DeLong on fiscal stimulus
In recent years, I get the impression that Keynesians are being more and more aggressive in their arguments for fiscal stimulus. I see economists arguing for stimulus when we are in a deep slump (2009), a sluggish recovery (2011-14) and even full employment (now.) Here’s an example from Brad DeLong:
When should you use fiscal policy to expand demand even if the economy is at full employment?
First, when you can see the next recession coming: that would be a moment to try to see if you could push the next recession further off.
Second, if it would help you prepare you to better fight the next recession whenever it comes.
The second applies now whether we are near full employment or not.
Under any sensible interpretation of where we are now, using some of our fiscal space would put upward pressure on interest rates and so open up enormous amounts of potential monetary space to fight the next recession. It would do so whether or not it raised output and employment today as long as it succeeded in raising the neutral interest rate . . .
Like Mae West, they seem to think that too much of a good thing is . . . wonderful. In my view, one of Bob Lucas’s greatest insights is that you need to think in terms of coherent policy regimes, not just gestures that might or might not seem appropriate at a given point in time. In the standard Keynesian model, fiscal policy cannot be consistently expansionary. For every year of expansionary fiscal policy, there’s another year of contractionary fiscal policy. This is due to two factors:
1. What matters is not the cyclically-adjusted deficit, but rather the change in the cyclically-adjusted deficit.
2. In the long run the federal budget has a constraint; interest as a share of GDP cannot increase without limit.
In fairness, DeLong is focusing not on the direct impact, but rather the indirect effect on interest rates and monetary policy. But I’d also make a practical argument here. Any sort of plausible fiscal stimulus that might have a prayer of getting through Congress today (when the deficit is already $650 billion and rising), is likely to be on the order of a few hundred billion dollars/year, at most. And that’s simply not going to move the equilibrium interest rate enough to have a meaningful impact on the potency of conventional monetary policy. Big Japanese fiscal deficits also failed to significantly boost their interest rates. It will also mean less fiscal “ammunition” in the next recession, when it might be even more necessary (from a Keynesian perspective.)
Nor is the direct impact of stimulus likely to be significant. The Congress suddenly cut the deficit from about $1050 billion in calendar 2012 to about $550 billion in calendar 2013, and it had zero impact on growth (for standard monetary offset reasons). Fiscal policy is a waste of time and money.
I explored these issues in more detail in my recent Lake Wobegon post, and in an even more recent post I pointed out that Janet Yellen suddenly seems to have changed her mind, and now agrees with me.
I’d really like these new Old Keynesians to write down a model on exactly:
1. When do they want the cyclically adjusted deficit to be smaller than average?
2. And exactly when do they want it to be larger than average?
I think I know their answer to the latter question (always), it’s the former question that has me perplexed.
HT: Bob Murphy
Tags:
22. November 2016 at 15:37
I would like to see some clarification on what is the meaning of fiscal stimulus when interest rates approach zero. That is, can we move to helicopter drops or permanent quantitative easing as a form of conventional stimulus?
Why tax if a government can issue 50-year bonds at no interest? What is the point of unnecessary budensome taxes?
Also, I disagree that our economy is anywhere near full capacity. I do not agree that keeping unemployed one out of twenty Americans who want work is a good policy. We are seeing more people attracted into the labor force as demand emerges.
Demand creates supply.
22. November 2016 at 16:08
Fiscal policy is a waste of time and money.
Is there such a thing as no fiscal policy? The state is always spending money (and plenty of it). So in other words: You defined an upper limit of fiscal policy (=interest as a share of GDP) but what is the lower limit? If (expansionary) fiscal policy is a waste of time then why should the state tax and spent at all? At one point is it not a waste of time and money?
22. November 2016 at 16:34
@Christian List
“At one point is [fiscal policy] not a waste of time and money?”
My two cents: it is not a waste when the service being paid for is required (e.g. paying a federal appellate judge’s salary so he/she can rule on a backlog of federal cases). It is a waste when the money is being spent in hopes of creating a side effect (e.g. building a bullet train to the federal appellate judge’s vacation cabin in order to stimulate the economy).
22. November 2016 at 17:28
Christian, I meant fiscal policy in the sense of tax and spending decisions aimed at impacting AD.
22. November 2016 at 18:17
Sorry for the remedial question and I don’t intend to start a long back and forth, just looking for a direct answer here. If fiscal policy can increase money in the pockets of individuals who would spend it thereby increasing the velocity of money, wouldn’t this increase output in the economy?
I understand it may not be the most efficient way to increase output, but it seems to be a way to do it. And if bridges and roads need to be rebuilt it would also seem good.
The appropriateness of the timing in this stage of the business cycle aside, of course. I agree, I don’t see the rationale behind fiscal stimulus for an economy that doesn’t need any stimulating. Like giving an AED to a living functioning body.
22. November 2016 at 19:04
Recently, Bank of Japan Governor Haruhiko Kuroda announced he would buy Japanese ten-year Japanese Government Bonds at 0%. That’s a ceiling on interest rates, not a floor.
The US Fed bought long-term federal bonds during WWII at a ceiling of 2.5% (the Treasure-Fed Accord), effectively monetizing a portion of wartime outlays.
Of course, WWII was incredibly effective as economic stimulus. Was it due to fiscal or monetary stimulus? Are we pretending now WWII federal spending and monetary policy did not boost AD?
Seems to me helicopter money (as we employed in WWII, successfully, through the Accord) is now the way to go. Whether you call that fiscal or monetary stimulus is up to you.
Add on David Beckworth: Beckworth has pointed out about 40% of global economy is “central banked” from the US. Raghuram Rajan says central bankers have global obligations.
The Fed should probably be much more aggressive is promoting growth.
Larry Summers says demand creates supply, and I think he is right.
The world is already glutted with supply of everything, from steel to cars to clothes to smart phones.
It is AD that needs boosting. We could see an extended global boom, if only central bankers would take off the monetary noose.
BTW, Larry Summers pointed out there was a whole range of defeatist economic literature printed near the end of the Great Depression, that limited economic growth was permanent.
Today, U.S. macroeconomists are similarly enfeebled by an ossified orthodoxy.
22. November 2016 at 19:40
Sumner. Ludicrous. Just another day in the office for Scott.
ssumner: “In my view, one of Bob Lucas’s greatest insights is that you need to think in terms of coherent policy regimes”. The RatEx guy said that? I thought our own Marcus Nunes showed that with the ‘genie in the bottle’ article (priceless graphics showing this, if you could visit his site before the paywall). Besides Lucas was prominent only up to two stages out of the four identified by Arnold Kling as: The Forgotten Moderation (1960-1969), The Great Stagflation (1970-1985), The Great Moderation (1986-2007) and The Financial Crisis Aftermath (2008-present). Monetarism only worked during the Great Stagflation (or towards the end) in taming inflation (actually not even then, since when the Volcker Fed raised rates inflation *rose*, and when it lowered rates inflation *fell*, but let’s not go there). During the Great Moderation Keynesianism worked, and during the Great Moderation the Taylor Rule worked. For a brief period (of a few months in 2008) arguably Sumner’s NGDPLT might have worked better than present Fed policy, according to our host. So when did Dr. ‘Bob’ (Robert) Lucas say what he did? Inquiring minds want to know.
ssumner: “For every year of expansionary fiscal policy, there’s another year of contractionary fiscal policy.” – well Keynes did say this in his tome, but in fact we’ve had since I believe FDR, and maybe one year of Eisenhower, perpetual gov’t deficits. Sumner’s never heard of a perpetual bond, which actually pays for itself as long as the growth rate is greater than the interest rate (the inverse of Piketty’s infamous r < g argument)? When and if growth is less than what govt' bonds pay in interest, the debt becomes unserviceable, but otherwise not. So you can actually have infinitely running budget deficits, contrary to what Sumner implies. The RatEx people argue (rather lamely IMO) that de facto people implicitly set aside money when there's a tax cut to pay for future tax increases, but bounded rationality seems to undercut this argument. In short, what the hell is Sumner talking about? Not even he knows.
ssumner: "The Congress [sic? Not sure if that's even correct grammar except in the UK] suddenly cut the deficit from about $1050 billion in calendar 2012 to about $550 billion in calendar 2013, and it had zero impact on growth (for standard monetary offset reasons). Fiscal policy is a waste of time and money." -there you go again. The old 'Fiscal Cliff in 2012 had no bad effect' argument. As Krugman points out, in a 15T economy back then, a mere 500 billion is not that much money to have any affect. But Sumner likes to cling to his myths, including the one about how going off gold helped the USA in 1933 (never mind that Argentina went off gold in 1929 to no avail, and other countries went off gold to no effect, and other countries stayed on gold and yet still came off the lows of the Great Depression). In fact, what happened in the USA in 1933/34 was the cessation of the bank failure crisis, and going off gold to explain the end of this bank crisis in 1933/34 is a simple post hoc ergo propter hoc fallacy.
Fallacy. That's Sumner's middle name.
22. November 2016 at 21:38
Then there’s the fact that everything in Keynesianism is nonsense on stilts from top to bottom, recession or not. It’s methodologically dead wrong and aggregated to the point of meaninglessness. History is squarely on the side of repeated falsifications of the theory, from Post WWII growth, to Post-sequester unemployment declines, to Canada expansionary fiscal contraction, to utterly failed Japanese fiscal stimulus and on and on and on and on.
The only thing Keynesian that rhymes with good macroeconomics is the age-old classical economist understanding that a collapse in total nominal spending can have bad effects.
One can’t be uncharitable enough regarding Delong and Krugman and their brand of defunct macro. It’s unquestionably little more than a thinly veiled technocratic cloak for an agenda built around enlarging the state. Krugman wants us to be the Sweden of the 1980 that Swedes have long rejected. It appears that Delong would like to as well.
I acknowledge that they believe a larger state that controls even more of the economy is, in their eyes, the path to a better world. But all kinds of insanity and evil has been justified by the desire to do good. However, when they keep moving the goal post well beyond the “special case” of the alleged “liquidity trap” and “slack capacity” to basically anytime, anywhere, any reason, I wonder if we should even grant them good intentions. Dishonesty isn’t a sign of good intentions. And considering the way Delong treats others in the field, he’s forfeited any benefit of the doubt even the most charitable critic should extend.
Keynesianism, like socialist central planning, is the economic equivalent of eugenics. A relic of the 20th century we should finally put to rest.
22. November 2016 at 21:51
Lastly, the truly Big Lie of Keynesianism, and one that even Jeffrey Sachs has repeatedly attacked, is the utter disregard for the return on investment of said “stimulus” spending.
It ALWAYS matters how we make use of our scarce resources. Waste is always waste. Excessive, policy-induced unproductive consumption, which I would argue is the overwhelmingly dominant percentage of federal spending, is ultimately a drag on our economy. It uses up our product, rather than adding too it. Keynes mocked a straw man of Say’s law. But insodoing, he undermined the most basic insight of economics: we must produce in order to have the means to consume. Our real income must come from somewhere real. Money wages are an IOU on the value of our contributions to the supply of goods and services. If our income is an IOU borrowed on the proceeds of the production of our children and we proceed to consume rather than invest, we are engaged in little more than a Ponzi scheme.
22. November 2016 at 21:52
That’s my anti-Keynesian rant for 2016. Thanks for indulging me. It’s been a while.
22. November 2016 at 22:02
Professor Sumner
I think Delong thinks that greater supplies of US and other blue chip country debt would b beneficial in a world desperate for safe assets
He has given examples of activities that he thinks better done by governments, like health insurance and retirement funding. I think he’s also talked about universal pre-k. He expects them to cost a lot of money
I would add enough nuke plants or subsidized natgas plants to replace coal, beefier solar and wind subsidies, and EV chargers and fusion moonshot funding and whatever u need to use natural gas as our number 1 transport fuel
they say the grid and our concentrated, high efficiency food system r vulnerable to attack so that might b a nice, expensive thing to change
I would like a biotech industrial policy consisting of mass procurement of drugs/vaccines, including/especially for people in poor countries, and promises to buy up new antibiotics. the market will always underrepresent the social benefits of these techs because so many sick folks r dirt poor. they r expensive to develop – and getting easier to bootleg, I expect – and it seems useful to have more researchers working on the job instead of taking jobs in fields where their long-run contributions will mostly b less important
I would like an agriculture industrial policy that involves large subsidies to make multicrop agriculture more profitable than monocrop agriculture and large Canadian/GWB style cash and food transfers to food insecure societies
I want Morgan Warstler’s welfare program
I want more uni study grants for students from here and elsewhere who want to study engineering, computers, pharma and other techs that seem likely to improve human welfare
I defer to experts who say we need more spending in other fields that I can’t fathom
defense spending seems to have led to some cool stuff, historically, so let ‘er rip
I mean that’s a lot of other people’s money that I’m desperate to spend. Does that qualify as fiscal policy, or does AD have to b the first order concern in each case?
22. November 2016 at 23:30
Government spending: No negative NPVs! And even at 0% interest rates, projects with 30 year life expectancies need to provide 3.33% per year to retire the debt. 30 year Treasuries aren’t at 0%, they’re at 3% and to amortize that over 30 years requires 5.06% payments per year. So if a solar project or train or whatever is to be built, the benefits should be held to that standard. At a minimum.
23. November 2016 at 04:14
Boring. Your argument is correct and obvious.
23. November 2016 at 07:06
“I’d really like these new Old Keynesians to write down a model on exactly:
1. When do they want the cyclically adjusted deficit to be smaller than average?”
When they were Treasury officials in the 90s. In other words, when bond market forces the issue.
23. November 2016 at 07:24
> I’d really like these new Old Keynesians to write down a model on exactly:
> 1. When do they want the cyclically adjusted deficit to be smaller than average?
> 2. And exactly when do they want it to be larger than average?
Start with a standard monetary offset argument and add a zero lower bound. Monetary policy might still be effective at the ZLB, but it’s uncomfortable and policy suboptimal because it makes Serious People write Serious Editorials in the WSJ about punishing savers.
In this model, the Wicksellian natural real rate is a function of the desire to save wealth (not just money) for the future versus the set of available investment opportunities. This rate is not constant.
When the Wicksellian rate is positive, ‘conventional’ monetary policy works away from the ZLB, and in full employment short-term rates are greater than the inflation rate so nobody writes angry WSJ editorials.
The government’s fiscal deficit acts as negative savings. After monetary offset, the nominal steady-state interest rate increases for the same overall inflation rate, indicating that the Wicksellian rate has gone up. This is obvious to anyone who remembers the phrase ‘crowding out’. At the same time *if* we see an undersupply of government investment then increased spending can shift the supply of investments, as the private sector can’t build public works. However, this is not necessary.
In this Old Keynesian-plus-offset napkin model, the answer to the question should be:
1) The cyclically adjusted deficit should be smaller than average when the Wicksellian natural rate is high, and
2) The cyclically adjusted deficit should be larger than average when the natural rate is low.
The core idea is that there’s no harm in crowding out bad, low-return private investments. There may even be a benefit to doing so if you think the market underprices risk, but again this is not crucial.
In your view, this would be a third-best policy option, behind adopting a level target and behind increasing the inflation target. I would argue that it is an improvement over the status quo because of the extra monetary ‘room’, in that central banks are far more comfortable with cutting rates than with OMOs that don’t affect interest rates.
At the moment, the natural real rate is probably somewhere around -1%, so this would suggest more stimulus is warranted.
23. November 2016 at 07:58
Jpd, No, because the Fed will offset any rise in V with a reduction in M.
Dots, I understand that that is his view, and I tried to explain in this post why it’s wrong.
dtoh, You said:
“Boring. Your argument is correct and obvious.”
Not to Delong, who’s smarter than I am. And what other bloggers are making my arguments?
Majromax, But Keynesians have already shown themselves to be unable to determine when interest rates are above and below average. What makes you think they’ll be any more successful in the future?
23. November 2016 at 08:37
> Majromax, But Keynesians have already shown themselves to be unable to determine when interest rates are above and below average. What makes you think they’ll be any more successful in the future?
Keynesians have trouble defining whether interest rates are higher or lower than they should be, but that’s mixing the nominal and real.
Define a consistent real rate, such as the 10-year constant maturity rate less the 10-year breakeven rate. (Using 10-year rate less 1-year CPI growth for earlier data, beginning in the 1980s as the real beginnings of monetary offset).
Compared to a 2-2.5% range prior to the 2008 recession, we see a recent range of -0.5 – +0.5%. Additionally, this drop coincided with the dramatic fiscal austerity.
With a market 10-year real rate around 0%, the private sector feels that there are no real-terms risk-adjusted worthwhile investments to be had on the margin.
23. November 2016 at 08:44
I don’t believe Keynesians exist. I only read about people that claim to be Keynesians to justify increased government spending. When I see one of them suggest cutting spending during good times, I’ll start believing in them. The Delong article you referenced just helps solidify my view.
23. November 2016 at 14:59
Mark Barbieri wins the internet for today …
24. November 2016 at 08:23
Majromax, Keynesians have trouble with both the nominal and the real rate. They have no idea what it means to be above average, going forward.
24. November 2016 at 10:49
Sorry Professor Sumner, I understood your argument as an argument about the usefulness of fiscal stimulus, as such, in expansionary times
25. November 2016 at 07:32
Dots, Well he did suggest stimulus would be appropriate now, which is certainly an “expansionary time”.