Scott Sumner  

The old rules still apply (What the rest of the profession could learn from Ben Bernanke)

PRINT
Blame the Republicans... Milton and Rose Friedman on Hi...

During the days of William Jennings Bryan it was pretty well understood that deflationary monetary policies helped bondholders and inflationary monetary policies helped debtors. That's why the rich favored the gold standard and lots of indebted farmers and small merchants favored a bimetallic system (which would have been slightly more inflationary.)

Mark Thoma links to a post by Atif Mian and Amir Sufi, which seems to suggest that the old rules no longer apply:

The Federal Reserve may help in boosting the net worth position of households. But does it boost household net worth where it is needed the most? Unfortunately, quite the opposite is true. The Fed directly controls short term interest rates, and hence has the strongest and quickest influence on bond prices. Bond prices are inversely related to interest rates because lower future interest rates make the future coupon payments paid by existing bonds worth more today. The value of long-term bonds can increase substantially if the Fed can lower expectations of future interest rates.

The chart below shows that at the outset of the Great Recession, both house prices and stock prices tanked. As we have done in the past, we index the values to be equal to 100 in 2006, and so the percentage change from 2006 to any year is just the point for that year minus 100.

While stock prices later recovered, house prices remained depressed for an extended period. However, the one asset that did remarkably well was long term bonds. Those holding long term bonds profited handsomely from the decline in interest rates.

[graph here]

Unfortunately for the macro-economy, the gains in long-term bonds were a unique benefit to creditors. Debtors with a levered claim on house prices remained stuck. This was one of the great limitations of how effective the Federal Reserve could be in the midst of the Great Recession.

Many have placed much blame on the Federal Reserve for increasing wealth inequality. That is unfair -- it is not the Fed's fault that only the very rich hold bonds and other financial assets. But it is true that a by-product of looser monetary policy is a rise in wealth inequality-the Fed was unable during the Great Recession to boost the net worth of debtors.


This is a common view--the Fed's easy money policy helped wealthy bondholders. But is it true, or is the 19th century view more accurate?

The modern view is based on people looking to either interest rates or the money supply as an indicator of the stance of monetary policy. And by those criteria policy was indeed quite loose. But in 2003 Ben Bernanke said neither criterion was reliable, and investors instead needed to look at NGDP growth and inflation. If one averages Bernanke's two criteria, then monetary policy in the 5 years after mid-2008 was the tightest since Herbert Hoover was president. So there is no mystery that needs to be explained. The reason easy money seemed to defy orthodox economics and help creditors is that monetary policy after 2008 was in fact not easy. Indeed it was highly contractionary.

People sometimes roll their eyes at my endlessly repetitive harping on how monetary policy was not expansionary after 2008. They say it doesn't really matter what you call it, the Fed did what it did. But the Mian and Sufi post shows it does matter. They draw conclusions about the effect of loose monetary policy on income inequality that are quite misleading. And as an aside this has nothing to do with me being a "conservative" who "doesn't care about inequality." Matt Yglesias is quite progressive, and has frequently complained that the Fed's policy was too tight and was hurting the working class by keeping the job market slack.

Faster NGDP growth (easier money) after 2008 would have helped borrowers. The old rules still apply.

Economists need to take Bernanke's 2003 claim more seriously. And that includes Ben Bernanke himself, who no longer says what he said ten years ago. When an economist has been both an academic and a policymaker, trust what they say when they have their academic hat on. That's when they are free to say what they really believe.

HT: TravisV

Off topic: I'd like to recommend this FT piece by Tim Harford. I believe that an NGDP futures market would provide the sort of real time information that he calls for.


Comments and Sharing


CATEGORIES: Monetary Policy



COMMENTS (18 to date)
Jon writes:

I think you mean unexpected inflation favors debtors and unexpected deflation favors bond holders.

So while the WJB gold standard vs bimetallic debate might have capture the zeitgeist, both sides were quite wrong then and now.

Scott Sumner writes:

Jon, Good point, but regarding 30 year bonds the "unexpected" refers to inflation that was not anticipated 30 years earlier. So the "short run" effects of a policy switch can last for a surprisingly long period.

Andrew_FL writes:
The modern view is based on people looking to either interest rates or the money supply as an indicator of the stance of monetary policy. And by those criteria policy was indeed quite loose. But in 2003 Ben Bernanke said neither criterion was reliable, and investors instead needed to look at NGDP growth and inflation. If one averages Bernanke's two criteria, then monetary policy in the 5 years after mid-2008 was the tightest since Herbert Hoover was president.

You know, this bothers me. Because it reminds me of something:

go to about 14:10

Walter Heller: "What has happened in these three years, actually, from 75 yo 78, was that we still had a policy a great deal of slack in it. A lot of unemployment a lot of unused capacity-"

Marina Whitman: "You mean an economy with a lot of slack in it."

Walter Heller: "Yeah, that's right. And uh, inflation got stuck. It got stuck right around five percent level-"

We are talking about Market Monetarism here, right?

dannyb2b writes:

"Faster NGDP growth (easier money) after 2008 would have helped borrowers"

Only if they borrowed more first. If debtors dont increase their debt the broad money supply doesnt increase and neither does inflation or NGDP.

Since the 50's in the US there is rarely any growth without increase of total debt.

W. Peden writes:

"If debtors dont increase their debt the broad money supply doesnt increase and neither does inflation or NGDP."

That all rests on four contingent ceteris paribus assumptions-

(1) No variation in the demand for broad money.

(2) No variation in the LRAS curve.

(3) No change in the money supply due to the balance of payments.

(4) No shift of assets into broad money.

dannyb2b writes:

Yes ceteris paribus.

dannyb2b writes:

4) What do you mean shift of assets into broad money?

W. Peden writes:

Dannyb2b,

Start with an example of an intra-broad money shift: imagine government restrictions on interest rates on transaction accounts are abolished. Then, the opportunity cost of holding M1 falls, and savers will shift their funds into demand deposits.

Similarly, imagine the opportunity cost of holding all deposits relative to near-substitutes for deposits (short-term bonds, commercial paper, bankers' acceptances, repos etc.) falls. Savers will shift their funds into broad money.

Because the relationship between credit/debt on the one hand and NGDP on the other is so subject to so many ceteris paribus assumptions, it's almost always better to focus on bank LIABILITIES rather than assets. (Except the monetary base, of course.)

dannyb2b writes:

In my post I was assuming just convention monetary policy.

What would extra demand for demand deposits affect?

Does the fed directly control these?

(1) variation in the demand for broad money.

(2) variation in the LRAS curve.

marcus nunes writes:

[Comment removed for policy violations.--Econlib Ed.]

Scott Sumner writes:

Andrew, Heller was wrong, the economy didn't have slack in 1978. Not sure why you call that market monetarism.

danny, Correlation doesn't prove causation. You can have more NGDP without more debt, just as you can have more NGDP without more restaurant meals. But you rarely see either occur. The correlation doesn't prove that either debt or restaurant meals cause NGDP.

W. Peden writes:

dannyb2b,

My point is that this is a case of probable inference, rather than an accounting identity or a matter of mechanistic necessity. An increase in domestic debt (or even the existence of a credit/debt market) is not a necessary condition of an increase in NGDP.

dannyb2b writes:

I understand correlation doesn't always mean causation. The reality is that ngdp is influenced (to varying degrees) by lending because most deposits come into effect from loans. Thats one of the main purposes of adjusting the fed funds rate.

marcus nunes writes:

Scott Sumner writes: “Faster NGDP growth (easier money) after 2008 would have helped borrowers. The old rules still apply.”
The link below indicates that using the level of interest rates (or even inflation) to gauge the stance of monetary policy can be very misleading.
http://thefaintofheart.wordpress.com/2014/03/01/identifying-the-stance-of-monetary-policy/

steve writes:

You, not surprisingly, focus on the loose vs tight money issue. I think you are correct that money was tight. However, you seem to be missing the broader point. Bondholders came out ahead. If you look at the results of Fed policy, not what they say they are doing or think they are doing, it seems to nearly always benefit the bond holding class, i.e., the creditors. Why does the Fed preferentially meet the needs of creditors over debtors?

Steve

Mark A. Sadowski writes:

dannyb2b,
"The reality is that ngdp is influenced (to varying degrees) by lending because most deposits come into effect from loans."

The velocity of broad money (which includes deposits) is highly correlated to bond yields:

http://research.stlouisfed.org/fred2/graph/?graph_id=87718&category_id=0

And bond yields are highly correlated to NGDP expectations:

http://3.bp.blogspot.com/-fQOh2jjNS_8/T2kdMyGHxAI/AAAAAAAACXI/y5k6QONVFGM/s1600/treasyield_NGDP.jpg

So, all other things being equal, this implies that expansionary monetary policy should lower the ratio of outstanding commercial bank loans to GDP.

Andrew_FL writes:

@Scott Sumner-I'm sorry, I should have been more clear and less...gotchay.

To be clear, I am accusing you of identifying slack policy by a slack economy. I believe this to be philosophically troubling. It's saying, you know policy was not sufficiently expansionary, because if it was the economy should have grow faster or such and such. But that is coming from a position of assuming your theory is true at the outset. It is a line of thinking I associate with Keynesianism and I am kind of cheekily suggesting it is odd, and somewhat unsettling, to see it coming from a Monetarist.

Isn't there some way to judge the tightness or looseness of policy objectively?

dannyb2b writes:

"So, all other things being equal, this implies that expansionary monetary policy should lower the ratio of outstanding commercial bank loans to GDP. "

Doesnt seem like it in the data.

Comments for this entry have been closed
Return to top