David Beckworth on helicopter drops
David Beckworth has an excellent post explaining why “helicopter drops” (meaning money financed deficits) are not a cure-all for AD shortfalls. Paul Krugman agrees.
I’m actually surprised (but pleasantly surprised) that there is agreement on this point among diverse ideologies. When I did this post I thought I was just about the only one who didn’t see helicopter drops as foolproof way to boost AD. I had thought economists of both the Keynesian and monetarist stripe would say “one thing both Keynesians and monetarists agree on is that a ‘helicopter drop’ would work.” Indeed if you read the post you’ll see me mildly criticizing David Beckworth for what I thought was endorsement of that view. Rereading it I may have missed his point, as he was combining a “helicopter drop” with NGDP targeting. In any case I’m glad to see the helicopter drop myth has been put to rest.
PS. Of course although a “helicopter drop” does not solve the problem, a helicopter drop (no quotes) would work. Shoveling cash out of an actual helicopter would be a very credible “promise to be irresponsible.” And it would be irresponsible, as it’s not needed. NGDP level targeting is enough.
PPS. SG sent me to a Brad DeLong post criticizing Beckworth:
What David Beckworth misses is that if quantitative easing is used to fund expansionary fiscal policy-if the government buys not long-term Treasury bonds but, rather, bridges and NIH research and the human capital of twelve-year olds-then those asset purchases are going to be permanent: you cannot unwind those transactions. Hence, by Beckworth’s logic, that policy will be effective.
Yes those specific real output transactions are not reversible, but the money used to finance them can be later withdrawn via open market sales, and expected future fiscal policy can tighten creating somewhat of a Ricardian equivalence problem. Now in fairness Ricardian equivalence only applies 100% to taxes and transfers, not real expenditures. So I agree with DeLong that as a practical matter his plan is likely to be effective.
But not necessarily. If the government announced a 3% of GDP money financed fiscal stimulus for 2014, and also announced that beginning in 2015 they would reduce M2 at 20% per year for 4 years, you’d have a Great Depression expected in 2015, and that expectation would cause a Great Depression to begin in 2014, as the crash in business investment in 2014 overwhelmed the puny fiscal stimulus. As a theoretical matter it’s still all about the future path of policy; current actions are relatively unimportant.
PPPS. If Congress doesn’t renew extended UI benefits, as now seems likely, we’ll have an interesting experiment in 2014. I’m on record as believing extended UI raised the unemployment rate by around 0.5% (DeLong seems to agree with me on this point), but have an open mind on the issue.
There’s also the issue of whether monetary offset applies to this sort of policy change–I say partly but not completely as these are lower skilled workers (on average) and hence RGDP changes less than employment when they find jobs. Also it’s a positive supply shock that does not raise inflation. Of course monetary offset does apply to the demand-side effects of lower UI payments, and with Yellen coming in I think monetary policy will be as aggressive in doing offset, if not more so, than under Bernanke. My hunch is RGDP growth might pick up slightly in 2014 because of:
1. Yellen.
2. Less UI supply-side effects.
3. Possible European recovery boosting US manufacturing.
4. Stock market seems optimistic. Rising long term rates.
I emphasize slightly faster–I’m not expecting a boom. We are also getting a slight reduction in austerity, but probably not a big enough policy swing there to have another offset test. Most forecasters expect a slight pickup in growth, so I’m not exactly going out on a limb here.
PPPPS. Alan Blinder says he was the first to recommend negative IOR. I don’t think so. I had two papers in The Economists’ Voice in early 2009 (especially the second), the earliest I could google Blinder promoting the idea was 2010.
HT: Frank McCormick
Tags:
11. December 2013 at 06:32
July 1st, North Carolina ended long term unemployment payments:
http://www.journalnow.com/news/state_region/article_7454f7d4-5391-11e3-8af5-0019bb30f31a.html?mode=image&photo=0
11. December 2013 at 06:35
Beckworth appears to be saying that a CB can always use monetary policy to reverse the effects of fiscal policy (helicopter drops in this case). This is not the same as saying that there are no circumstances where helicopter drops will be needed when monetary policy has reached its limits.
Sometimes at the ZLB monetary policy (swapping one asset for another) may have little effect on AD where fiscal policy will have a big impact. The fact that monetary policy could limit this effect is interesting but doesn’t negate this point.
11. December 2013 at 06:55
Scott,
Isn’t it enough to say that at the moment, the evidence indicates that the Fed is steering the nominal economy, and that all fiscal policy is therefore a pure supply-side issue? I don’t understand why it’s even necessary to discuss Ricardian equivalents (though, in fairness, i’m not an economist so I probably don’t even understand what Ricardian equivalence even means).
11. December 2013 at 07:13
Morgan,
You should go to this site and view the NC labor participation and employment rates over the same period.
http://esesc23.esc.state.nc.us/d4/LausSelection.aspx
Much of the drop in the unemployment rate in NC can be explained by the lower labor participation rate. I suspect that some of those who were collecting UI benefits stopped claiming they were “looking for a job” after benefits were terminated and there was no continued reason to make such a claim.
On the other hand, total employment (particularly seasonally adjusted) is up, but not as much as the unemployment rate would suggest.
11. December 2013 at 08:09
a new paper with something to say about helicopter drops having effects in absence of nominal rigidities.
http://cep.lse.ac.uk/pubs/download/dp1249.pdf
11. December 2013 at 08:34
Vivian, what I’ve learned from economists is that they don’t count drop out rate.
11. December 2013 at 08:47
Scott, You say in reference to helidrops, “And it would be irresponsible, as it’s not needed. NGDP level targeting is enough.” That confuses two issues. First, there is the monetary versus fiscal argument. Second, there is the NGDP versus inflation targeting issue. To illustrate, NGDP targeting can perfectly well be done with either pure monetary policy or pure fiscal policy.
Monetary policy is defective in that it channels stimulus into the economy just via borrowing and investment, and there is no prima facie reason to assume a recession is caused by a fall in borrowing and investment rather than by a fall in consumer spending, public spending or exports.
Moreover, and as regards the recent crisis, while a drop in investment was obviously a major cause, it was a fall in totally irresponsible borrowing and investment (NINJA mortgages, etc). To cut interest rates in those circumstances (rather than go for helidrops or more conventional fiscal policy) will just reflate housing bubbles. And that’s exactly what’s happening, at least according to Nouriel Roubini. See:
http://www.project-syndicate.org/commentary/nouriel-roubini-warns-that-policymmakers-are-powerless-to-rein-in-frothy-housing-markets-around-the-world
11. December 2013 at 08:59
Morgan,
If your point is that the unemployment rate does not consider “the drop out rate” that’s clear. The unemployment rate is the percentage of those actively seeking work that don’t have jobs. Actually, that was precisely my point and one doesn’t need an economist for that.
On the other hand, economists (and others) are normally concerned, in assessing the labor market condition, with the number of persons seeking work *and* those who have are not looking. There may be a number of reasons for the latter, including lack of available jobs, but also lack of appropriate incentives that would encourage them to do so.
In the case of NC, I’ve seen a few posts online that contend the unemployment rate drop in NC is not significant because it was accompanied by a drop in the participation rate post July 2013. That may be a legitimate critique of the point you appear to want to make with the data. All of those who were previously claiming to be in the workforce had to be in order to get UI. I would contend that this critique is perhaps unwittingly damaging to the critic’s side if the reason for the post-July drop in the participation right is, in fact, mostly people who where on the UI roles claiming to be looking for work only because they had to in order to get UI benefits and now that that pretense has been dropped there is no reason to maintain it.
Let me state this differently: Of all the persons previously on NC UI, 100 percent of them, almost be definition, were in the “available workforce”. Post July 2013 some of those would have actually started to look for work (rather than pretend to). Others would have dropped out. In any scenario I can think of, with respect to this cohort, the participation rate would decline as a result of this change.
For the above reasons, I prefer, and I think most “real” economists do, too, the percentage of working age adults who have jobs to those who merely claim they are looking for jobs.
11. December 2013 at 09:17
Ralph: “Monetary policy … channels stimulus into the economy just via borrowing and investment”
Nope, that’s not the monetary policy transmission mechanism. The two most important mechanisms are future expectations (medium/long-term) and the HPE (short term). Given that you misunderstand the mechanism, it’s no surprise that you come to the policy conclusions that you do.
11. December 2013 at 09:41
As to UI if it’s not extended and next year unemployment goes down .5% how to we know it’s the UI? The unemployment rate has been going down about the same amount every year anyway. It’s down about that amount this year.
Forecasters already expect it to drop further next year.
11. December 2013 at 09:59
I am dubious that a literal helicopter drop of cash, would actually have a meaningful impact on output or employment.
It is not an experiment I would like to see carried out, but alas, its just not going to happen.
11. December 2013 at 10:15
Market Fiscalist, No, monetary stimulus is just as effective at the zero bound as at other times. It just might require a bigger base.
Ralph, You said;
“To illustrate, NGDP targeting can perfectly well be done with either pure monetary policy or pure fiscal policy.
Monetary policy is defective in that it channels stimulus into the economy just via borrowing and investment, ”
Both statements are false. Fiscal policy alone cannot target NGDP. As far as I know no economist has ever proposed using fiscal policy to target inflation or NGDP. And monetary policy does not work through borrowing.
You said;
“To cut interest rates in those circumstances (rather than go for helidrops or more conventional fiscal policy) will just reflate housing bubbles.”
Cutting interest rates is not a policy it’s an effect of an expansionary monetary policy. More money lowers short term rates regardless of whether it’s injected via OMOs or helicopter drops.
Mike, The test is whether the drop is more or less than expected. Of course as with any macro test, you never “know” the answer for certain.
Doug, A large one certainly would.
11. December 2013 at 10:30
Don,
Where’s the evidence that expectations are of overriding importance? There’s a Bank of England publication on the monetary transmission mechanism which is very unsure as to what the main mechanisms are. Plus there’s an economics tutorial article which doesn’t mention expectations. See respectively:
http://www.bankofengland.co.uk/publications/Documents/other/monetary/montrans.pdf
http://www.tutor2u.net/economics/content/topics/monetarypolicy/policy_transmission_mechanism.htm
As to HPE, what’s that? Higher Price Expectations?
11. December 2013 at 10:48
Here’s something odd that I came across by Scott Fullwiler and Stephanie Kelton:
“This all confirms Krugman’s point – there is no meaningful difference between what the government is doing today (case 1) and what Market Monetarists are urging it to do via helicopter financing via direct money creation (case 2). As Krugman says, “the results are the same.”
http://neweconomicperspectives.org/2013/12/krugman-helicopters-consolidation.html
To my knowledge only Market Monetarist that has advocated helicopter drops is David Beckworth, and only then when done in conjunction with NGDPLT. My impression is that David offered that proposal in a moment of weakness, giving in to pressure from Cullen Roche and the Monetary Realists (MR). (If I am wrong then somebody please correct me.)
Now I understand that MR likes to distance itself from MMT, and L. Randy Wray considers MR to be an inferior imitation of MMT:
http://www.economonitor.com/lrwray/2013/12/11/mmt-often-imitated-never-duplicated/
But it’s interesting that Fullwiler and Kelton seemed to totally miss the point that Krugman was agreeing with Beckworth, and by extension the Market Monetarists, that helicopter drops by themselves can’t make up for bad monetary policy, and that the pressure for helicopter drops has mostly come from a group they consider to be a substandard copy of themselves.
11. December 2013 at 10:56
Scott,
I’m baffled as to why fiscal policy cannot target NGDP. Assuming crowding out doesn’t TOTALLY negate the supposed stimulatory effects of fiscal stimulus (and most economists agree that crowding out is not that serious), then a government could perfectly well aim to achieve an X% nominal expansion in GDP per year using fiscal stimulus alone. (Incidentally by “fiscal stimulus” I mean: government borrows $X, spends the money and gives $X of bonds to those it has borrowed from.)
I’m sure you’re right to say that targeting NGDP using fiscal policy alone has never been tried. Indeed, I wouldn’t favour that policy. But there is no reason it couldn’t be done: in a serious recession, it would be better than nothing.
Re your claim that “monetary policy does not work through borrowing” that’s not the view of the Bank of England publication (and the other publication) mentioned in my answer to Don above.
You then say that interest rate cuts result from money supply increases. Agreed. Then you say, or seem to say that disproves my claim that interest rate cuts may reflate housing bubbles.
I’m baffled: surely it doesn’t make a blind scrap of difference what causes an interest rate cut – one effect will be that house buyers will tend to borrow more.
11. December 2013 at 11:06
It seems to me one difference between QE and Helicopter Drops is the central bank’s credibility in being able to subsequently withdraw the base money. Obviously, with QE they can withdraw the money by selling back the bonds they have purchased. But with a Helicopter Drop wouldn’t it depend on how big the central bank’s balance sheet is? If the CB kept financing gov’t deficits with money instead of the Treasury with bonds wouldn’t you reach a point where they no longer had the credibility or maybe even the capacity to withdraw all of the additional base money? At that point some fraction of the additional money would have to be permanent and the price level would respond accordingly. Now, maybe if the CB can pay IOER then this is another means of providing monetary contraction to keep the price level in check. But what if the CB is not allowed to pay IOER? This is a point of confusion I’ve always had with “helicopter drops”. Could someone clear this up?
11. December 2013 at 11:09
Ralph Musgrave
“Monetary policy…channels stimulus into the economy just via borrowing and investment”
There are nine channels of the Monetary Transmission Mechanism (MTM) as enumerated by Frederic Mishkin. The following paper by Mishkin gives an overview of the MTM:
http://www.iset.ge/old/upload/01%20Mishkin.pdf
The Channels of Monetary Transmission: Lessons for Monetary Policy
Frederic S. Mishkin
May 1996
Abstract:
“This paper provides an overview of the transmission mechanisms of monetary policy, starting with traditional interest rate channels, going on to channels operating through other asset prices, and then on to the so-called credit channels. The paper then discusses the implications from this literature for how central banks might best conduct monetary policy.”
You might find the following table, found in the author’s best selling intermediate monetary economics textbook (“The Economics of Money, Banking and Financial Markets”) useful:
http://economistsview.typepad.com/.a/6a00d83451b33869e201901c401aea970b-500wi
Yes, Mishkin’s paper is a trifle simplistic, but it provides a nice frame of reference. And yes, four out of the nine channels involve loans or lending activity, and six out of the nine channels impact physical investment (along with other types of spending). But it’s simply wrong to say that monetary policy only works through borrowing and investment.
Moreover, one of the key points that MMT correctly makes against Austrian Business Cycle Theory (ABCT) is that investment and consumption are positively correlated, not negatively. You seem to be making an argument that has more in common with ABCT than with MMT.
11. December 2013 at 12:06
Ralph Musgrave,
“To cut interest rates in those circumstances (rather than go for helidrops or more conventional fiscal policy) will just reflate housing bubbles…Then you say, or seem to say that disproves my claim that interest rate cuts may reflate housing bubbles.”
The IMF has made it clear that loose monetary policy is not the primary cause of asset price bubbles (Page 106-107):
“If monetary policy were the fundamental cause of house price booms over the past decade, there would be a systematic relationship between monetary policy conditions and house price gains across economies. Certainly, average real policy rates were low and even negative in some economies, and Taylor rule residuals were mostly negative, suggesting that monetary policy was generally accommodative across economies during this period. But there is, at best, a weak association with house price developments within the euro area (Figure 3.13, blue lines).22 And there is virtually no association between the measures of monetary policy stance and house price increases in the full sample (Figure 3.13, black lines). For example, whereas Ireland and Spain had low real short-term rates and large house price rises, Australia, New Zealand, and the United Kingdom had relatively high real rates and large house price rises. Moreover, the association between measures of the monetary policy stance and real stock price growth is extremely weak, whether assessed during the global house price boom (2001:Q4-2006:Q3; not shown) or during a later period, when stock markets rallied from their troughs (2003:Q1) through the stock market declines of 2007 (Figure 3.14).
The fairly regular behavior of inflation and output and the fact that Taylor rule residuals were not associated with recent asset price rises across economies in the sample suggest that monetary policy was not the main or systematic source of the recent asset price booms.23”
http://www.imf.org/external/pubs/ft/weo/2009/02/pdf/c3.pdf
Figure 3.13 is similar to the following graph (note the near zero coefficient of determination):
http://www.federalreserve.gov/newsevents/speech/bernanke20100103slide9.gif
11. December 2013 at 14:19
“No, monetary stimulus is just as effective at the zero bound as at other times. It just might require a bigger base.”
I’m missing this point. If you swap bonds for money then you have just changed people’s assets allocation. People may feel that they are holding too much money and create a HPE on other assets that will end up increasing asset prices. However while asset prices increase , the revenue stream they generate does not, so its not clear that this will increase AD.
Against this: Unless you believe in full Ricardian equivalence then fiscal deficits are guaranteed to increase AD.
11. December 2013 at 18:55
Mark, None of that confusion surprises me.
Ralph, You said;
“I’m baffled as to why fiscal policy cannot target NGDP. Assuming crowding out doesn’t TOTALLY negate the supposed stimulatory effects of fiscal stimulus (and most economists agree that crowding out is not that serious), then a government could perfectly well aim to achieve an X% nominal expansion in GDP per year using fiscal stimulus alone. (Incidentally by “fiscal stimulus” I mean: government borrows $X, spends the money and gives $X of bonds to those it has borrowed from.)”
Mishkin’s textbook explains why that’s impossible. V would have to rise every year, meaning interest rates would have to rise every year, meaning real interest rates would have to rise every year, meaning the budget deficit would just explode.
Regarding bubbles, I thought you were distinguishing between different types of monetary expansion.
Danny, It’s confusing because with “helicopter drops” the central bank accumulates bonds, but with helicopter drops (no quotes) they don’t.
Market Fiscalist, the expansionary effect of monetary policy does not come from people having more wealth or income. It comes from the hot potato effect.
11. December 2013 at 20:24
The HPE will cause other assets to rise in value but unless this causes peoples expected income stream to rise why will this be expansionary ?
Monetary policy st the ZLB seems to rely upon some “psychological” effects that depend upon people thinking its going to work and acting like it will work even though there is no actual reason why it would work beyond these ungrounded expectations.
11. December 2013 at 23:50
Scott,
Yes I realise that, as you put it, “budget deficit would explode”. Nevertheless, fiscal policy alone could be used for a year or two to counter a recession, especially if a country had a low national debt to start with. But to repeat, I’m not serious advocating that policy: I’m just saying it’s a possibility.
Mark Sadowski,
Thanks for all that detailed information. Obviously it will take me time to digest it.
12. December 2013 at 03:23
Scott,
“V would have to rise every year, meaning interest rates would have to rise every year, meaning real interest rates would have to rise every year”
Why would V have to rise every year?
12. December 2013 at 06:26
Market Fiscalist, The exact same thing that causes asset prices to rise will cause expected future NGDP to rise. That causes current NGDP to rise. If wages are sticky that causes output to rise.
Ralph, I misunderstood you, I thought you meant for the long haul. Economists usually think in terms of long run policy targets, not just for a recession.
Philippe, He was holding monetary policy constant, and doing it all with fiscal stimulus.
12. December 2013 at 07:22
“The exact same thing that causes asset prices to rise will cause expected future NGDP to rise. That causes current NGDP to rise. If wages are sticky that causes output to rise”
Why do rising asset prices cause future (and therefor current) NGDP to rise? If you assume that people will base current consumption on expectations of future income streams then why do rising assets prices that are simply the inverse of falling returns on investment cause people to consume more ?
Is there an assumption here that people will spend more when their nominal wealth goes up irrespective of how they expect this to affect their future real income ?
12. December 2013 at 07:33
Mark Sadowski,
I had a look at that Miskin paper. I obviously over-stated my case when I said that “Monetary policy…channels stimulus into the economy just via borrowing and investment”. But I don’t think I need to make a HUGE alteration to that statement. I.e. I’ll change it to something like “Monetary policy brings stimulus in a way that is weighted towards borrowing and investment, and there is no obvious justification for that or any type of weighting.”
Next, you say that my points have something in common with ABCT Austrian stuff. I certainly didn’t intend to say anything like that. I disagree with Austrians 90% of the time. Perhaps you thought that in criticising interest rate adjustments I was saying they had no effect at all. Actually I think they do have an effect: however it’s the above “weighting” I object to. That is, if an economy is not producing enough goods and services to bring full employment, strikes me the best solution is to give consumers more of the stuff that enables them to buy goods and services, and that’s money. And the other very large consumer is government, so government or public spending needs to be boosted as well. And in reaction to that increased demand, some employers will invest more.
Re the effect of interest rates on housing bubbles, I accept there isn’t a close relationship there. On the other hand I find it difficult to believe that house buyers are totally indifferent to interest rates.
12. December 2013 at 09:07
Ralph Musgrave,
Thanks. I greatly appreciate your restatements and will let it go at that.
12. December 2013 at 11:04
Ralph: HPE = Hot Potato Effect. Increasing the monetary base results in consumers holding more currency than they desire at current levels of NGDP, and their attempts to “get rid” of the excess currency are a direct causal mechanism to increase NGDP.
Re: your BoE links: One of Sumner’s main claims is that the world’s central bankers (e.g. Japan, Europe) radically misunderstand their own jobs, and have deliberately done their economies harm. They certainly might be explaining what they think they are doing, but the point is that they are actually wrong about macro.
“if an economy is not producing enough goods and services to bring full employment”
The quantity of goods & services produced, is not a causal explanation for the level of employment.
12. December 2013 at 19:48
Market Fiscalist, You said;
“Why do rising asset prices cause future (and therefor current) NGDP to rise?”
Reread what I wrote, I did not say rising asset prices cause NGDP to rise (although I do think they may have some effect.) I said the thing that causes asset prices to rise (the hot potato effect) also causes NGDP to rise.
12. December 2013 at 21:45
Morgan & Vivian,
I have graphed some of the North Carolina labor data to try to see what’s going on.
http://idiosyncraticwhisk.blogspot.com/2013/12/a-natural-experiment-on-emergency.html
13. December 2013 at 07:28
Mark Sadowski…
I read the paper you linked and I still find the original comment you were responding to accurate. Although it may be too reductive a central bank that targets interest rates effects the economy through effects on lending (or expectations of future lending). To use the MMT type language I guess they would focus on how interest rate targeting means the money supply is endogenously determined by demand for loans (in an interest rate targeting regime). I don’t see how any of the other mechanisms mentioned don’t directly relate to that channel.
While I’m not a full on MMT person there is nothing wrong with certain points of theirs.
The existence of bubbles is perhaps interesting I feel as if it is mostly unimportant to macro with a competent bank?
Also we should stop calling it a housing bubble or housing price appreciation or anything with the word house. Houses are durable goods that depreciate. Replace the word house or housing with “land” and I believe people would be more rational discussing the issue.
13. December 2013 at 20:20
“Reread what I wrote, I did not say rising asset prices cause NGDP to rise (although I do think they may have some effect.) I said the thing that causes asset prices to rise (the hot potato effect) also causes NGDP to rise.”
I read the paper that Mark Sadowski linked to above and that described multiple channels through which monetary policy may work. I need to digest that. Potentially that may explain why the HPE on assets may cause NGDP to rise.
17. February 2017 at 06:34
[…] me on this point), but have an open mind on the issue.” http://www.themoneyillusion.com/?p=25288 Here’s the trouble though. He has an open mind on the […]