Wednesday, May 1, 2013

Small monetary policy question

A small monetary policy question has been bugging me this morning.

The Fed is buying long term bonds and agency debt, with stated desire to drive down long term Treasury and mortgage-backed security rates.

Why does the Fed not simply say "We are going to peg the 10 year Treasury rate at 1.5%. We buy and sell at that price." If the Fed wants the 10 year rate to be, say, 1.5%, then this seems a lot simpler than setting a quantity ($80 billion a month), and then enduring endless arguments with academics like me whether it's having any effect at all, or commissioning complex staff studies to determine whether the impact is 10 or 15 basis points and how long it lasts.

The answers I can come up with are not pretty. Perhaps the Fed understands that any "segmentation" is smaller than it thinks, and doesn't last that long. So, the required bond purchases would rapidly explode in size. If that's the answer, then the Fed isn't doing it so that it can continue to seem powerful when in fact it really is not.

Perhaps it's political. If the Fed can say "we're buying $80 billion a month. This is helping, but we don't know exactly how much," then it avoids responsibility for what the rate actually is. If it says 1.5%, then every car dealer and mortgage broker in the country wants to know, why not 1.4%? This is even more cleverly Macchiavellian. But such deeply political decisions are a long way from the benevolent independent central bank we write about in papers.

Any ideas anyone?  

While we're here, two great quotes from Fed economists (obviously un-named) I've talked to recently.

Me: Aren't you worried about big banks borrowing short and lending long? What happens when, inevitably, interest rates rise?

Economist A: "Don't worry, we'll let them know ahead of time."

This was some time ago. I think last week's announcements that the "stress tests" were (at last) going to include plain - vanilla interest rate risk might count as such warning.

Economist B: "Inflation is just not a concern. The Fed now is balancing growth against financial stability."

A new dual mandate, and a fascinating can of worms. I'd love to see that Phillips curve.

 

29 comments:

  1. the quantitative policy does not just work through the rates but also through equity and real estate markets - hence simply looking just at one treasury yield would be a mistake

    ReplyDelete
  2. Would Economist A's statements suggest that large banks enjoy a second "too big to fail" premium in that not only are they getting cheaper interest rates due to implicit backing, but they are also getting insider information with regards to future rate changes. Given that the research suggests that following yield is a good strategy until it isn't, the banks seem to be getting all the benefits of writing out of the money options and being pulled out of the way of the steam roller by the central bank right before they get hit.

    ReplyDelete
  3. "Me: Aren't you worried about big banks borrowing short and lending long? What happens when, inevitably, interest rates rise?
    Economist A: "Don't worry, we'll let them know ahead of time."

    Apparently, the big banks are unconcerned. See

    http://www.federalreserve.gov/releases/z1/Current/z1.pdf

    Page 77 - L.109 - Private Depository Institutions - 4th quarter (2012)
    Total Financial Assets - $14.99 Trillion
    Total Liabilities - $15.26 Trillion

    Most of the run up in depository liabilities has come from the Fed's quantitative easing program - deposits up from $6.6 Trillion in 2008 to $9.4 Trillion in 4th quarter of 2012. This has essentially rendered the entire depository banking system insolvent (liabilities exceed assets).

    Your best investment opportunity here would be to short the commercial banks.

    ReplyDelete
  4. The flippant "monetarist" answer would be: We are trying to increase the money supply (and expectations of the future money supply), not drive down long-term interest rates. That's why it's called quantitative. Maybe the tranmission mechanism is via long rates, maybe other asset prices, maybe it's immaculate. Whatever. We don't care.

    ReplyDelete
  5. I would argue its simply a balance sheet issue, they want to keep control of their balance sheet size and now they can pretend it's their new tool for monetary policy.

    ReplyDelete
  6. I'm not sure if this is what you mean by ''segmentation'', but if the Fed did that, the increase in the monetary base would sooner or later lead to inflation, which would raise yields, which would force the Fed to engage in further open-market purchases to meet its target, and so on. Eventually we would get hyperinflation. Because the economy needs a little inflation now, pegging the Treasury rate would temporarily be a good policy, but an unsustainable one.

    ReplyDelete
  7. I'd say the answer is that the Fed's ultimate objective isn't to target a particular long-term interest rate, its to manage the price level. I also think that the public (and many economists) have the inapropos view that the main channel through which the Fed effects the economy is through lower interest rates rather than through the price level. Thus I think that all the rhetoric about policy lower treasury or mortgage rates simply caters to that belief and is more like p.r. relations than anything else.

    ReplyDelete
  8. "pay no attention to that man behind the curtain..."

    Paul

    ReplyDelete
  9. Maybe those Fed bond purchases are a way for the Fed to commit to keeping the short-term rates low for a long time. They have to fund these bond positions at the short-term policy rate and if they become insolvent (not mark to market, but eventually when the smoke clears) they have to ask for more money from the Congress. The Fed doesn't want to go ask for more money, market knows that, and hence the commitment works.

    ReplyDelete
  10. Didn't they peg yields after WWII and just let inflation rise?

    Economist B may be correct given the recent PMI, ISM, EIA and ADP numbers. I think higher taxes since January are going to bite hard. Maybe this process started earlier but taking money out of consumer hands just wasn't very bright.

    ReplyDelete
  11. Setting a particular rate like 1.5% might make it more difficult to get out of that rate -- how does one unwind that policy? By stating that they're raising the rate? How do they pull the plug all together?

    Surely it would be nice to return nominal interest rates, i.e. the mechanism for allocation of capital, back to the market! Stating the quantity of asset purchases may make the interest rate target more fuzzy -- but it's exactly that flexibility that will be needed during the unwind...

    ReplyDelete
  12. Do you mean peg it there forever, or just for awhile?

    The latter seems tough to implement. How do you change the peg? Since there is no permanent commitment, huge pressure will develop as the next move becomes increasingly inevitable and the Fed will then get run over when it finally acts. It's like an exchange rate peg: you either have to mean it or forget it. Prices have to move continuously as expectations change or not at all.

    The former, of course, requires them to follow a path of the short rate that compounds to 1.5% over 10 years or it creates an unlimited arbitrage opportunity. So it's equivalent to 100% guaranteed forward rate guidance, which is a really bad idea because forward rate guidance should be conditional.


    Regarding Fed economist A: It's broadly suspected that banks, and especially clearing banks, have a huge informational advantage in rates markets. While FOMC members probably can't disclose how they or other committee members intend to vote, there is nothing preventing them from discussing with their friends how they "feel" about possible Fed actions. Just witness the extent to which markets responded in the run-up to each QE announcement. By the time each announcement was made, the general features had somehow essentially been fully anticipated by the financial press, despite the lack of official disclosures from the Fed.

    ReplyDelete
  13. I'm confused, seems to me you're phrasing the question incorrectly.

    The goal isn't lower long-term rates and has been fairly explicit (bond buying until PCE and URate are @ predetermined levels) despite the lack of accuracy in hitting said goals. Stating the persuit of lower long-term rates is the Fed's way of explaining to an ignorant policy how bond buying can (hopefully) impact the goals of 6.5 URate and 2% inflation. Fed doesn't care if, to obtain their goals, it takes a 10 year Treasury rate at 1.5% as you mention or if it is 0.5% or 5.5%.

    ReplyDelete
  14. John, the answer is simple. The Fed is a legislative body and thus the policy has to be presented in a written "form" that will get enough votes to pass or perhaps even more than enough, in which case they really have to be dumbed down.

    The same is true with appellate court opinions. They are, in effect, "bills." They have to be written in a way to gather votes.

    ReplyDelete
  15. John, the answer is simple. The Fed is a legislative body and thus the policy has to be presented in a written "form" that will get enough votes to pass or perhaps even more than enough, in which case they really have to be dumbed down.

    The same is true with appellate court opinions. They are, in effect, "bills." They have to be written in a way to gather votes.

    ReplyDelete
  16. John, it also seems to me that the Fed is targeting or budgeting for the creation of HQC or safe assets, w/o regard to the interest rate.

    "QE is a transformation of non-cash HQC to cash HQC," This is from page 66 of the current, Office of Debt Management (Treasury) Fiscal Year 2013 Q2 Report

    If you look at this report, it has projections or budgets of the demand for HQC and projections or budgets for the possible sources to meet such demand, at lot of which is driven by bank regulation (all of which the report labels as pro cyclical). IOW complete madness.

    ReplyDelete
  17. I'm probably missing something, but wouldn't the Fed need to peg prices of all nominal yields to make this work? What if I wanted to sell 9 year bonds to the Fed? If they only accept 10 year bonds in exchange for money, wouldn't this lead to a kink in the curve at 10yr?

    ReplyDelete
    Replies
    1. They could peg the whole curve. But they wouldn't have to. Are markets so segmented that nobody is left who will short 10 and buy 9? Are Treasury swaps totally dead?

      It's actually a bigger question. Why, back when rates were not zero, did the Fed set daily reserve quantities and not simply borrow and lend overnight at its target rate?

      Delete
    2. I see. If the Fed said: we will buy/sell at fixed prices, but only bonds with, say, 3589 days to maturity, that might be complicated if only a small number of those bonds actually exist.

      Delete
  18. I think the more relevant question is why the Fed states that it wants to lower long-term interest rates at all. Tight money has an a priori ambiguous effect on interest rates. Yes, it's true that in the market for short-term bonds, when the fed reduces the size of its balance sheet by selling gov't debt and taking cash out of the economy, it lowers the price and increases the interest rate on that debt.

    But, I think most economists would agree that if the Fed were to announce today a 300 basis point hike in its target for the fed funds rate, that it would sharply lower long term interest rates, as the market would begin to expect deflation.

    Think about what Volker did in whipping inflation in the early 80's- by RAISING short term interest rates to extreme levels, he caused long term interest rates to COLLAPSE, as market expectations of inflation finally went away.

    I assume that when the Fed wants easier monetary policy, it actually wants to do the opposite of what Volker did, and have long term rates to go UP. So I find it highly confusing for them to say they want long term rates to go down.

    ReplyDelete
    Replies
    1. Actually, Bernanke has discussed this several times. He argues that the Fed is not to blame for "punishing savers with low rates" because if they engaged in contractionary policy, then interest rates would eventually -- if not immediately -- only fall further. When he says they want to lower rates, he means expansionary policy which lowers rates through one channel (the traditional channel that they teach in undergrad econ) but might actually raise rates in the net by boosting NGDP.

      Delete
    2. This illustrates why the Fed would not want to target a particular long-term rate. If the Fed chose a rate, such as 1%, then the boost to NGDP might be large enough to push long-term rates up to the point that 1% becomes increasingly inflationary. Moreover, this might all happen very quickly and well before the interest rate ever actually hit 1%. Long-term rates are influenced in multiple ways by expansionary monetary policy, so the Fed shouldn't care where they actually end up.

      Delete
    3. I have heard that Bernanke has acknowledged that low long term interest rates primarly reflect weak growth. But that only reinforces my confusion about why the Fed is targeting low long term interest rates.

      Delete
    4. The point is that the Fed is not targeting low long-term rates. The Fed is taking actions that lead to lower rates through one well-known effect, but should ultimately lead to higher rates. Perhaps Bernanke has said he wants lower rates when he explaining how his asset purchases will boost the economy. But, based on talks he has given, it is clear that he believes monetary contraction would lead to lower rates. Note that the Fed does not specify some long-term rate they want to hit, but rather an unemployment rate or inflation rate that will cause a change in policy.

      Delete
  19. I agree with K. The Fed might be concerned about fixing short-term T-bills forward rates, which would be a side-effect of pegging T-notes and T-bonds yields. I don't know the literature on the subject, this might or might not be a big problem.

    Otherwise, it seems to me to be a rather effective instrument to reach core-CPI and employment goals (which the Fed has failed at in the last 5 years). It is probably a better instrument than buying (or selling, if the economy overheats) assets at a fixed and arbitrary rate. But it is strange to buy and sell government debt in order to hit a 10-year yield target, which is chosen so to reach inflation and employment objectives. It is quite a detour.

    Can it be as effective as the Swiss method, which was to buy a ton of foreign assets with printed money?

    ReplyDelete
  20. The larger the Fed balance sheets, the larger the shareholders' dividend payment. It specifically states on their site the 6% remittance of capital. That's an ungodly amount of money now that we're talking trillions. They're laughing at the Congressman who demand for more reckless QE.

    ReplyDelete
  21. Initially, I thought there was a pretty obvious answer to this question. The Fed isn't going to engage in rate pegging because other Central Banks / Foreign Governments would probably highly object to that (What if rates in every highly developed welfare state were pegged?) However, the question is probably even more complicated than this. It's not clear that the Fed's buying of longer-dated securities has had any effect on the term structure since a) long rates have fallen in countries like Germany and Japan in the absence of QE on the same scale as the US (until recently in Japan's case) b) the Fed has basically been continually rolling forward its pledge to keep short rates 0-25bps for the next two years since 2010. So ultimately, I think it's probably a combination of politics (the effect of QE on FX markets is a bit more straightforward - take a look at the 20% drop in JPY since October) and powerlessness (the Fed has absolutely *NO* ability to raise interest rates anytime soon - the debate about Central Bank independence is entirely semantic).

    Anyway, I think the really interesting question brought up by the Fed economist is... why is everyone at the Fed so confident in their ability to signal "well in advance" to market participants that rates are going up? Personally, I hope the Fed blows some white smoke out of their chimney when they're 100% sure that rates are changing to an upward trajectory.

    ReplyDelete
  22. I agree that the policy is in place to increase the money supply. The lower long term rates are an unfortunate side effect. I say unfortunate because the lower long term rates actually diminish the banks' incentive to loan and increase the money supply -- a bit of a catch-22 going on there.

    Also, have you seen our banks' balance sheets recently? Google "excess reserves graph" and check it out. There will be no European contagion here.

    ReplyDelete
  23. Core PCE has been below 2 for years.

    I think the answer is that this Fed is very small c conservative, tentative, and interest rate pegging for longer rates is not something that they've done before. They've mostly spent the past five years twiddling their thumbs while the economy stagnates, occasionally doing things that seem big but in fact likely to have modest impact. For much of that time, the Fed believed low nominal rates by themselves constituted loose monetary policy, nevermind how low inflation was.

    ReplyDelete

Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.