Snarking Bloomberg

A headline on Bloomberg:

Fed Loans Guided by Raters Grading Subprime Debt AAA

Oh, my God! You mean the Fed is actually relying on ratings provided by major ratings agencies? Can you imagine such a thing? And did you know there are homosexuals in Iran? Shocking, isn't it? Hard to believe, but it's true.

Seriously, dude, this is not news! It would be fine if you labeled it as commentary and called it something like, "Why is the Fed Relying on Raters Who Graded Subprime Debt AAA?" There is a good case to be made that the Fed should be doing its own credit research rather than relying on agencies that have both conflicts of interest and abysmal recent records. And someone else can perhaps counterargue about the dangers of allowing public institutions to pick private sector winners. Whatever. It's all very interesting, but it's not news.

Krugman on Steinbrueck

If I were a currency trader, I'd be selling Euros.

Phillips Curve Presages Deflation

Back of the envelope analysis:


UE rate is now 6.7%.
Almost certain to rise over next couple of months.
Probably at 7% early in '09.

The most optimistic estimates have recession ending about halfway thru '09.
Stimulus program heavy on public works will be slow to implement.

In recent recessions, UE has continued to rise for at least a year afterward.
So a fairly optimistic scenario would have UE rising to 8% by YE '09 and staying there for a year.
That would give average UE around 7.5% for '09 and 8.0% for '10.

Typical estimates put NAIRU at 5%.
So UE gap is 2.5% for '09, 3% for '10.

Typical Phillips curve coefficient is around 0.5
which implies inflation rate will fall by 1.25% in '09
and an additional 1.5% in '10
and continue to fall.

Current inflation rate is around 2%


You do the math

I’m with Mephistopheles

Paul Krugman makes the case for deliberately providing too much of an economic stimulus. Here I want to examine the counter-case, for which I think there is a valid argument, although it doesn’t seem to be the argument that anyone is actually making. In the end, though, I agree with Professor Krugman: too much is better than too little.

Here’s the thing, though: there is one big advantage in doing too little economic stimulus. I mean, doing some, doing a significant amount, but not quite enough. The advantage (as I have argued in many of my recent posts) is that it doesn’t cost anything, because it can be financed with seignorage.

Professor Krugman makes the case that, from the point of view of the condition of the economy – the tradeoff between output and inflation, which is what macroeconomists usually care about – it is better to do too much, because, if you do too much, the Fed has the power to undo the excess by raising interest rates (whereas, if you do too little, there is nothing the Fed can do about it, since we’re in a liquidity trap). But as most people writing for the public acknowledge (or insist), there is more to care about than the condition of the economy. There is also the condition of the government’s finances.

As far as the government’s finances are concerned, it is not just a bad idea to do too much; it is a bad idea to do enough. As long as you don’t do enough, standard textbook macroeconomic theory (at least the Keynesian kind, which is the one most influential among economic forecasters) says that there will be disinflationary, and ultimately deflationary, pressure. It’s a pretty simple point of logic. “Doing enough” is defined as getting the economy on a track where the unemployment rate will go down to the NAIRU (or possibly lower, if deflation takes hold and we are trying to reverse it). What other definition could there be? Then by definition of the NAIRU (the “non-accelerating inflation rate of unemployment,” which, if looked at from the other direction, is the non-decelerating inflation rate of unemployment, or eventually the non-accelerating deflation rate of unemployment), the inflation rate will tend to keep falling. Therefore, the Fed can keep creating money, financing the Treasury, and there will be no inflation (or any other deleterious effects that I can think of).

Once you do enough, not only does the government have to start doing real borrowing to finance its deficit; it also has to pay real interest on the outstanding debt. The government is like a monopolist facing a kinked demand curve. As long as you stay below the kink, the more you produce, the better. But as soon as you get to the kink, all Hell breaks loose. Not only are you unable to sell your marginal “new” products for a profitable price; the price of your inframarginal “old” products starts to go down too.

This puts us in a position rather like Goethe’s Faust. (I’ve only read selected passages, and those in very loose translation, so I’m relying on someone else’s analysis here, and probably an inaccurate memory even of that.) Faust is granted unlimited knowledge, until the point where he can say it is finally enough, that he has reached the culmination of human experience. At that point his soul belongs to the Devil. Similarly we are granted unlimited economic stimulus, until the point where we can say it is finally enough. At that point we are damned to Eternal Debt.

But as I argued in an earlier post, even with a debt-financed stimulus, we have a net free lunch, in the same sense that Ricardians argue that international trade provides a net free lunch. With the stimulus, there is more produced today than there would otherwise be, and at no point in the future will production have to be reduced in order to pay for the increased production today. There is unambiguously more. There may be issues about distribution, as there are with trade, but overall there is unambiguously more. Perhaps the current generation benefits at the expense of future generations, but the current generation gains more than the future generation loses.

I want to say one other thing. Even with the biggest fiscal stimulus anyone can imagine, our debt-to-GDP ratio will still end up lower than it was at the end of World War II. As I recall, that situation didn’t work out too badly. In fact, as a representative of the Future Generations that were affected by that debt – a representative who has a self-defrosting refrigerator-freezer, an automatic washing machine and dryer, a dishwasher, an air conditioner, a flat-screen color television set, a VCR/DVD player, several personal computers, an air-conditioned car with automatic transmission and power brakes that gets 35 MPG on unleaded gasoline, a blackberry, wireless internet access, immunity to polio, the ability to make a living without putting my pants on, and the opportunity to go to sleep after dusk in New York and wake up before sunrise in London despite the time difference – I would like to thank our grandparents for choosing damnation. Hell doesn’t really seem all that bad.

And in Goethe’s version, doesn’t Faust end up going to Heaven?

My Fantasy

As a postscript to my last two blog entries, I want to imagine a possibility.

Suppose that the Treasury were to retire all its outstanding long-term debt, and suppose that the Fed were to buy enough of its short-term debt to keep the interest rate at zero. And suppose this were all done without causing inflation. I realize, for institutional reasons, this is all highly unlikely to happen, and it is also somewhat unlikely for economic reasons. But it isn’t inconceivable. In fact, leaving aside the institutional issues, I’d give maybe a 25% chance that it’s economically possible today, provided that the fiscal stimulus is not large enough (which is likely to be true in any case). You can disagree, but anyhow, this is only a fantasy, and I’m pretty sure I can get quite a few reputable economists to agree that my suppositions are far from inconceivable, economically speaking.

So here’s my fantasy:

Federal Budget, FY 2010

National Defense and Homeland Security  $730,213,476,219.56
Domestic Discretionary Spending         $773,154,217,326.40
Social Security                         $653,207,125,217.23
Medicare                                $417,316,224,327.01
Other Entitlements                      $615,222,143,177.03
Interest on the National Debt                         $0.00

OK, it’s only a fantasy. But as fantasies go, I think I’ve got a better shot at this than at dating Angelina Jolie. Or Isobel Wren, for that matter.

Behold the Power of Seignorage

For Sepetmber 2008, the monetary base was reported at $903.5 billion. For October, it was reported at $1.1285 trillion. That's an increase of $225 billion. That's how much money the Fed created directly in one month. A couple of months of that would be enough to finance last year's federal deficit. Three or four months would be enough to finance this year's likely deficit, including the first half of the TARP.

Granted, not all the money was actually used to finance the federal deficit. Much of it was used for even more "inflationary" purposes, such as emergency lending to banks. But the money was created, and it could have been used to finance the federal deficit. And if the Fed keeps creating money at anywhere near that rate, chances are that the portion created by buying T-bills and other Treasury securities will be enough to finance the current deficit, even if that deficit is larger than we expect.

And yet not many people seem to be worried about inflation. TIPS are selling at historically high yields relative to nominal Treasuries. Commodity prices are crashing. And the great debate of the time is about whether we will experience deflation.

Yes, Virginia, there is...

Treasury Finance Policy

Just a quick thought. Ben Bernanke suggested today that the Fed might start buying longer maturity Treasury securities now that it has done about all it can at the short end of the yield curve. Sounds like a good idea, but here's my question: If reducing yields on longer-term Treasuries is a desirable policy objective (and I think it is), then why is the Treasury still selling the damn things?

Granted, the Fed may ultimately need to buy even more than the Treasury is currently issuing, but wouldn't the first step, before having the Fed do something unconventional, be for the Treasury to shift its financing dramatically toward the shorter maturities? That would put downward pressure on longer-term yields and upward pressure on short-term yields, which the Fed could then relieve by buying more of them and bringing the yield back down to zero. (In other words, the Fed would do what it normally does but would have the chance to do more of it.)

From the public's point of view, it should be exactly equivalent. Under the policy Chairman Bernanke is suggesting, the Fed buys longer-term Treasuries, and it is just as if they had never been issued, while the additional short-term T-bills don't get issued in the first place. Under the policy I'm suggesting, the longer-term Treasuries don't get issued in the first place, and the Fed buys the additional short-term T-bills, so it is just as if they had never been issued.

I'll anticipate the answer and say that the availability of a certain amount of on-the-run Treasuries is important to the proper functioning of markets, so the Treasury doesn't want to stop selling them altogether. But it could reduce the size of its issues. And it could buy back some of its own securities on the open market. If the Fed confines itself to buying short-term T-bills, that would leave it more flexibility in the future if it should become necessary to withdraw some of the base money it is now creating.


UPDATE: People don't seem to be getting the point that these policies are equivalent. A shift to shorter-term financing by the Treasury would not put the Treasury at any greater risk (with respect to the need to refinance at possibly higher future interest rates) than would the Fed's contemplated policy of buying longer-term Treasuries. If the Fed buys the new T-bills (which it would, unless the economy starts to recover, in which case the Treasury could shift back to longer-term financing), the Treasury may never have to pay back or refinance the money at all. Possibly it would have to refinance, if the Fed ends up selling the T-bills (instead of rolling them over) to slow down the economy later on. But in that case, it's just a question of where the government takes the loss: if the original strategy was for the Fed to buy longer-term bonds, then the Fed would be liquidating those bonds at a loss, the loss would pass through into the Treasury, and the Treasury would have to borrow the money to make up the loss; if the original strategy was for the Treasury to shorten its financing term, then the Treasury would have to borrow the same money to refinance the expiring T-bills.

You might say that either one is bad policy for the same reason. I can appreciate the logic of that argument, but then what would you do to revive the economy? You can try a fiscal stimulus, but then the government has to "borrow" that money, so it's potentially on the hook for that repayment. (Again, though, it may not ever have to pay the money back, if the Fed finances it.) You can have the Fed buy some other kind of asset, such as agency securities, but if the securities are fully guaranteed, that's pretty much the same as buying longer-term Treasuries, and if they aren't fully guaranteed, then the Fed is taking a risk that passes through to the Treasury. You could argue that the Fed should take that risk anyhow, but that is similarly equivalent to having the Treasury take the risk (for example, by fully guaranteeing the assets). There's just no way that the government can both revive the economy and get complete assurance of not having to refinance for 30 years. That's the bad news. The good news is that there's a pretty good chance that the government would never have to refinance.