Thursday, September 17, 2015

Washington: We have a problem

John Cochrane makes the case in the WSJ that everything is back to normal. Hunky Dory, rosy tinted, don’t panic, keep-calm-and-carry-on normal. He points out that inflation is under control. We have not entered a deflationary death-spiral and unemployment is back in reasonable territory.

Here is what John learned from the Great Recession.
The [QE] experiment was huge, and the lessons are clear. The economy is stable, not subject to Keynesian “spirals” requiring constant Fed intervention. And when reserves pay the same rate as bonds, banks do not care which one they hold. So even massive bond purchases do not cause inflation. Quantitative easing is like trading a $20 bill for $10 and $5 bills. How would that make anyone spend more money?
John sees the world through the lens of a model where QE can’t matter. Doubling down on this view…
As then Fed Chairman Ben Bernanke said in January 2014: “The problem with QE is that it works in practice, but it doesn’t work in theory.” That’s a big problem. If we have no theory why something works, then maybe it doesn’t really work.
My emphasis.

I'm not sure we’re living on the same planet. In fact, I know we’re not living on the same planet. John is on planet Chicago. I'm on planet UCLA.

On my planet, QE has some pretty big effects. By increasing the size of the monetary base, the Fed averted an even bigger deflation than the one that occurred.

Here’s the evidence for that. Figure 1 shows how the Federal Reserve Board responded to the financial crisis. The solid line, measured in percent per year on the right-hand axis, measures the expected rate of inflation. The boundary of the shaded region, measured on the left-hand axis in millions of dollars, is the size of the Federal Reserve’s balance sheet.

Figure 1
From January of 2007, through September of 2008, expected inflation fluctuated between two percent and three and a half percent. When Lehman Brothers declared bankruptcy in September 2008, expected inflation fell by nearly eight hundred basis points in the space of two months and by October of 2008 it reached a low of negative four and half percent.

Immediately following the Federal Reserve purchase of one point three trillion dollars of new securities, expected inflation went back up into positive territory.

The Fed’s actions did not completely prevent deflation, and the CPI inflation rate fell to negative two percent in July of 2009. But the Fed’s actions did turn around inflationary expectations and it is likely that QE prevented a much larger deflation that would have had catastrophic effects on unemployment, had it been allowed to occur.

What about the composition of the balance sheet? By intervening in the MBS market, the Fed turned around a stock market crash and held the unemployment rate at 10%. Pretty bad, but not the 25% of 1933. 

Figure 2
Figure 2 contains the same information on asset purchases as Figure 1 but instead of plotting expected inflation on this chart, the solid line is the value of the stock market. I want to use this chart to make a point about the effects on markets of the type of assets that central banks buy.

Figure 2 shows that the turn around in the stock market that occurred at the beginning of 2009 coincides closely with the Fed’s intervention in the MBS market. Further, when asset buying was suspended temporarily, in the second quarter of 2010, the stock market resumed its downward spiral, picking up again only when the Federal Reserve announced at the Jackson Hole conference in the autumn of the same year, that large-scale asset purchases would resume.

Here’s a link that explains how QE works. And here is a link that presents the evidence for a causal connection from the stock market to the real economy. John is aware of my argument that financial instability is Pareto inefficient. He either disagrees or chooses to ignore it. I'm not sure which.

So, is there a case for raising interest rates and getting ‘back to normal’. John thinks not
For interest rates, the Fed has set itself a nearly impossible task. Fed officials need to know what the correct, or “natural,” real rate of interest is. Is there a “savings glut” or another recession on its way, driving the correct real interest rate down? Or are tight markets for skilled workers and large corporate profits a sign of high “natural” rates? Setting the right price of tomatoes is hard enough, let alone divining the right real interest rate for an entire economy.
John is wrong to think that we do not have a problem when rates are at zero. And he is wrong for two reasons.

First. Even if we accept that the Fed has no business trying to influence the real economy it does have a responsibility to control inflation. And the lever that controls inflation is the money interest rate. That lever did a pretty effective job for thirty-five years. Right now, it’s set at full speed ahead with no room to maneuver.

Second. The Fed does influence the real economy; not just in the short run: But in the long run. Financial markets do not allocate capital efficiently and the Fed has an important role as lender of last resort. We left markets to themselves in the nineteenth century and that didn’t turn out too well.  Lets not forget that lesson.

So what should we do? Raise rates. And raise them soon. But raising rates is not a get out of jail free card. A higher nominal rate will drive the economy back into recession by triggering a fall in the stock market.  

Financial wealth goes up. Financial wealth goes down. And with it - so goes the real economy. Movements in financial wealth do not reflect future booms or busts. They cause them. The lesson from Figure 2, is that the Fed can, and should, stabilize asset markets by actively trading the risk composition of its portfolio.  So - yes: raise rates. But; at the same time, absorb the risk that the private sector is unwilling to bear by trading equities.

Perhaps that sounds too radical? Intervening in the asset markets in any capacity was a radical proposal when the Fed was created in 1913. If you think the last crisis was bad: Wait and see what the next one will bring.


  1. "Movements in financial wealth do not reflect future booms or busts. They cause them."

    That is correct. Unfortunately most economists do not seem to have identified the mechanism that leads from one to the other. In the Great Depression and the Great Recession the mechanism was through the effect on consumer expenditure which in turn caused a fall in aggregate demand. In Japan it was through the effect on company finances and thus a fall in investment demand. That was the first stage. The next stage was through the effect on banks (and shadow banks) and total lending.

    My book "Redefining Macroeconomics" has the mathematical details and why Keynes was wrong on so many points, though the Keynesian method is basically right. The book also explains many other things: e.g. why recoveries that follow financial asset market crashes are of necessity prolonged, why keeping interest rates low does not help and indeed leads to the next crash, etc

    1. Thanks for your comment Philip. I will take a look at your book.

  2. Roger, you might be interested in this post inspired by your post here.

  3. " does have a responsibility to control inflation. And the lever that controls inflation is the money interest rate. That lever did a pretty effective job for thirty-five years. Right now, it’s set at full speed ahead with no room to maneuver."

    Roger, just to be clear, in the above your worry with controlling inflation must be on the downside, correct? Because there's plenty of room to maneuver on the upside for rates (i.e. putting the brakes on inflation) and we're currently undershooting the 2% inflation target.

    But if you're arguing that QE is effective (in this age of ZIRP) then there's room to maneuver outside of the rate lever (equivalent to room on the downside for rates). What am I missing?

    1. Tom
      Your graph is interesting. And suggestive.

      Yes. The current problem is deflation. More generally; it is that interest rate control is the lever we use to control inflation. Both on the upside and on the downside. That lever has been effective. But it has been directed at two targets. Inflation and unemployment'.

      For thirty years, the right way to address both targets had coincided. That is no longer the case. We need a second instrument. That instrument is control of the risk composition of the Fed balance sheet.

    2. Roger, thanks for the clarification. You write:

      "Your graph is interesting. And suggestive."

      If you're referring to any of the graphs on the post I linked to, those aren't mine: those were created by Jason Smith, who writes that blog.

  4. Roger, you stated that we should raise the rates and raise them soon, but that will lead the economy into recession. What, if anything, can the government do to prevent the economy from going into a recession, and instead reach prosperity?

  5. Prevent a stock market drop by buying risky assets.


Note: Only a member of this blog may post a comment.