people think it is also fiscal policy, and I don't object to that view. I don't favor helicopter drops, BTW.

Oops, I just noticed that Nick disagreed later on. The helicopter drop can be regarded as a tax on current cash balances. It's "dilution" to use stock market terminology. So it isn't a burden on future taxpayers, like deficit spending. But I agree that this is a judgment call, as with Ricardian equivalence even deficits fall on the current generation. You're probably better off relying on Nick's view.

Nick, You are right that none of the bits are new. I think combining the Woodford emphasis on changes in the future path of policy with the monetarist excess cash balance transmission mechanism may be new, but then again it probably was discussed by King, or McCallum or someone like that. But I don't recall seeing the combination. I hope this doesn't sound arrogant, but I think you could pretty much say the same thing about books like the General Theory. Is there any specific building block that was truly novel in the GT? People were already kicking around all those ideas in the early 1930s. Of course I am not claiming to have developed a sweeping new paradigm like Keynes developed. Although I talk big, I think we actually were pretty close during the Great Moderation, "target the forecast" and "level targeting" would finish the job, in my view.
I was the first person to promote the idea of a penalty rate on excess reserves. But there was one offhand mention in another blog before my blog got going. On the other hand the "oral tradition at Bentley" goes back to October 2008. :)

JKH, What you say about tying down the expected inflation (or NGDP) path sounds right. It's not my area of expertise, however.

JKH, I think you are right about the distinction, but only once you start paying interest. Before that point the level of ERs was generally near zero, and they couldn't target the two separately, except through reserve requirements. But most monetarists were only interested in M2, and so they only cared about the Reserve Req. to the extent it impacted M2. Holding reserve requirements fixed, most monetarists assumed a fairly predictable relationship between changes in the base and M2. Obviously that's not true with interest on ERs, and may not be true at the zero rate bound either. I'm not a monetarist so I don't worry about those issues. I just want the Fed to do enough OMOs to get NGDP expectations on target, and let the market sort out how the base will be allocated.

[Too Much Fed]

All quotes are from winterspeak's post at 10:09 a.m.

"If the Govt writes you a cheque for building a stealth bomber, it's fiscal policy. If the Govt writes you a cheque for being old, it's fiscal policy. If the Govt drops a cheque on your head from a height it's... monetary policy?!? Complete nonsense!"

OK I can see all three(3) being fiscal.

"You take that cheque and you deposit it in a bank, and *that* is what creates additional reserves."

Does it matter if I cash it in for currency and don't deposit the currency in another "bank"?

"The government creates the initial money,"

Is that creates initial debt?

"and the t-bill purchase changes its term structure."

By that, do you mean create time differences?

"Banks were bailed out by the Treasury -- TARP's $800B was booked as deficit spending, it was (again) fiscal, not monetary."

But was it enough (as in what if the fed did not create excess reserves)? I'm assuming that the fed used excess reserves to buy some of the "garbate debt". I could be wrong about that.

I keep coming back to the idea that excess reserves MIGHT be future gov't currency denominated debt in disguise.

"Obama has made it clear what he thinks of the lower and middle classes (and even the non-banker upper classes)."

Agreed. He listens way too much to geithner, summers, bernanke, and his GS chief of staff.

[Too Much Fed]

Scott Sumner said: "Too much Fed, I usually define money as the base. But in the example you cite M1 or M2 would also work, but the numbers would be different.

With legal tender laws, I consider money to be currency plus all the debt based on the currency (gov't and private) because that is the "fungible" money supply, the one that can be used to purchase either financial assets or goods/services.

[winterspeak]

Scott:

You are incorrect, bank lending is not reserve constrained. It is capital constrained. If you want a link explaining how banks handle reserves, I can get you one, although it is well documented elsewhere. I understand why you want to believe that bank lending is reserve constrained -- monetary policy loses a critical (if not its only) operational channel if it were true. Unfortunately, it is.

Seriously though, let me know if you're looking for a link.

The helicopter drop absolutely impacts the size of the national debt, the same way buying a stealth bomber, or running SS, or any form of deficit spending impacts the national debt. I don't know if you've ever worked for the Govt, but you probably know someone who has. This is how Government spending works:

1. You do some work for the Govt
2. The Govt credits your bank account

This credit it counted as Govt spending, G. It will add to the National debt, unless the Govt debits bank accounts through taxation by the same amount (T). The Govt may or may not choose to swap reserves for Treasuries, depending on their monetary policy at the time.

Too Much Fed: Be careful, if you continue down this path, you too may understand how the financial system works!

If you deposit your cheque and take out cash, then the cheque does not add to reserves. Cash currency exists independent of the banking system. The Government creates the initial money (makes the initial credit in the banking system) when it deficit spends. Look at how I explained Govt spending to Scott just a few lines above. You may have direct experience with this and can vouch for my description.

A t-bill purchase just swaps reserves for Treasury bills. The duration of a t-bill is called its "term structure", but I think it means the same thing as your "time structure".

Excess reserves do nothing. The fiscal TARP was a complicated way to implement regulatory forbearance on capital requirements for banks. The Fed's policy with toxic assets was a complicated way to effectively lend unsecured to member banks. Excess reserves are inconsequential.

[Nick Rowe]

Winterspeak: If you define the debt to include currency in public hands, then of course a helicopter operation increases the national debt. But if you exclude currency from the national debt, because unlike bonds currency pays no interest, then a helicopter operation does not increase the national debt.

[JKH]

Scott Sumner,

“I think you are right about the distinction, but only once you start paying interest.”

That’s right, and that was the context of my question, although it has to do more directly with the size of excess reserves, which is reflected indirectly in the Fed’s decision to pay interest on them. I think it’s worth pointing out in discussions of massive base growth during the credit crisis that an extremely disproportionate amount is coming from excess reserves. That the reserve growth is a disproportionate contributor (and that currency has been a relative “underperformer” in the sense of growth) should in itself be a warning signal that the interpretation of the excess reserve component may not be “textbook” boiler plate, and that the purpose of such reserves may not be what at first blush seems natural. (There are more fundamental reasons as well, but this circumstantial divergence is a flashing signal, IMO.) If behaviour in the two base components is so wildly divergent, it should call into question just why that is. This is the sense in which I think differentiation is critical. Not to get into it again, but my interpretation of the excess reserve component is described several posts on here; I’ve probably noted it on your blog several times in the past as well.

BTW, with a NORMAL Fed balance sheet, there are two types of targeting. The Fed ACTUALLY targets excess reserves on a daily basis in order to steer the overnight trading rate (effective Fed rate) toward the target policy rate. Conversely, I assume that monetarists, to the extent they believe the base should be targeted longer term (limited as per your M2 comment), would be inclined naturally to use the total without reference to composition, given that the reserve components are usually a relatively minor dollar contributor to base growth. In that sense, I agree that differentiation is a less important or more predictable explanatory factor in normal times.

“I just want the Fed to do enough OMOs to get NGDP expectations on target, and let the market sort out how the base will be allocated”

I understand that, particularly if it refers to your NGDP futures targeting idea. Also, the idea of letting the market sort out how the base will be allocated is consistent with an emphasis on distinction. Excess reserves would be created initially via OMO, and then converted to currency partially, according to market demand at the teller’s wicket or at the ATM.

Sorry, I’d either forgotten or didn’t appreciate you weren’t a monetarist. If I’d been reading your blog more regularly of late, I’m sure I would have been reminded of that. My crude idea of a monetarist is that of an association with the idea of the importance of the stock of money in some sense as opposed to other variables (such as the interest rate). No doubt that’s wrong or oversimplified.

[Too Much Fed]

Nick said: "Winterspeak: If you define the debt to include currency in public hands, then of course a helicopter operation increases the national debt. But if you exclude currency from the national debt, because unlike bonds currency pays no interest [MY EDIT: and has no repayment terms END MY EDIT], then a helicopter operation does not increase the national debt."

If the treasury/gov't won't let the fed fail and if excess reserves are actaully debt instruments (even if they did not pay interest until recently), does that mean that helicopter dropping excess reserves is increasing the national debt (excess reserves are gov't debt in disguise)???

[Too Much Fed]

winterspeak said: "You are incorrect, bank lending is not reserve constrained. It is capital constrained. If you want a link explaining how banks handle reserves, I can get you one, although it is well documented elsewhere."

I'll take some links, please.

And, "Too Much Fed: Be careful, if you continue down this path, you too may understand how the financial system works!"

That's why I'm posting, to learn.

I'm still thinking that excess reserves allowed the fed to overpay for "toxic" debt assets. That allowed the banks to "maintain" bank capital. The fed then hoped for their usual solution, create more currency denominated debt to solve a bad currency denominated debt problem.

[Too Much Fed]

JKH said: "I think it’s worth pointing out in discussions of massive base growth during the credit crisis that an extremely disproportionate amount is coming from excess reserves. That the reserve growth is a disproportionate contributor (and that currency has been a relative “underperformer” in the sense of growth) should in itself be a warning signal that the interpretation of the excess reserve component may not be “textbook” boiler plate, and that the purpose of such reserves may not be what at first blush seems natural. (There are more fundamental reasons as well, but this circumstantial divergence is a flashing signal, IMO.) If behaviour in the two base components is so wildly divergent, it should call into question just why that is."

Let's stay with this topic.

Let's also add why is the "fungible" money supply (the money supply that can be used to purchase either financial assets or goods/services) skewed towards so much currency denominated debt.

[Patrick]

Ok, I had to revisit Adam P's blog:

http://canucksanonymous.blogspot.com/2009/05/what-do-liquidity-traps-have-to-do-with.html
http://canucksanonymous.blogspot.com/2009/05/consumption-euler-equations-explained.html

to remember why I had a sick feeling about the effects of high and rising unemployment on real expectations:

"A negative real interest rate comes about via the consumption Euler equation from some combination of expected consumption growth being low (possibly negative) and (for risk averse agents) consumption growth being perceived as very uncertain (volatile). In particular, a negative natural real rate of interest can occur even in a world without money."

So, I think I'm thinking starting at low interest rates, unemployment feeds into creating risk aversion and low (or negative) expected consumption growth, which produces the dreaded negative real rate of interest.

But not having any real econ background, Adam's (and Nick's!) posts are admittedly a little over my head at times, so I may not be making any sense.

[Nick Rowe]

Patrick: Think about diminishing marginal utility of consumption. People would normally prefer to have the same consumption today and tomorrow. You can only persuade them to have higher consumption tomorrow than today (postpone consumption) with a high real interest rate. Conversely, you can only persuade them to consume more today than tomorrow with a low (or even negative) real interest rate. Therefore, today's desired consumption depends (positively) on tomorrow's expected consumption, and (negatively) on the real interest rate.

It's a bit more complicated, but you get the same thing with investment.

Since C+I=Y (ignore G and NX), today's desired expenditure (=Y) depends positively on tomorrow's expected expenditure (=Yt+1) and negatively on the real interest rate. A dynamic IS curve.

[Too Much Fed]

Patrick said: "So, I think I'm thinking starting at low interest rates, unemployment feeds into creating risk aversion and low (or negative) expected consumption growth, which produces the dreaded negative real rate of interest."

Are you sure the fed did not produce it?

[Too Much Fed]

For Patrick, Nick, & others:

Nick's post said: "Patrick: Think about diminishing marginal utility of consumption. People would normally prefer to have the same consumption today and tomorrow. You can only persuade them to have higher consumption tomorrow than today (postpone consumption) with a high real interest rate. Conversely, you can only persuade them to consume more today than tomorrow with a low (or even negative) real interest rate. Therefore, today's desired consumption depends (positively) on tomorrow's expected consumption, and (negatively) on the real interest rate."

I'd rather view the situation like this. There are the few who have extremely postive real earnings (think someone working at goldman sachs making more than $1 million per year who spends say $500,000 per year or less). Then there are many others who have negative real earnings (think someone who makes $20,000 per year and needs to spend $30,000 per year). Lower interest rates don't get the excess saver to spend while the excess debtor can only go so far into currency denominated debt and be able to make the interest payments. If the lower interest rates don't get the excess saver to spend and the excess debtor can't go further into currency denominated debt, what happens?