Scott Sumner  

Mission creep

Imprisoning Immigrants: What t... Totally Conventional Views Whi...

During periods where there is an enormous demand shortfall, such as the 1930s and the period after 2008, there is a resurgence of interest in demand-side models. This is good. But it also produces an unfortunate side effect---mission creep.

Demand stimulus can help reduce unemployment. It can also ease debt burdens. But that's pretty much all it can do. Unfortunately, many pundits are claiming that demand stimulus is some sort of magic potion, which can cure structural problems, boost productivity, increase labor force participation, etc. If we learned anything from the 1970s, it's that demand stimulus cannot solve supply-side problems.

Here's Barclay's Bank:

"We estimate that core inflation ex-VAT effects eased to 0.3% y/y in January. With the sharp drop in oil prices, we believe the BoJ's 2% target will remain out of reach even toward end-2016. We estimate that unemployment stayed at 3.4% while jobs/applicants improved slightly. Policies are needed to resolve labor shortages and alleviate structural unemployment, turning attention to the third "arrow" of Abenomics."
I think this is right. Demand stimulus is not likely to boost growth in an economy with "labor shortages." Or to be more specific, we should only expect a slight boost to growth. And by the way, although the "third arrow" is needed, the terminology is unfortunate. The "second arrow" was supposed to be fiscal stimulus, but Japan did not do fiscal stimulus, they sharply raised taxes. So it's all about monetary stimulus and structural reforms, as it should be.

Although further demand stimulus will not create many jobs, the BOJ should continue to boost the inflation rate, until they reach their 2% target (or better yet a 2% or 3% NGDP growth target.) This will reduce the burden of Japan's huge public debt. Normally the benefit of higher inflation is offset by a higher nominal interest cost. However Japan is at the zero bound, so they can raise inflation without raising nominal interest rates. Indeed they have already done so, and it worked. They also brought the unemployment rate down to 3.4%. With more demand stimulus it might fall a bit further, but not much.

Mission creep doesn't just occur in Japan, many British Keynesians suggested that a demand shortfall explained Britain's low growth a few years back. But the data showed exactly the opposite---jobs were growing, but productivity was declining. Demand stimulus does not address productivity problems.

In the US, demand proponents have occasionally argued that demand stimulus can boost the labor force participation ratio. It's conceivable that it could have a very small effect, but I've been skeptical all along. And now as the unemployment rate approaches the natural rate, it's looking less and less likely that there is a hidden pool of workers just waiting to jump back in when jobs become available. If there were, they would be looking for jobs right now.

To summarize:

1. Where there is lots of involuntary unemployment, demand stimulus might help.

2. If you are at the zero bound and face a large public debt problem, demand stimulus might help.

Otherwise focus on supply-side reforms, meaning privatization, deregulation, tax simplification, and lower MTRs on capital income.

Comments and Sharing

COMMENTS (14 to date)
Roger McKinney writes:
This will reduce the burden of Japan's huge public debt.

Which would impoverish bond holders. Why not be honest and just raise taxes the retired, pension funds and insurance companies that bought the government bonds? It's the same thing.

Andrew_FL writes:

@Roger McKinney-If someone takes on debt and there is an unexpected deflation that increases the burden of the debt, surely it is not "impoverishing the bond holders" to restore the inflation rate to what they already accounted for in the inflation premium? It's just reversing an earlier redistribution to the bond holders, back to the debtors.

Kevin Erdmann writes:

Scott, I hope you don't mind my constant updates on this, but I am now up to Post #15 on the housing series, and #15 shows how inflation targeting was creating monetary errors as early as 2005.

Rajat writes:

And yet wages growth remains slow. Perhaps some workers are moving directly from out of the labour market into jobs?

Kevin Erdmann writes:

Why does everyone keep saying that wage growth is slow?

Rajat writes:

Kevin, isn't it? Seems to be still running at less than 0.2% m/m, down from about 0.3% m/m pre mid-2008.

Jesse writes:

Definitely guilty of this myself. With that said...

I think there is a pretty solid argument that there is significant involuntary unemployment in the US. There is an unusual amount of people that went on disability during this period. Also, the prime age work force participation has fallen. Some of that can be attributed to increases in school enrollment, but not nearly enough. I have to believe that this is not supply side in nature. As for wages, the argument is that real output per hour has increased during this crisis while real wages has been flat. Stocks have recovered well and corporate profits are strong.

It's a hard line to draw to determine whether or not these issues are structural or demand-driven; it's probably a bit of both. It is easier to try to boost the demand side than it is to create structural reform; so why not try to feather out the supply side vs demand side argument while pursuing a pro-growth strategy, especially in light of weak inflation and an economy way below its long-term trend NGDP?

ThomasH writes:

I think you failed to mention another "supply side" factor which looks superficially like demand management: public investment. Governments should be investing in activities that have future benefits and present costs when the NPV of such activities in greater than the LT borrowing rate. Since this decreases during recessions, such investments should increase.

Ray Lopez writes:

Great post by Sumner on why demand stimulus does not always help.

I'm holding my breath and hoping Sumner also concedes money is largely neutral, the Fed follows the market (and when it doesn't, it doesn't usually matter), and monetarism, a close cousin of Keynesian-ism, is largely a shell game. Here's hoping.

Kevin Erdmann writes:


Nominal wage growth is slow. Real wage growth is pretty strong, and has been since the recession, relative to where we might have expected it to be, given unemployment.

Jacob writes:

Scott, I don't really understand the zero lower bound implies that the BOJ can raise inflation without raising interest rates? Doesn't the fisher equation claim that nominal rates should rise in correspondence?

Scott Sumner writes:

Roger, With all due respect I don't think we need to worry about Japanese bond holders. They've received a huge undeserved windfall---if they give a bit back due to sound macro policy, that's not worth worrying about.

Thanks Kevin.

Rajat, Possibly.

Jesse, We are better off aiming for stable NGDP growth. If you are right then the jobs will appear.

Thomas, You said:

"Since this decreases during recessions, such investments should increase."

That doesn't follow at all. It depends why rates fell during the recession. Maybe they fell because there are fewer good investment opportunities (public and private), perhaps due to slower population growth.

Jacob, No, the Fisher equation is an identity. It makes no claims about causation. In fact the proof is in the pudding---the Japanese succeeded in raising inflation, and interest rates have not risen at all, even long term rates. That slightly reduced their debt burden. They should do it again.

ThomasH writes:


I could have been clearer, I was talking about a recession such as the US and Europe had after 2008. And monetary policy reduced short tern and later longer term interest rates (not as much as would have been required to get the price level/NGDPL back on track, but lower. It's pretty unlikely that the future value of new transportation, water and sewerage infrastructure and upkeep of the same simultaneously fell, so an increase in such investments seems likely to be welfare improving. (It is also possible that recessions also create a gap between the market price of project inputs and their opportunity cost, which will also raise NPVs.)

But the fundamental supply side policy is to actually apply the NPV rule, recession or no recession, and not REDUCE investments because of deficit fears, what I call "austerity." That's basically the Law and the Prophets of Keynesian economic insight.

Michael Byrnes writes:

Can we add a little corollary to this post:

"In a situation where demand stimulus is unlikely to help (ie supply side reforms are needed), monetary tightening is not the answer - that just adds a demand-side problem to a supply-side problem."

Comments for this entry have been closed
Return to top