Is QE Risky?

Paul Krugman says yes, and when I first read his post I thought he was right.  Now I’m not so sure.

Here is Krugman’s argument:

The big policy news this week has been the Fed’s decision to buy $1 trillion of long-term bonds, going beyond the normal policy of buying only short-term debt. Good move “” but it’s probably worth pointing out that yes, this does expose the Fed, and indirectly the taxpayer, to some risks. . . .

The Fed is, however, creating a new liability: the monetary base it creates to buy these bonds. In effect, it’s printing $1 trillion of money, and using those funds to buy bonds. Is this inflationary? We hope so! The whole reason for quantitative easing is that normal monetary expansion, printing money to buy short-term debt, has no traction thanks to near-zero rates. Gaining some traction “” in effect, having some inflationary effect “” is what the policy is all about.

The problem may come when the economy recovers, and inflation starts to become a problem rather than a hoped-for outcome. Basically, there will come a time when the Fed wants to withdraw that extra $1 trillion of money it created. It will presumably do this by selling the bonds it bought back to the private sector.

But here’s the rub: if and when the economy recovers, it’s likely that long-term interest rates will rise, especially if the Fed’s current policy is successful in bringing them down. Suppose that the Fed has bought a bunch of 10-year bonds at 2.5% interest, and that by the time the Fed wants to shrink the money supply again the interest rate has risen to 5 or 6 percent, where it was before the crisis. Then the price of those bonds will have dropped significantly.

And this also means that selling the bonds at market prices won’t be enough to withdraw all the money now being created. So the Fed will have to sell additional assets; if the rise in interest rates is at all significant, it will have to get those assets from the Treasury. So the Fed is, implicitly, engaged in a deficit spending policy right now.

My back of the envelope calculation looks like this: if the Fed buys $1 trillion of 10-year bonds at 2.5%, and has to sell those bonds in an environment where the market demands a yield to maturity of more than 5%, it will take around a $200 billion loss.

I’m not complaining; I think quantitative easing (it’s really qualitative easing, but I give up on trying to fix the terminology) is the right way to go. But we should go into it with our eyes open.

Just thinking out loud, but doesn’t this policy actually reduce risk to taxpayers?  The Federal government has something like $8 trillion in outstanding Treasury debt.  The taxpayers face a huge risk that the real value of this debt could rise through deflationary monetary policies.  Indeed that is exactly what has happened recently, as tight money has caused deflation, reducing nominal interest rates and sharply raising bond prices.  If the Fed bought $1 trillion of the T-bonds held by the public, doesn’t that reduce risk for the consolidated government balance sheet?  If the Fed does succeed in reflating the economy, and long term yields rise back to normal, then the Fed may lose $200 billion (although I doubt it will be that much) but the Treasury would gain much more.  Hence taxpayers would still be better of with an expansionary policy by the Fed.

I’m not sure if Krugman understood this point when he wrote the post (I certainly didn’t when I first read it.)  It is possible that Krugman is using the term “risk” in a very loose sense, more akin to how ordinary people use the term.  Average people think of risk asymmetrically, as risk of loss, but not risk of gain.  Krugman might argue that if the Fed engaged in QE, and was successful, there is a risk the Federal Government might not gain as much from reflation as they would have without the QE.

Consider two possibilities.  One is that the Fed has inside information about its intention to reflate, and should use that information to its advantage.  Krugman’s column occasionally hints at this perspective, but he also doubts about whether QE will work.  Even if the Fed has inside information, and knows QE will work, QE might be worth doing if it is the only way to reflate.  In that case the only question is which type of bond purchases will minimize the losses to the Fed.  (BTW, if this was Krugman’s view, so far he has been right.)

In other posts I have expressed skepticism that the Fed really has inside information.  I have argued that although common sense suggests they should have inside information, they are actually a big lumbering bureaucracy and the financial markets are just as able to predict the Fed’s future actions as the Fed itself is.  I argued here that in December 2007 the financial markets out-forecast the Fed regarding the future path of the fed funds rate.   If the EMH holds, the Fed is just as likely to gain as to lose from its bond purchases.  And this means that QE will reduce aggregate risk to the taxpayer, under any definition of ‘risk.’

BTW, Krugman’s argument is just one more reason to do QE by charging an interest penalty on excess reserves.  That would allow for $800 billion in QE without buying one more bond.

P.S.  Now that school’s out I am starting to catch up with other blogs.  I was embarrassed to notice that Krugman also had a “Being There” post, and even quoted the same passage.  Now I understand why famous people get caught plagiarizing so often—it’s easy to do without even realizing it.



7 Responses to “Is QE Risky?”

  1. Gravatar of Nick Rowe Nick Rowe
    9. May 2009 at 06:30

    Your argument makes sense Scott. But here’s another way of thinking about it:

    There’s “loss” relative to where we are now; and there’s “loss” relative to where we would be in the future if we didn’t do QE, and so the economy took a lot longer to recover, and the price level would be lower than it would be if we did QE now (and it worked, of course). It’s that counterfactual conditional loss that matters.

  2. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    9. May 2009 at 07:20

    Gees, Krugman changes his mind more often than a teen-age girl. But, that aside, ‘Franco Modigliani, call your office.’

  3. Gravatar of ssumner ssumner
    9. May 2009 at 09:17

    Patrick, The Modigliani angle is interesting, but it’s over my head. Finance is not my strong suit.

    Nick, I agree that your second counterfactual is the more reasonable way of looking at things. I also forgot to mention that if the Fed had inside information, which led it to expect a capital loss, the government could probably hedge that risk in the futures markets. Even better, the Fed should just buy short term bonds. I don’t see their moving into long term debt as accomplishing much of anything. The point is not to lower long rates, it is to raise them. (Which they finally seem to be doing, thank God.)

  4. Gravatar of Nick Rowe Nick Rowe
    9. May 2009 at 10:13


    “The point is not to lower long rates, it is to raise them. (Which they finally seem to be doing, thank God.)”

    Exactly! It is both hilarious and depressing to read the UK press saying that the Bank of England’s QE (buying long bonds, or “gilts”) has failed, because the yields on those gilts are rising, after an initial fall.

    It’s rising stock prices that seem to me to give the relevant signal (touch wood!)

  5. Gravatar of 123 123
    9. May 2009 at 11:59

    “P.S. Now that school’s out I am starting to catch up with other blogs. ”
    You might be interested in these posts where Delong gets lost in the IS/LM model:


  6. Gravatar of Devin Finbarr Devin Finbarr
    9. May 2009 at 13:57

    The whole line of thinking seems pretty crazy to me. Take a dollar bill out of your wallet. It says “Federal Reserve Note”. The idea of the Fed or USG worrying about dollar losses is non-sensical, since USG has an infinite supply of dollars. The point of any monetary or fiscal policy is to maximize the real goods and services those dollars can buy. Nominal values do not matter to a currency issuer.

    A major source confusion of economic commentary, including commentary by the politicians and Fed officials, is still thinking about economics in terms of a gold standard or fixed exchange system. People talk as if we are still on some sort of Bretton Woods era, hard money standard. Like it or not, the U.S. monetary system has been a de facto “soft currency” system for four decades. The best source of how to think of economics from a “soft currency” perspective is Mossler:

  7. Gravatar of ssumner ssumner
    9. May 2009 at 18:24

    Nick, Yes, I know that you and I both made that point back in March. Just so I don’t sound like a “know it all,” the big drop in yields after QE was announced caught me off guard. The model I have in my head would not have predicted that market reaction. And I can’t argue the announcement had zero credibility, because the dollar fell dramatically against the euro on the news. But I do feel better about things today.

    The world is endlessly complex. Here’s an off the top of my head hypothesis. The decision of the PBOC to freeze the yuan at 6.8 last year has meant that Chinese monetary policy is becoming effectively more expansionary every day. This is due to their rapid productivity growth, plus the Balassa-Samuelson effect. There is some recent evidence that China may recover quickly. The Fed’s March QE decision probably didn’t help much, but was at least a Bernanke put against deflation. As East Asia starts growing faster the equilibrium world real interest rate will start rising. That plus slightly rising inflation expectations in the U.S. will make the effective stance of monetary policy more expansionary. Stocks see this happening, and rally strongly. Of course it all may fall apart for some reason next week that both I and the markets missed.

    One irony if I am right, it suggests the world economy might have recovered quicker if the U.S. and Europe actually encouraged the Chinese to devalue last year. As I get older the income effect (in both directions) seems much more important than the terms of trade effect. (Look at the Can$ since last year when it was at parity–and look at the change in Canadian exports to the U.S. in that period of time.)

    Thanks 123, I’ll check out DeLong (although I hate IS-LM.)

    Devin, Here’s how I interpret the gold standard legacy. The press says the Fed has made a good effort to reflate–gave it the old college try. How is any M increase less than infinity adequate, if NGDP is below target? Yes I know there are limits (short of infinity) to what the Fed can buy in practice, but they have never come close to exhausting their options.

Leave a Reply