Scott Sumner  

Can the economy handle a 20% fiscal contraction (at near zero interest rates?)

PRINT
Weekend Grab Bag... The U.S. Postwar Miracle...

Here's Matt O'Brien:

Libertarian presidential candidate Gary Johnson is a friendly guy, seems pretty moderate. But he could tank the economy.

That's what trying to balance the budget all at once would do. Which, of course, is what Johnson says he would. He wants to cut spending by 20 percent next year to get the government back in the black, and then veto any legislation that would make the red ink return.

This probably wouldn't end well. The problem is the Federal Reserve might not be willing or able to really counteract this. In normal times, you see, the Fed cuts interest rates when the government cuts the deficit so that the private sector can pick up the slack for the public sector. But even eight years after the Great Recession, these are still not normal times. The Fed can't cut interest rates right now, because they're barely above zero. Now, it's true that the Fed could print money instead -- that's how it stopped austerity from starting a recession in 2013 -- but Johnson doesn't want the Fed to do that. He's said that quantitative easing, which is when the Fed buys bonds with newly created dollars, is just an attempt to "override the free market" that will only lead to "malinvestment, inflation, and prolonged unemployment." And since he would not only get to pick two Fed members in 2017, but also a new Fed chair in 2018, what he thinks matters.


I mostly agree with this, but not entirely. I certainly agree that the Fed could offset fiscal austerity with a more expansionary monetary policy, just as they did in 2013. But it's not quite right to say that the Fed cannot cut interest rates. The current rate of interest on reserves is 0.5%, and that rate could be cut by 100 basis points, to minus 0.5%. Nonetheless O'Brien is right that interest rate cuts might not be enough.

But I also think he slightly exaggerates how much we know about this issue. Let's take the policy environment at the end of WWII. Here are some relevant facts:

1. There was massive austerity, as government spending fell by far more that 20%. We suddenly went from a deficit of 20% of GDP, to a surplus.

2. Several top Keynesian economists warned that this austerity would lead to another post-war depression.

3. Short term interest rates were very low, but slightly above zero (0.38% on T-bills, for instance.)

Sound familiar? Here's what happened next. After WWII, the Fed did not cut rates at all to offset the fiscal austerity. Indeed after holding them at 0.38% for about 2 years, they began gradually raising them in mid-1947. Nor did they do any QE. And despite all that, the economy remained fine, with the unemployment rate fluctuating between 3% and 5% throughout 1946, 1947, and 1948, despite millions of men suddenly being discharged from the military. The Keynesian predictions did not come true.

I recognize that there are lots of differences between 1946 and today. But even so, it should give us all pause to consider that well-informed Keynesian economists got this wrong. Maybe there is something wrong with the model.

And let's not forget that Johnson is also proposing many positive initiatives that would boost aggregate supply, such as tax reform.

This is not to say that I agree with everything Johnson is proposing. I disagree with him on the wisdom of balancing the budget so quickly, and I disagree with him on QE, at least as an option. Nonetheless, the post-WWII experience should make us all very cautious about predicting the impact of fiscal austerity.

Screen Shot 2016-09-23 at 4.18.40 PM.png
PS. A few months back Trump proposed paying off the entire national debt in 8 years, which is an even more contractionary proposal. These sorts of proposals need to be taken with more than a grain of salt.

PPS. I recommend this David Beckworth interview of Jason Taylor, which touches on some of these issues. Taylor says that Keynesians predicted 25% to 35% unemployment if the government suddenly discharged 10 million soldiers, and also suddenly slashed massive military spending. The government did exactly that, and unemployment averaged 3.9% in 1946 and 1947. (The specific discussion occurs after the 47 minute mark.)

Screen Shot 2016-09-23 at 4.22.16 PM.png


Comments and Sharing






COMMENTS (17 to date)
Brian Weiner writes:

I could be mistaken but in at least one
interview recently Gary clarified that he did not expect a balanced budget in Year 1.

He says they would propose a balanced budget to congress and allow them to justify where they need deficits. Willing to sign a budget with a deficit, but not a tax increase.

Ben writes:

Sorry Sumner, I normally agree with you but this post is just ignorant.

Real GDP declined an unprecedented 12.7% from 1944 to 1946. Hardly "fine". Sure, unemployment didn't increase substantially at all which was likely due to a high demand for workers as women returned home and left the workforce again.

You can safely say that a sharp 20% reduction in government spending would lead to massive decline in RGDP as seen in 1946 and unemployment would certainly spike. To say otherwise would be crazy.

Khodge writes:

More recently, we heard same argument when the 2013 sequestration went into effect. At that point the economy showed the beginnings of healing.

BC writes:

I have a related question on balanced budget amendments. The usual argument against is that a balanced budget amendment prevents fiscal stimulus during a recession. However, this argument seems to ignore monetary offset. At least away from the ZRB, and maybe even at zero interest rates if the Fed adopts level targeting, what are the arguments against a balanced budget amendment?

Scott Sumner writes:

Thanks Brian.

Ben, You said:

"Real GDP declined an unprecedented 12.7% from 1944 to 1946. Hardly "fine"."

Of course it was "fine", as living standards (consumption) rose sharply. Who cares if we are building less military equipment if the war was over?

You said:

"You can safely say that a sharp 20% reduction in government spending would lead to massive decline in RGDP as seen in 1946 and unemployment would certainly spike."

Keynesian economists were very confident in predicting high unemployment in 1946; what was wrong with their model?

Roughly 350 Keynesian economists signed a letter warning of a possible recession if fiscal austerity was implemented in 2013. Instead growth sped up. What was wrong with their model?

I concede that RGDP might well fall with a sharp fiscal contraction, and I would oppose reducing the deficit that quickly, but I'd have more confidence in that prediction if I knew what went wrong with these earlier predictions. Instead I see Keynesians simply ignore the issue.

Khodge, Good point.

BC, The classic argument is that you want to run deficits during recessions for "tax smoothing" reasons. It helps to minimize the deadweight loss of taxes. But yes, the argument four automatic stabilizers is weaker with monetary offset.

B Cole writes:

The Bank of Japan has paid off 1/3 of Japan's national debt. Paying off the national debt through QE appears a viable option. In eight years? Seems aggressive. As a longer term schedule? Why not?

MikeP writes:

Real GDP declined an unprecedented 12.7% from 1944 to 1946. Hardly "fine".

I admire your confidence in determining the real value of war materiel that has one producer and one consumer and is intended to be shipped overseas to be destroyed.

Scott Sumner writes:

Ben, Actually they have not done so, as they've replaced interest bearing bonds with interest bearing reserves.

Radford Neal writes:

"they've replaced interest bearing bonds with interest bearing reserves"

An interesting point, but it does seem like there's a difference here. Deciding not to pay interest on bonds would be a default, and as a result have various bad consequences. But am I right in assuming that deciding to no longer pay interest on reserves would not be a default? Does the Japanese (or other) central bank have any legal obligation to keep paying interest on reserves?

Jon Murphy writes:

@Ben:

Real GDP declined an unprecedented 12.7% from 1944 to 1946. Hardly "fine".

Remember how GDP is measured: Y=C+I+G+NX, where:
Y= GDP
C= Consumption
I= Investment
G= Gov't spending
NX = Net Exports (Exports - Imports)

All else held equal, GDP will fall when government spending falls, even if such a fall increases the welfare of the economic actors (as Scott points out). And interesting corollary to this is, as Bob Higgs pointed out, during World War 2, GDP rose (due to the G spending), but standards of living fell dramatically: rationed food, rationed goods, etc. If we were to only look at the GDP, it'd look like the US was doing amazing during World War 2 (in the neighborhood of 5% growth). However, the exact opposite is true.

GDP is a useful measurement, but one must be careful of its limitations, especially when discussing economic well-being.

Scott Sumner writes:

Radford, No, it would not be obligated to. But if it did not do so when interest rates were positive, then you'd have a lot of inflation. Any government with its own money has the option of monetizing its debt, if it is willing to accept high inflation.

Thaomas writes:

If Mr Johnson can find 20% of the Federal budget that is producing no benefits to anyone (the output of the expenditure, not the income of persons producing it and the Fed were committed to maintaining NGDP, then no problem. In the space-time continuum in which we live, it's a different story.

Jon Murphy writes:

@Thaomas:

If Mr Johnson can find 20% of the Federal budget that is producing no benefits to anyone...

Such a condition would be impossible to obtain as everything benefits someone, even if it's just one person. After all, thievery benefits the robber at the expense of the robbed. So, I humbly recommend a different criteria.

That said, eliminating 20% (approximately $760 billion) wouldn't be too hard to do.

-The GAO reported $125 billion in waste, so that gets us down to $635 billion.

-The wars in Iraq and Afghanistan are costing some $400 billion a year, so ending those could get us down to $235 billion.

-Cutting back operations in Syria and Libya could easily knock that down to $200 billion.

-Medicare and Medicaid reform could easily knock that down to $100 billion.

-Eliminating corporate subsidies like the farm subsidies, Ex-Im bank, etc could same some $20-30 billion, so we're down to $70 billion.

-A GAO report (separate from the one I mentioned earlier) reported some 22% of federal programs provide no positive impact on the communities they serve. Eliminating those could save another $123 billion. So now I've covered the 20%, and now am saving an additional $53 billion.

-The CBO also has a list of useless budget items, and additional $100 billion in cuts can be made there. So now we're positive $153 billion.

These are items I've picked just by reading the GAO and CBO reports. Low hanging fruit that could be picked on day one.

But why aren't they? These reports are no secret. They're presented to Congress. The figures I quote are in the abstract or executive summary. But, as I said earlier, someone always benefits and they will fight tooth and nail to keep those benefits.

Maurizio writes:

Fantastic post!

D Stonebraker writes:

I've seen it said that the "Keynesians predicted disaster after WW2" claim is a myth. Is it not?

What is the source of this claim?

Scott Sumner writes:

D. Stonebraker, The source is the Beckworth interview of Jason Taylor, linked to in this post.

Justin writes:

--"The GAO reported $125 billion in waste, so that gets us down to $635 billion."--

Much of that $125 billion consists of improper payments, particularly from Medicare and Medicaid. I suspect it will be difficult to reduce this amount meaningfully, and certainly not by anything close to 100%. To be very generous, let's assume the government can aggressively go after improper payments and other spending universally regarded as waste and reduce this value by $50 billion/yr.

--"The wars in Iraq and Afghanistan are costing some $400 billion a year, so ending those could get us down to $235 billion."--

I'm not sure what your source for that number is, but spending for overseas contingency operations (OCO) requested for FY2017 is just $58.8 billion. To the degree you're including veteran costs and interest on past spending, that's a sunk cost we can no longer cut.

http://comptroller.defense.gov/Portals/45/Documents/defbudget/fy2017/FY2017_Budget_Request_Overview_Book.pdf

--"Cutting back operations in Syria and Libya could easily knock that down to $200 billion."--

Operations against ISIL are included in the $58.8 billion of OCO funding, so no additional savings there.

--"-Medicare and Medicaid reform could easily knock that down to $100 billion."--

We're already assuming significant declines in improper payment for those programs - what program do you think it is possible to politically enact that would reduce spending by $100 billion/yr right out of the gate? Immediate benefit cuts probably aren't possible. The only thing I can think of is allowing Medicare to negotiate prescription drug prices. For context, if Medicare were able to negotiate drug prices as well as Denmark, that would save an estimated $541 billion over 10 years, so probably something like $40-$45 billion in year 1. I don't see this being likely, but let's say $45 billion in prescription drug savings.

http://healthaffairs.org/blog/2016/02/01/how-would-government-negotiation-of-medicare-part-d-drug-prices-work/

--"-Eliminating corporate subsidies like the farm subsidies, Ex-Im bank, etc could same some $20-30 billion, so we're down to $70 billion."--

I'm fine with this, I'm not sure if it's something Johnson will be able to pull off politically, but we'll add $30 billion to the total.

So by my accounting, we are still short $576.2 billion.

Comments for this entry have been closed
Return to top