Thursday, December 9, 2010

QE2 and Rising Yields

More folks are now considering the possibility that the rise in long-term interest rates may be a sign of economic recovery.  For example, see Ryan Avent, Paul Krugman, Cardiff Garcia, and David Andolfotto.  This is a welcome reprieve from the those observers who point to the rising yields as a sign that QE2 is failing. 

Recall that the recovery view begins with notion that a successful QE2 will first raise inflation expectations.    The increase in  inflation expectations, however, also implies higher expected nominal spending (i.e. higher future nominal spending means higher future inflation).  Higher expected nominal spending in an economy with sticky prices and excess capacity should in turn lead to increases in expected real economic growth.  Finally, this higher expected real economic growth should increase current real long-term yields.  Given the fisher equation, this understanding implies that the rising long-term nominal yields are occurring because of both higher expected inflation and higher real yields.

Though it too soon to know for sure,  the data seem to support the recovery interpretation of the rising nominal yields.  Below is a figure showing the 10-year expected inflation rate and the 10-year real interest rate from the TIPs market.  This figure shows that inflation expectations pick up first and eventually the real interest rate does too: (Click on figure to enlarge.)

If this understanding  is correct, then QE2 seems to be doing some good.

Update:  Ryan Avent is now fully on board with the recovery view of rising yields:
BUTTONWOOD continues to be sceptical of the Fed's new security purchases. And, you won't be surprised to hear, I continue to think his fears are somewhat misplaced. He writes, for instance:
It is possible that the Fed's actions have actually been counter-productive, since 10-year bond yields are actually higher than they were when the second round of QE was announced. In part, this may be a case of "buy on the rumour, sell on the news". But the 30 basis point rise in bond yields over the last two days has revealed signs of investor cynicism.
Reduced bond yields are not the Fed's goal. The Fed's goal is to facilitate recovery, so as to move inflation and unemployment closer to the central bank's target levels. Beginning in late August, the Fed signalled its intent to do more to achieve its goal through additional purchases of Treasury securities. And indeed, the Fed's messaging was successful; Treasury yields were lower in early October than they were in late August. But lower yields were the means, not the end. The promise of more Fed action boosted markets and expectations, and before long actual economic data was following suit. But of course, we'd expect an improving outlook for the American economy to lift American government bond yields. Yields were low, aside from Fed activity, because investors were uninterested in putting their money in private projects. That's no longer the case; with rising growth expectations comes rising interest in private investment, which makes for falling bond prices and rising yields. Yields are rising because QE2 has been successful.


  1. Is there any independent evidence that expected real GDP growth is increasing?

  2. Nick,

    The only solid evidence I know is that U.S. consumer confidence is up.

    Others are looking to the new fiscal stimulus bill as improving the outlook. For example, the well-known economic forecaster Mark Zandi increased his 2011 forecast of real GDP from 3% to 4%.

    So yes, my recovery interpretation of the rising yields is tentative at best.

  3. Look at a longer time frame on the yield chart for 10-yr treasuries. Delong conveniently has one on the top of his page.

    Looks like just another tooth in the saw blade.

    In fact, current yield is about in the middle of the trend channel. It's hard for me to get excited until something dramatic happens, and this is a long way from it.


  4. Reduced bond yields are not the Feds goal? I beg to differ. If the Fed is concerned with deflation,(Bernake has stated such), and is trying to increase inflation, real existing long term rates should decline. Granted, new issues may in fact offer higher nominal yields as recent market activity indicates. However; this could be attributed to the signal from the White House and Congress that more deficit spending is forthcoming in the form of the extension of the Bush tax cuts. (Wrong type of stimulus.)

    Everyone seems to be focused on the 10 year rates. Look at recent activity in the 20 and 30 year rates. The 20 year is up 24 basis points since 12/01, and the 30 year is up only 19 basis points in the same time period. Interestingly, the 5 yr. bonds have risen approximately the same amount as the 10 yr during the period in question. (last 10 days.)

    The question in my mind, is, will nominal rates exceed the rate of inflation going forward? If inflation erodes nominal rates look for real rate declines.

  5. I should have added one other point: The velocity of money (M1 Multiplier), is a key factor with regard to the inflation dynamic. Without velocity, inflation will be difficult to materialize. Are current conditions a new paradigm? Let's call it a velocity of money trap. (Credit to Jazzbumpa for bringing the M1 Multiplier condition to my attention.)

  6. nanute:

    You need to think about this just a bit more.

    While I don't know what Bernanke is trying to do for sure, the reality is that an expansion of the quantity of money _can_ raise real rates through expectations of higher levels of output and employment in the future.

    I think your error is holding credit demand for credit (or, equivalently, the supply of saving and demand for investment) constant. The Fed impacts (in this situation, increases) the supply of money (and credit.) This reduces real interest rates, _given_ the demand for credit.

    But in reality, the demand for credit isn't given. Suppose two years from now, the demand for credit will be substantially higher. The level of interest rates at that time consistent with real expenditures equal to productive capacity and inflation on target will be higher.

    If this is correctly forecast, then real interest rates on securities of longer than two years term to maturity will rise.

    The only extra step is to see that the Fed's purchases of long term bonds today can make that scenario two years from now more likely. And so, long term real rates rise in response to the Fed's purchases of long term securities. This simply requires that those holding the securities sell more than the Fed buys.

    While Bernanke says that he wants higher inflation expectations, which would, if successful, raise long term nominal rates as well, even if inflation were on target and there was no plan to raise the target, the higher level of future output due to recovery could result in higher real rates.

    By the way, none of this makes any sense unless the current level of prices (and wages) is too high, so that real expenditures is below the productive capacity of the economy. With market clearing, it makes no sense. And so, there are plenty of economists today, who either cannot, or will not, see this possibility.

    Now, how are higher interest rates consistent with bringing about recovery now? Those selling the bonds to the Fed and to others (which is what raises their yields) must intend to spend the funds they obtain on consumer and capital goods. All of those firms with lots of "cash" (which are securities) must sell some of them and buy capital goods. They will do this because they expect higher levels of output (really, sales) two years from now. Of course, firms can also sell newly issued bonds too, borrowing to fund the purchase of capital goods. These involve a decrease in the supply of credit from everyone other than the Fed and an increase in the demand for credit. These are an increase in the demand for investment.

    Similarly, households can sell off securities they hold and fund consumer spending. Or, some households may borrow. Also, this raises the demand for credit and decrease in the supply of credit. It is a decrease in the supply of saving.

    If none of these expectations changes occur, then the Fed's purchases of securities will reduce real interest rates, and given expectations of future sales, production, and employment, the lower real interest rates will make increased spending somewhat more attractive. But to the degree people realize this, and so expect that spending will be higher in the future, other elements of what people do change, in such a way that real interest rates actually rise.

  7. This comment has been removed by the author.

  8. This comment has been removed by the author.

  9. This comment has been removed by the author.

  10. Dave,

    What I struggle with is the rising real yield part of the story. Sure, the Fed isn't targeting lower nominal yields - but it should at least be looking for falling real yields, at least if it wants to stimulate macroeconomic activity it will (as much as it can stimulate macroeconomic activity through a credit channel that remains quite clogged).

    Perhaps the real yield simply reflects improved growth prospects, but it could also offset those prospects as well. The Fed's stoked inflation expectations - but unless economic activity does rebound smartly, I suspect that nominal yields will compress somewhat.

    Rebecca Wilder
    p.s. Confidence is up, but only back to levels las seen in June of this year (when it hit 76). The S&Ps had quite a ride of late - very likely correlated to giddy consumers.

  11. This comment has been removed by the author.

  12. P.S. David, sorry for addressing you as Dave earlier! Rebecca

  13. John Cochrane is a boob-Nipponista.

    Mild deflation is good? Has he never heard of Japan?

    Mild inflation is bad--as in the 1980s-1990s in the USA? Oh, we suffered so much.

    The Nipponistas are the mose dire threat America faces today--they make the Taliban look microscopic, in comparison.

  14. JDTapp: I was disappointed with his explanation of how QE2 could work.

    Rebecca: Dave or David is fine, no worries. In regards to your questions, first, I readily admit this is a highly speculative post. I would want to see a lot more data points before become truly convinced myself. Jazzbumpa may be right that the the latest movements may end being inconsequential.

    With that said, I view monetary policy as influencing the real economy through many channels. The dropping of real yields story of how QE2 could work only assumes the interest rate channel. Even then, though, the liquidity effect doesn't last forever.

  15. How does the fact that bond yields jumped materially after no hints at an expansion of QEII in yesterday's FOMC press release impact your analysis. Seems to contradict some of those that you quoted...