Scott Sumner  

Both Barrels Blazing? What would that even look like?

The U.S. Postwar Miracle... Binding Arbitration: The Radic...

I recently listened to a podcast where David Beckworth interviewed Cardiff Garcia, of FT Alphaville. I agreed with most of the things Garcia had to say, including his observations on the recent growth slowdown. He discussed the way that top economists argue about whether it was a supply-side or a demand-side phenomenon, and then asked something to the effect, "Why can't it be both?" Indeed.

He also applied this quite reasonable and pragmatic logic to stabilization policy. If there's a debate about whether or not monetary policy is effective at the zero bound, why not use both monetary and fiscal stimulus? But in this case I'm not so sure, even though I understand where he's coming from.

Here's my claim; most economists don't know how to construct a policy that combines highly expansionary monetary policy with highly expansionary fiscal policy. They don't know what that policy would look like. I'm going to explain this by starting with an example, and then getting into the theory that I believe people get wrong. Let's start with an example:

Suppose that over the past 23 years, one country has run up astounding deficits, which pushed the ratio of debt to GDP up to about 250%. And suppose this country also adopted a monetary policy of near-zero interest rates, and an unprecedented increase in the monetary base---from roughly 10% of GDP to roughly 80% of GDP. Massive deficits, ultra-low interest rates and massive QE. My claim is that this policy combination is what most people have in mind when they think of a combined monetary/fiscal expansion.

And yet it failed miserably. Japan did basically what I've described above, between 1993 and 2016, and their aggregate demand showed perhaps the smallest increase ever seen in a developed economy over such a long period of time. If there is a worse performance for AD over a 23-year period, I'd appreciate if someone would point it out to me. NGDP was essentially flat, for 23 years.

Now I understand that Japan is just one data point, and there could have been mitigating factors. But still, an unprecedentedly large fiscal stimulus combined with near zero rates and unprecedented QE, and you get essentially NOTHING? Those mysterious mitigating factors must be incredibly powerful.

Now let me explain what I think actually happened. I don't believe that the BOJ adopted an expansionary monetary policy. Not at all. Nor do I believe that Herbert Hoover's ultra-low rates and QE of 1932 were expansionary. And that's because I don't believe that either interest rates or QE are the correct measure of the stance of monetary policy. And here's what's weird. On one level almost all economists agree with me. Hardly any economists called Fed policy "tight" in late 2007 and early 2008, even though the growth in the monetary base slowed to roughly zero. Instead, they focused on falling interest rates. But if you ask economists whether high interest rates during hyperinflation mean tight money, they will deny it.

So on one level, economists know that neither the monetary base (i.e. QE) nor interest rates measure the stance of monetary policy. But when it comes to Japan, most economists do believe the BOJ has had an expansionary monetary policy. And indeed I believe that their actual policy since 1993 is pretty close to what economists have in mind when they talk about a combined fiscal/monetary expansion---both barrels blazing.

Screen Shot 2016-09-22 at 1.44.34 PM.png
Instead of combining a fiscal stimulus with a monetary policy that seems expansionary, but actually is not, how about just doing a monetary policy that actually is expansionary. In that case, we wouldn't even need the fiscal stimulus.

So part of the reason I oppose fiscal stimulus is that it's wasteful, but another reason is because I fear that (as we saw in Japan) it would not even work. Instead, I'd like us to think very hard about what sort of monetary policy actually would work. That might be NGDPLT, combined with a "whatever it takes" approach to asset purchases.

Comments and Sharing

COMMENTS (6 to date)
Thaomas writes:

Great, but exactly what BOJ have done to be "actually" expansionary? Should they not just announce that it will continue to buy more and more LT assets until the price level returns to target?

Leave aside "fiscal policy" because we do not know if the expenditures passed an NPV test. [If a government do not do all the projects that DO pass NPV tests, should that not be considered contractionary fiscal policy or "austerity?]

Steve J writes:

I don't get the idea that deficits are equivalent to fiscal stimulus. The recipients of government money do not care if there is a deficit. The people who pay the taxes may be happy to pay less but often the reason they are paying less is because they are making less. If a deficit is created by a recession who involved is going to say they are the recipient of stimulus money?

Scott Sumner writes:

Thaomas, Fiscal policy refers to the budget deficit, not whether projects pass cost/benefit. You can run massive deficits doing sound projects or wasteful projects. In neither case do massive deficits make sense, in my view.

The BOJ should first determine the preferred size of its balance sheet. Then set a NGDP target, level targeting, which will give them a balance sheet of the desired size, and then buy up as many assets as needed to hit the target for expected NGDP growth. This could all be done in a week.

Steve, I'm not saying a deficit would actually stimulate the economy, I'm saying that deficits are how fans of fiscal stimulus define the concept. To be more precise, cyclically adjusted deficits.

Lorenzo from Oz writes:
most economists don't know how to construct a policy that combines highly expansionary monetary policy with highly expansionary fiscal policy.
Wasn't that what Weimar Germany did in the early 1920s? (But perhaps folk don't mean quite that "expansionary".)
Benjamin Cole writes:

Great post but…

"That might be NGDPLT, combined with a "whatever it takes" approach to asset purchases."


Just go to helicopter drops.

Send in the B-52s if you have to.

If asset purchases, then buy foreign bonds, put into the national treasury, and cut FICA (or equivalent in Japan) taxes by an equivalent amount.

No more tweetybirding around.

Jose Romeu Robazzi writes:

Mr. Sumner, I have been thinking about what you always say: QE with sovereign bonds is enough in order to implemente expansionary monetary policy. Then I read John Cochrane (and others) that say that bonds and money are the same thing, meaning, if you exchange bonds for base money nothing changes.

Well, if that is true, then, we should indeed consider the option of the Central Bank buying a variety of private assets, and offloading risk from the private sector. I think that when one starts thinking about the private sector competing for "safe" assets , ultimately what they are looking for is a way of reducing risk, before they start thinking about putting on more risk on their portfolios (i.e. making investment spending).

What I am saying is that buying private assets is much more effective as monetary policy tool than buying government bonds. Also, since the central bank has an infinite investment horizon, it is much easier to defend such a policy since in the long run private assts earn a risk premium, so the fiscal effect may actually be positive...

Comments for this entry have been closed
Return to top