Krugman and I

Here’s Paul Krugman:

On the whole, I’m sympathetic to skepticism about the effectiveness of QE, predictably. After all, I’ve been arguing for forward guidance instead for 15 years. On the other hand, right now investors are not making a clear distinction between QE and forward guidance; taper talk has been accompanied by a clear shift in expectations toward the notion that the Fed will raise short-term rates sooner rather than later. So I wouldn’t be tapering now “” it sends a bad signal at a time when recovery remains very weak and fragile.

And here’s my view:

On the whole, I’m sympathetic to skepticism about the effectiveness of QE, predictably. After all, I’ve been arguing for NGDP futures targeting instead for 27 years. On the other hand, right now investors are not making a clear distinction between QE and future expected Fed policy; taper talk has been accompanied by a clear shift in expectations toward the notion that the Fed will tighten policy sooner rather than later. So I wouldn’t be tapering now “” it sends a bad signal at a time when recovery remains very weak and fragile.

Same place, different language.

PS.  Aren’t the MBSs bought by the Fed already backed by the Treasury?  If so, why does it matter so much which asset is bought?

PPS.  Just to be clear, in this post I’m treating QE as temporary, as does Krugman.  We both believe that permanent QE is highly effective (at least I think we both do.)

PPPS.  I’m still worried people are going to get confused.  Krugman basically says; “QE helps, but only because it changes the expected future path of policy.”  I say “QE helps, because it changes the expected future path of policy.”  I leave out the “but only.”  That’s how slight our differences are today.


Tags:

 
 
 

23 Responses to “Krugman and I”

  1. Gravatar of jknarr jknarr
    13. September 2013 at 09:21

    GSE bonds dont matter, but then again, all of the Fed’s balance sheet assets don’t matter. They could buy Fukushima fish as assets and it would not matter.

    There is no deliverable for the FRN liabilities, and the Fed has never shrunk its balance sheet — i.e. engaged in net selling, so who cares what they have on the balance sheet?

    Still, the nice thing about gold was that they could value it at any price, and expand the monetary base accordingly. They could have a $5 trillion 1 ounce gold coin if they so chose. They can’t do the equivalent with self-extinguishing (maturing) cash-bearing assets that have a clear present value price.

    Currency and reserves (liability) provision matters a whole lot to the economy, while the Fed choice of assets is moot.

  2. Gravatar of Doug M Doug M
    13. September 2013 at 09:27

    MBS are backed by mortgages…

    The mortgages may or may not be insured by private insurance. The issuing bank provides a partial guarantee, and FNMA and FHLMC further guarantees the timely payment of interest and the eventual payment of principal.
    Some MBS is backed by FHA and VA loans. The underlying mortgages are guaranteed by the respective government agencies. GNMA guarantees the timely payment of principal and interest.
    FNMA and FHLMC are private companies currently under government receivership.
    GNMA is wholly owned by HUD. GNMA securities are assumed to have the full faith and credit of the US government.
    FNMA and FHLMC securities are not explicitly guaranteed.

  3. Gravatar of Michael Michael
    13. September 2013 at 09:43

    I like Woodford’s argument that forward guidance is more important than QE.

    But QE may be more effective than forward guidance RIGHT NOW, because of the succession uncertainty. Unlike QE, following through on today’s forward guidance depends to some extent on who the next Fed chair is.

    There has been speculation that Bernanke badly wants to begin the unwinding of QE before he leaves – but given the uncertainty about his successor, this may be the worst possible time to start the unwinding.

  4. Gravatar of TravisV TravisV
    13. September 2013 at 10:45

    I thought this new piece by Peter Beinart was a fun read:

    http://www.slate.com/blogs/moneybox/2013/09/12/egalitarian_politics_policies_needed.html

  5. Gravatar of Dan S Dan S
    13. September 2013 at 11:47

    Scott, this is only tangentially related, but this talk about Krugman reminds me of reading his book The Return of Depression Economics, and he talks about international currency/financial crises over the last few decades, which reminded me of your post about fixed exchange rate regimes: http://www.themoneyillusion.com/?p=14668

    I was wondering if maybe you could talk in more detail some time about the ’97 Asian crisis, and how you think fixed exchange rates influenced that event. Maybe they allowed too much so-called “hot money” to enter in the form of USD denominated debt, which then became unserviceable when the countries suddenly had to devalue?

  6. Gravatar of ssumner ssumner
    13. September 2013 at 12:19

    jknarr, You said;

    “There is no deliverable for the FRN liabilities, and the Fed has never shrunk its balance sheet “” i.e. engaged in net selling, so who cares what they have on the balance sheet?”

    US taxpayers care.

    Doug, I was told the Fed only bought MBSs that had a Treasury backing (i.e. the GSEs) Is that false?

    Travis, That’s Yglesias.

    Dan, I’m not really an expert on that crisis.

  7. Gravatar of jknarr jknarr
    13. September 2013 at 13:54

    Well, the owners of the US Federal Reserve, whoever they are, care about their $250b of equity, and so they ought to care about the quality of their assets.

    I’m just not certain why anybody else should — why taxpayers particularly? Yes, the Fed now defers losses into a Treasury liability account, so as to preserve notional capital.

    So Scott, isn’t that an argument for the Fed to hold less rate-sensitive assets, not more? Why should NGDP stimulative success destroy their capital or increase future taxpayer liabilities?

    It almost looks like conflict of interest — the Fed cannot stimulate effectively without destroying its own (and/or Treasury’s) capital and create an asset-liability mismatch. Rate sensitive assets are quite possibly the worst assets that the Fed could ever hold!

    Few will like hearing this, but gold would be a perfect CB asset — no mark to market problems or future taxpayer liabilities. If currency is undeliverable against Fed assets, they can hold any asset on the book. Why on earth hold rate sensitive exposure?!

    http://www.federalreserve.gov/publications/annual-report/files/2012-annual-report.pdf

    Balance sheet on page 323. A very interesting document, BTW, lots of data scraps that don’t make it to the front page.

  8. Gravatar of Tommy Dorsett Tommy Dorsett
    13. September 2013 at 14:22

    Scott, the the stock market is near record highs and bond yields are not plunging and credit spreads are not widening (unlike when QE1 and QE2 ended), so maybe the message of the markets is 1) taper is priced in and 2) it’s not expected to slow the recovery.

  9. Gravatar of Bill Ellis Bill Ellis
    13. September 2013 at 14:31

    If not for the “Sumner critique” you would be policy twins.

  10. Gravatar of Doug M Doug M
    13. September 2013 at 14:39

    Agency MBS isn’t really backed by the Treasury. I don’t think it is relevant from QE point of view.

    But, it is a relevant distinction from a regulatory point of view.

  11. Gravatar of Max Max
    13. September 2013 at 15:06

    jkn,

    The central bank can always raise its target, whether it’s gold or CPI.

    In both the gold and CPI cases, somebody could take the view that the change is only temporary, so credibility matters. But in the gold case, credibility is enhanced because the CB can offer to buy gold at the new price, which is riskless for the CB. There is no riskless real asset for a CPI-targeting bank to buy.

  12. Gravatar of jknarr jknarr
    13. September 2013 at 15:17

    Max — yes, my question is whether the nature of the Fed’s assets contributes or detracts from their policy goals.

    On one hand, fixed income assets mean that the asset side value ultimately moves inversely to the size of the liability side (legacy bond values down if they print currency to infinity). This will “break the bank” as it were — destroy shareholder equity, which then becomes a (likely-debt-funded) taxpayer liability.

    Or, alternatively, they could hold any amount of gold, and simply revalue the price (or buy additional units) to expand liabilities (the monetary base), with no residual equity capital loss. At the infinite currency point, assets and liabilities are in perfect balance. If liabilities are undeliverable, why not?

    Bottom line, bonds appear to be the inferior asset for central banks.

  13. Gravatar of TravisV TravisV
    13. September 2013 at 15:27

    “Democrats harden opposition against Summers for U.S. Fed”

    http://www.reuters.com/article/2013/09/13/us-usa-fed-summers-tester-idUSBRE98C0W620130913

    “Four Democrats on the Senate Banking Committee are now expected to vote “no” if President Barack Obama nominates former economics adviser Lawrence Summers to be the next chair of the Federal Reserve, complicating one of the most vital decisions of his second term.

    Jon Tester on Friday became the latest member to publicly announce his opposition, taking to three the number of senators on the committee who are known to be against Summers, while a fourth, Elizabeth Warren, is also expected to be a “no” vote.”

  14. Gravatar of TravisV TravisV
    13. September 2013 at 15:31

    Neil Irwin:

    “Is the Larry Summers nomination falling apart?”

    http://www.washingtonpost.com/blogs/wonkblog/wp/2013/09/13/is-the-larry-summers-nomination-falling-apart

  15. Gravatar of Max Max
    13. September 2013 at 15:58

    jkn, gold is only riskless if the target is gold. If the target is CPI, then gold is risky.

    A CPI targeting bank is inherently a zero-reserve bank. No reserve asset exists, because you can’t store CPI in a vault.

  16. Gravatar of benjamin cole benjamin cole
    13. September 2013 at 16:07

    Excellent post. BTW and worth noting, Krugman recently blogged that Summers had only a window to make an impact snd so should consider something big.
    That says to me that Krugman is advocating a more-aggressive growth policy and perhaps a radical step (by the Milquetoast standards of central bankers).
    Will Summers listen?
    And I hope QE is permanent…my bet is that it is QE that works, not pious posturing…

  17. Gravatar of Jon Jon
    13. September 2013 at 22:19

    The issue with the MBS is the duration.

    The fed is very long on the yield curve staying flat. So on a mark to market basis they stand loose a lot of accounting value.

    Other issue I have with twist and the MBS program is that term structure of treasury debt on a consolidated basis has become very short. This is not the right way to finance a long term liability.

  18. Gravatar of lxdr1f7 lxdr1f7
    13. September 2013 at 23:04

    “QE helps, because it changes the expected future path of policy.”

    Does QE help because it is a better transmission mechanism than targeting the interest rate on reserves therefore the future path of policy is easier to infer?

  19. Gravatar of ssumner ssumner
    14. September 2013 at 05:47

    jknarr, You don’t seem to realize that the taxpayers effectively own the Fed. Fed profits go directly to the Treasury. Bad investments reduce those profits.

    Tommy, Good point. Here’s what I have been saying:

    1. Taper is clearly a net negative, markets fall on taper news. Yes, it’s mostly priced in now.

    2. Other powerful factors are pushing stocks up. That might be faster growth expectations, although personally I don’t see many signs of fast growth. So that is a bit of a puzzle to me.

    Doug, I had thought the Treasury guaranteed GSE bonds during the 2008 crisis. Is that wrong?

    Travis, Thank God for Tester.

    Jon, I’d add that it’s not just a mark to market problem. If rates rise the Fed must either sell off a ton of MBSs, maybe at a loss, or raise the IOR. So that’s a problem for the Fed. But it’s not a problem for the consolidated (Fed + Treasury)

    lxdr, Right now that’s really hard to say. I would not rule out the possibility that a negative 2% IOR would be more effective than recent QE. It depends on lots of factors.

  20. Gravatar of jknarr jknarr
    14. September 2013 at 07:48

    Scott, come on, I clearly know that losses accrue to the treasury. I stated that above – the fed has a deferred loss account against the treasury.

    Can you address whether bonds are a good asset for central banks? They are liquid, which is the most important characteristic ( so the cb does not distort prices).

    Bonds as assets and currency as a liability clearly sets up for a present value asset-liability mismatch, i.e losses forced onto fed capital or the taxpayer. I don’t see the treasury saving the interest remittances against the inevitable future losses.

    These losses are not necessary. CBs could hold a high quality liquid asset where the present. Gold would work quite well – they could call the asset value of gold at any price to balance the amount of liabilities they wish to print. They already price gold wherever they like on their balance sheet ( interestingly, the ECB has market prices of gold, which is a really interesting dynamic on their balance sheet).

    I feel that bonds on the balance sheet hampers effective monetary policy. I’m afraid that NGDP futures may suffer the same asset-liability mismatch, though would be most excellent for guiding policy, and would so keep a better asset-liability balance sheet than is the case presently.

  21. Gravatar of Luis Pedro Coelho Luis Pedro Coelho
    14. September 2013 at 15:17

    Let me try my version of Scott:

    On the whole, I’m sympathetic to skepticism about the effectiveness of QE, predictably. After all, I’ve been arguing for NGDP futures targeting instead for 27 years. On the other hand, right now it is impossible to use aggregate investor behavior to make a clear distinction between effects of QE and future expected Fed policy; taper talk has been accompanied by a clear shift in expectations toward the notion that the Fed will tighten policy sooner rather than later. So, while *I* could be tapering now while maintaining future expectations intact, the fact that *they* are tapering now is a bad signal at a time when recovery remains very weak and fragile.

    (I did get the point of trying to change as few words in Krugman’s post as possible.)

  22. Gravatar of Geoff Geoff
    14. September 2013 at 16:45

    NGDP futures will have no value apart from what discretionary interest rate the Fed pays on margin accounts.

    They will therefore not reflect NGDP expectations, but simply the difference between the rate the Fed would pay, and other rates.

  23. Gravatar of ssumner ssumner
    14. September 2013 at 16:59

    jknarr, Sorry, but when someone says they can buy Fukushima fish and it wouldn’t matter I do not assume they understand the Fed’s situation. Gold is a risky asset with a low rate of return, I see no reason to buy it, or silver or zinc either. If they want to gamble, buy stocks.

    Luis, Yes, much better.

Leave a Reply