Sunday, April 24, 2011

Dissatisfaction with QE2

This Sunday New York Times piece captures some disquiet about what has been, or has not been, accomplished as the result of the Fed's QE2 asset purchases. The piece illustrates all of the problems with the policy that one might have anticipated at the outset.

1. Central bankers should not claim credit for things they cannot control. Bernanke told us here how QE2 would work:
Stock prices rose and long-term interest rates fell when investors began to anticipate the most recent action. Easier financial conditions will promote economic growth.
So, where is the economic growth we were promised? The employment/population ratio is still in the toilet, and by all reports the upcoming first-quarter 2011 GDP numbers will be weak. For anyone paying attention, Bernanke has now lost credibility.

2. Economists who might have supported your policies will run the other way when the going gets tough. When QE2 was announced, Mark Thoma seemed to like it, but had some quibbles. He thought it wasn't big enough, and that the Fed should be purchasing longer-maturity Treasuries. The New York Times now quotes him as saying:
It’s good for stopping the fall, but for actually turning things around and driving the recovery, I just don’t think monetary policy has that power

3. The Fed is seen as screwing up, so everyone wants to butt in with their own stupid suggestions. Here are some examples. The first is the suggestion, again, that it's not working because the Fed is not doing enough, i.e. it should buy all of the available Treasury debt:
The Fed limited the program to $600 billion under considerable political pressure. While that sounds like a lot of money, the purchases have not even kept pace with the government’s issuance of new debt, so in a sense the effort has amounted to treading water.
So, quit treading water, and swim across the Pacific Ocean. Here's another wonderful idea:
a growing body of research suggests that the Fed could have had a larger impact by spending more money on a broader range of debt, like mortgage bonds, as it did initially.
So, quit buying Treasury bonds, and go back to purchasing mortgage-backed securities (MBS), or some other private assets. The QE1 program (purchases of MBS and agency securities of more than $1.2 trillion) set a dangerous precedent. Clearly, if you do this once, you can do it again, on demand. Programs like this can be used to bail out sectors of the economy or individual firms, and using them threatens the Fed's independence.

Charles Plosser certainly comes off well. The guy has good sense, and integrity:
“I wasn’t a big fan of it in the first place,” said Charles I. Plosser, president of the Federal Reserve Bank of Philadelphia and one of the 10 members of the Fed’s policy-making board. “I didn’t think it was going to have much of an impact, and it complicated the exit strategy. And what we’ve seen has not changed my mind.”


  1. Stephen,

    There seems to be a dis-connect between market views of QE success and those of many observers. In other words, asset prices seem to be discounting much stronger nominal growth than these "QE was unsuccessful" comments imply. There are a number of potential explanations: 1) asset markets are right about QE; 2) they are discounting a strong recovery driven by non-QE factors; and 3) asset markets discount the Fed's ability and willingness to raise asset prices by encouraging speculation.

    One hypothesis related to 3) above is the following: QE/ZIRP has little impact on the real economy, but an out-sized impact on market psychology. One might attribute the housing bubble, partially, to this phenomenon (i.e. in that the Fed produced rising house prices without a commensurate increase in rents). This implies that the rise in asset prices might be unsustainable, and might therefore lead to a sharp correction.

    The above is not quite pointless speculation. Monetary policy now proceeds mostly through the financial speculation channel. To the extent that higher asset prices do not produce sustainable economic effects (such as higher business investment and hiring), monetary policy is bound to produce unsustainable asset price "boomlets" followed by deflationary busts, followed by fresh attempts to inflate asset prices.

  2. If you are trying to suggest I've been inconsistent, then you need to read more of what I've written. I've been making the same argument for years, and have done so in many, many posts.

  3. Hi Mark,

    No, you'll notice I didn't actually accuse you of anything outright. On the face of it, it looked inconsistent, but maybe you were quoted out of context, or there is some line of reasoning that takes you from November 4 to the quote. I have not been reading you for years, so I don't know your views on this. What should the Fed do now? More QE, sell some assets, raise the interest rate on reserves, hold steady for a while?

  4. anon1:

    If you want to dissect movements in the stock market over a period of a few months, attribute the causes, and work out the effects, good luck to you. More generally, on the effects of QE2, I think we are still looking for a good theory of liquidity, the maturity structure of the consolidated government debt, and how that relates to the term structure of interest rates and asset prices. Without that, we cannot understand QE2 or quantify its effects.

  5. Perhaps you can education me a little bit since I'm still trying to grasp this stuff. But shouldn't an action like QE2 have been expected to have small effects? Currently the Fed is paying interest-on-reserves and we have a huge amount of excess reserves, so aren't you just swapping interest-bearing assets around without changing much of anything? I guess you have the difference in maturities, but I can't imagine that the effects of this will be that significant. I'm just confused why QE2 would be expected to be that effective. Does the Fed really think that whatever hypothesized yield channel is that significant for aggregate economic activity? Also, doesn't this require some kind of market segmentation anyway? Perhaps I just embarrassed myself and there is some obvious reason why QE2 was expected to work. But, at a minimum, it looks to me there was no particularly compelling theoretical reason as to why QE2 would have such large, positive effects and I'm not a huge fan of policy makers running a massive experiment that nobody really have a good idea what the outcome will be.

  6. Ted,

    Exactly. Under current circumstances, QE2 is essentially just a swap of overnight interest-bearing government liabilities for long-maturity interest-bearing government liabilities. Why should that do anything? Only if the Fed is somehow better than the private sector at this brand of financial intermediation. But many private financial intermediaries are perfectly capable of acquiring long-maturity Treasury bonds, and financing that portfolio by rolling over overnight repos, which is essentially what the Fed is doing. But borrowing short and lending long is a risky activity, so maybe the Fed unloaded some risk from the private sector, and that matters? However, the risk has to go somewhere else in the private sector, ultimately. I think the working hypothesis is that QE2 should be neutral, except for the fact that the reserves issued by the Fed to finance the purchases can be withdrawn as currency. And no private sector entity can issue currency. The Fed can of course prevent that from happening by raising the interest rate on reserves. Thus, I think the hypothesis is: QE2 is neutral, under the assumption that the Fed manipulates the interest rate on reserves so that financial insitutions hold the reserves issued to finance the asset purchases.

  7. "So, where is the economic growth we were promised?"

    What's your counterfactual, Steve?

    We have good evidence that QE 1 reduced domestic and international interest rates, as well as the value of the dollar.

    The effects of QE 2 are harder to measure because expectations are harder to measure but one would assume that they are the same.

    It seems that the most likely effect of QE 2 was to raise GDP growth by aabiout 1.5 percentage points or so. Too small to be obvious but well worth doing.

  8. 1. "We have good evidence that QE 1 reduced domestic and international interest rates, as well as the value of the dollar."

    What makes you think that evidence is any good?

    2. "It seems that the most likely effect of QE 2 was to raise GDP growth by aabiout 1.5 percentage points or so. Too small to be obvious but well worth doing."

    Where does that come from? What's the theory?

  9. Because you didn't make the false accusation "outright" it's okay?

    It's not my problem that you haven't read enough of what I've said to fully characterize my views ("I don't know your views on this"). Since you admit that's the case, why the false insinuation which is, essentially, continued in your follow up comment?

    Go read the post you linked to again and see if I actually say this will have a big effect. I don't -- I've been consistent in saying that monetary policy alone won't be enough from the very start, and I haven't waivered from that position (and I have the emails from Joe Gagnon telling me I'm wrong to prove it if you won't take my word for it, he has not been pleased with my lackluster support).

    So please stop insinuating that I've been inconsistent when the real problem is you making accusations when, as you admit, you don't actually know my views.

  10. Mark,

    I'm just trying to learn something here. Don't get too bothered about it. On November 4 you say there is some stuff you don't like about QE2. The asset purchases should be larger and the maturity should be longer. Then, the other day, you tell the New York Times reporter that, during a recovery, monetary policy is generally irrelevant. Does that mean that it's irrelevant during the recovery as usually practiced? Does that mean that QE2 is irrelevant? Surely you did not have a view on QE2 from years' back, unless you have wonderful foresight, which of course could be the case. Please explain.

  11. Please quit mischaracterizing me. Stopping a downward fall is far from irrelevant.

    There's nothing special about QE2, it's just open market operations on a different part of the yield curve, so prescience isn't really at issue. The question is about the ability to change real rates (which I think the Fed can do) and the response to it (I hang my hat on the interest elasticity of investment falling in recessions - you can disagree, but the point is that I've been making this argument for decades in my classes, and certainly since prior to the recession).

    I have an entire category called "monetary policy," and there are lots and lots of posts making this point.

    More recently, I've made the point again and again that I think monetary policy can have some effect, though not a large one, and that I am not particularly worried about inflation, so the benefits -- though moderate -- still exceed the costs (hence the call for more aggressive policy, particularly since additional fiscal boosts are out of the picture -- something, however moderate, is better than nothing given my assessment of the costs). That is very far from being an ardent supporter. In any case, you seem more interested in playing gotcha than accurately reflecting my views, so there's not much point in continuing this. But what you've claimed is far from accurate.

  12. And when you make false insinuations in public, and then persist in comments while admitting you don't actually know what I believe, please don't tell me not to get bothered about it.

    What would be nice is if it bothered you to engage in such behavior.

  13. Mark, honestly, I'm not interested in making you look bad. I wanted to figure out what you were thinking, and made this a little provocative to, admittedly, get your attention. You've done a good job of explaining yourself. Thanks.

    1. "I hang my hat on the interest elasticity of investment falling in recessions..." Do you think that has something to do with perceived risk over the business cycle and lending? Some people talk about that, but I can't remember seeing it modeled. This depends on what interest rates you are looking at. Do you mean interest elasticity with respect to the risk-free rate? Spreads tend to increase in the downturn (and of course much more so in the last recession), in part due to default premia.

    2. For me, QE2 is fundamentally different. The usual open market operation is a swap of reserves for T-bills in a world where excess reserves are zero overnight. Now we are swapping reserves for long Treasuries in a world where the system is awash with reserves. I know Bernanke wants us to think it is the same, but I don't think so.

    3. I guess I worry more about the inflationary potential than you do.

  14. What the hell is the matter with Mark Thoma? Seems to becoming as insufferably arrogant as the infamous Brad deLong. He goes off at the deep end when you call him on things he thinks he is an expert on, as when the chartalist school showed up his ignorance about the money multiplier in a modern fiat system.

  15. No, Thoma and DeLong are different breeds of cat. Thoma is basically well-meaning and seems to be providing a useful public service. He's also sensitive. It might have been better for DeLong, and for humanity, if he had gone to Washington State University for his PhD, like Thoma, and not to Harvard. He would probably know more, and would be more humble.

  16. Steve, What makes me think the evidence is any good? [sigh] Why do you think that changes to real asset prices are neutral?

    Over a series of QE events where initial and final expectations were approx zero we observed sudden asset price changes immediately after QE announcements. Agents face lower long-term interest rates, higher stock prices and a lower real value of the USD. Growth rises, unless you think that changes in real asset prices are neutral.

    My exact numbers are estimates from various studies, extrapolated. Space and time preclude restatement.

    If you accept that money can have real effects in the short-run, then you should accept that QE did have real effects.

    PS: Don't start on my alma mater.

  17. Without QE2, economic growth would have been lower than it was, oil and prices would have risen even more than they have, and the number of jobs created would have been less than it has been.

    This--the 'Obama measure' (that cannot be measured)--confirms that QE2, like the stimulus, was an unqualified success.

    Obama in 2012!

  18. I could say that the sky is green, but that would not make it true. If Obama is smart, he will keep his distance from the Fed.

  19. "Without QE2, economic growth would have been lower than it was, oil and prices would have risen even more than they have, and the number of jobs created would have been less than it has been."

    Part 1: Correct.
    Part 2: Obviously wrong.
    Part 3: Correct again.

    Sorry, 66.7%, you get a D.

  20. How do you people know these things? You'll have to tell me what line of reasoning leads you to these conclusions.