
Yesterday I stumbled upon a strange sight, a couple of black swans swimming above some eels in a New Zealand lake. The swans would occasionally lift a leg out of the water, perhaps to avoid getting bitten by the eels. (At least that’s what the locals said.) That got me thinking about the financial crisis of 2008.
Most observers of this scene probably have more sympathy for the noble and elegant swans than the dark and slithery eels. When looked at objectively, however, both are rather primitive creatures with small brains, which act instinctively. We project the nobility and the deviousness. And I believe the same is true of financial crises.
When people discuss Fed policy during a financial crisis, there is often an implicit framing that portrays the central bank as a hero and the economy as a villain. Large injections of liquidity are seen as a sort of “rescue”, in response to sharply elevated demand for reserves during a financial crisis. The economy is seen as wild and out of control, something that needs to be stabilized by the central bank. This framing isn’t just wrong, it leads to bad consequences.
Let’s back up and think for a moment about the nature of fiat money. In the US and in most other countries, the central bank has been given a monopoly on the production of fiat money. Along with that monopoly, the central bank has been assigned the responsibility to stabilize the value of money according to some metric.
In the past, governments often told the central bank to make sure the value of fiat money was stable in terms of gold. More recently, they’ve been assigned the duty of stabilizing the value on money in terms of an index of goods and services prices. I favor stabilizing the value of money as a share of NGDP. Monetarists such as Milton Friedman wanted to the central bank to stabilize the quantity of M1 or M2.
Almost no one favors a policy of stabilizing the monetary base. Almost everyone believes it is the duty of the central bank to adjust the monetary base in such a way as to stabilize some other variable, whether it be gold prices, the CPI, NGDP or M2. Thus, it’s the job of the central bank to adjust the monetary base to accommodate changes in the demand for base money.
Each November, the turkey industry boosts the supply of turkeys to meet the public’s demand for those tasteless birds around Thanksgiving. We don’t think of the turkey industry as valiantly rescuing a public with a recklessly unstable demand for turkeys. Rather, the turkey industry is just doing its job.
If the Fed did not meet an increased demand for reserves with injections of new base money, then they would not be able to stabilize gold prices, the CPI, NGDP, M2, or any other plausible goal of monetary policy. When they are doing their job, a monetary injection is not some sort of heroic action that bails out an evil and irresponsible financial system, rather it is simply a way of meeting the demand for reserves, in order to stabilize the central bank’s goal variable. That’s equally true of an increased demand for reserves that originates in the banking system, and an increased demand for cash that results from people in Venezuela hoarding US dollars. But for some reason, many people view these two situations quite differently. The latter case is seen as normal policy, while the former is viewed as extraordinary.
When people ask me if the government should bail out the financial system in a crisis, I reply that the answer is no. Instead they should supply enough reserves to keep NGDP growing at around 4% a year. It just so happens that if they did so there would never (by definition!) be the sort of financial crisis that reduced aggregate demand. (There might be other sorts of financial crises.) In addition to not bailing out banks, the Fed should not try to reduce unemployment, or inflation, or pop asset bubbles, or undertake any other objective. Just keep NGDP growing at about 4%.
The hero/villain framing is unfortunate as it leads to bad monetary policy. With this framework, people expect too little of the central bank, as I explained in this recent post. If we expect too little of the central bank, then they are more likely to take their eye off the ball and pursue tangential objectives. They will fail to stabilize aggregate demand. Thus in 2008, the Fed did not see itself as causing a fall in NGDP with a tight money policy, and that failure of imagination contributed to their unwillingness to do “whatever it takes” to hit their policy targets.
Recessions in the US are not caused by private sector “shocks”, they are caused by bad monetary policy. As we expect more of monetary policy, these imaginary “black swans” will become much less frequent. Ironically, in Australia (home of the black swan) they already are quite infrequent.
PS. People on the right often unwittingly play into the hands of statists when they complain that reserve injections during a financial crisis are some sort of “bailout” or “rescue” of the banking system. That framing makes the free enterprise system seem feeble, and in need of government help.
READER COMMENTS
Brian Donohue
Jan 13 2020 at 2:42pm
Excellent post.
Lliam Munro
Jan 13 2020 at 5:06pm
Hope you’re enjoying New Zealand, Scott. Will you be in Wellington at all?
Scott Sumner
Jan 14 2020 at 3:45pm
I was just there for one day, but greatly enjoyed the city.
Lliam Munro
Jan 14 2020 at 8:48pm
Would have loved to have met you for a coffee. Another time, perhaps.
Scott Sumner
Jan 15 2020 at 4:06am
Let me know if you ever get to Orange County.
Alan Goldhammer
Jan 13 2020 at 5:32pm
This is somewhat off topic. There was an article in either the NY Times or WaPo this morning noting that more money has been spent on bailing out farmers for losses from the ill-advised tariffs against China than was spent in total on the 2008 bail out.
To Scott’s main point, we cannot know what would have happened if the banks along with AIG were not bailed out or if the takeover of the shadow banks such as the mortgage originators who were also virtually bankrupt. This is an experiment that I’m thankful did not happen. If it did, the recession could have been much worse.
Scott Sumner
Jan 13 2020 at 11:45pm
The relevant counterfactual is no bailout combined with NGDP level targeting. That is, no drop in aggregate demand.
Thaomas
Jan 14 2020 at 5:38am
Whether or not the 2008 crisis was the result of the fed failing to keep the PL going and employment full before the crisis, that cannot excess the less that “what it takes” to restore the PL to trend and return employment to something like full employment . (We now suspect that 4% of U6 is not “full employment,” but THAT’s an excusable error; we have new information.)
Michael Rulle
Jan 14 2020 at 12:31pm
Powell’s reaction 13 months ago was extraordinary—even if the result of a gradual movement toward NGDP style thinking.
Question—but first a backdrop
I was always against bailouts,including in 2008, but had no clue at all about NGDP targeting or money demand per se. The only analogies I could muster were:
1) 1907. I believed, from reading the Bruner and Carr book—-that banks were “cajoled” or “forced” or “threatened” by Morgan and cohorts to provide liquidity to each other—-which ended the crisis within a year of when it began. I thought that was a good model to follow
2) 1998. I thought Greenspan’s actions were very similar to Morgan’s in 1907. (after the fact of course—I had not known anything about “1907” in 1998) At the time, many were saying that Greenspan “bailed” the banks out from the Russian/LTCM crisis and should have let them figure it out.
I did not think Greenspan bailed them out. I thought he did the opposite. I thought he threatened them with a promise of not bailing them out and thus “coerced” them into providing liquidity to each other. That too was a crisis which ended quickly.
My question is—–do either of these two examples have similarities with—say Powell in 2018 (recognizing I am hypothesizing he prevented a crisis rather than fix one)
Scott Sumner
Jan 14 2020 at 3:48pm
I don’t see a lot of similarity, except that both cases avoided bailouts.
Bob Murphy
Jan 15 2020 at 1:14pm
Scott Sumner wrote: “People on the right often unwittingly play into the hands of statists when they complain that reserve injections during a financial crisis are some sort of “bailout” or “rescue” of the banking system. That framing makes the free enterprise system seem feeble, and in need of government help.”
Scott, I am splitting hairs perhaps but I disagree with the statement above. If it’s someone COMPLAINING about a bailout or rescue, that suggests that the critic doesn’t think it was a good policy. So surely the critic from the right isn’t saying, “I agree the banking system would’ve collapsed in 2008 without TARP and the Fed’s new policies, but I oppose a bailout so bring on the collapse!”
In any event, the Fed didn’t merely expand the monetary base in late 2008 forward. They were specifically buying MBS, not haircuts sold by single mothers in the inner city. And of course the institutions allowed to borrow at the new Lending Facilities etc. weren’t mom & pop retailers.
Scott Sumner
Jan 16 2020 at 4:28am
Some on the right worry that bailouts create moral hazard. They would prefer that the government allow banks to collapse in order to purge the excesses from the system.
I’ve seen people bemoan the fact that QE merely postponed the day of reckoning.
Roger McKinney
Jan 16 2020 at 6:47pm
Decades of bailouts in Japan have created armies of zombie companies that eat most new money just to make payments on old debt and rendering monetary policy impotent. The US is now where Japan was 30years ago.
tpeach
Jan 15 2020 at 5:52pm
Swans and eels…… does anything not remind you of monetary policy lol?
Are you coming to Australia while you are on this side of the world?
Scott Sumner
Jan 16 2020 at 4:25am
Not this trip.
IVV
Jan 16 2020 at 11:13am
…Isn’t that what Bitcoin does?
(Also, turkey isn’t tasteless if you know what you’re doing with it. Although I suppose the white cage-farmed birds aren’t the best examples of turkey.)
Scott Sumner
Jan 17 2020 at 3:05pm
Bitcoin isn’t money, at least in the sense of “monetary policy”.
IVV
Jan 17 2020 at 3:51pm
True enough. One would think that the way it’s pitched is misleading, then.
bill
Jan 16 2020 at 4:45pm
Excellent post. One could say that the Fed was both the arsonist and fire department in 2008. The evidence that makes sense to me is that the foreclosure crisis really hit once unemployment rose. If the Fed had simply hit a steady 4% NGDP target, there would have been no foreclosure crisis, thus no banking crisis, and so on. Even if somehow a recession (negative RGDP growth) was going to happen in spite of 4% NGDP growth, it’s easy to see that housing prices would have taken a different path if inflation had been in excess of 4% (if the sum is 4%, it’s just math). And if housing prices had held up, then there are fewer foreclosures at any given unemployment rate. Also, back then, I think we tended to speak of 5% as the target for NGDPLT which would have made any crisis even smaller.
nobody.really
Jan 16 2020 at 6:00pm
Please break this down for the layman: What SPECIFIC action should the Fed have taken to achieve a 4% NGDP growth rate? And what action did the Fed take instead?
Here’s my (limited) understanding:
To achieve greater NGDP growth, the Fed would buy back bonds, thereby flooding the market with liquidity and stimulating inflation. As the value of cash fell, firms would have been eager to trade them in for assets–even, perhaps, troubled assets. Problem solved.
Instead, the Fed bought some bonds, but also bought up–well, everything else, which we call Quantitative Easing. Is that so different?
Scott Sumner
Jan 17 2020 at 3:08pm
The most important specific action would be setting a level target for NGDP. After that, adjust the monetary base up or down until expectations of NGDP growth are on target. The specific actions are open market purchases and sales.
bill
Jan 16 2020 at 8:03pm
@nobody really,
Well, to start, instead of raising IOR, which had been 0% for 95 years, they could have cut it to negative 5%. They could have announced a 5% NGDPLT target. Level targeting is more powerful than rate targeting – most people agree on that. Look at the first QE program and see that it could have been designed more like QE3. Improve that QE type from $85 billion per month to one that would have grown exponentially (the monthly amount could increase by 50% every month, for instance) until we were back at the 5% growth level target. I have more ideas.
Comments are closed.