February 05, 2003
Deficits and Interest Rates Once Again

Ah. Here's what's going on:

Rex Nutting of CBS Marketwatch asks Bill Gale of Brookings and Glenn Hubbard of the CEA about the effects of "an extra $200 billion in deficits."

Glenn Hubbard interprets the question to be, "What would be the consequences of increasing the deficit by $200 billion above the previously-anticipated baseline this year only, and then reversing the increase and returning the deficit back to its original baseline next year, and leaving it there into the indefinite future?" As Nutting writes, Glenn "suggested that an extra $200 billion of deficit that lasted a year would only raise interest rates by a tenth of a percentage point at the most, hardly enough to crimp investment."

Bill Gale interpets the question to be, "What would be the consequences of increasing the deficit by $200 billion above the previously-anticipated baseline and keeping the deficit higher than the baseline into the indefinite future?" As Nutting writes, "William Gale, an economist with the Brookings Institution, says the probable impact is about 10 times higher. Gale figures a $200 billion permanent increase in the deficit could boost rates by 1 to 2 percentage points."

Note that there is no analytical disagreement at all. If asked, Hubbard would say that an anticipated $200 billion deficit next year would raise long-term interest rates this year by about 0.1 percentage points. And he would say that an anticipated $200 billion deficit the year after that would raise long-term interest rates this year by about 0.1 percentage points. And so on, and so forth. And by the time one has gone out 10 to 20 years Hubbard's answer is the same as Gale's: a permanent increase in the annual deficit of $200 billion relative to the previously-anticipated baseline raises interest rates by 1 to 2 percentage points.

It's very nice to see the reappearance of Glenn Hubbard Mark I: the Hubbard who wrote that you could see the pressure of the budget surpluses of the late 1990s lowering interest rates, and whose estimates of the effects of deficits are in the same ballpark as those of Peter Orszag, William Gale, Doug Elmendorf, Greg Mankiw, and company. Welcome back.

It's very nice to see the disappearance of Glenn Hubbard Mark II: the Hubbard who said that there was "no evidence" that deficits on the scale seen in the U.S. had any effect on interest rates. Good riddance. It's nice to see that Glenn's line in the past that budget deficits had no or small effects on interest rates are, as Richard Nixon's staff used to say, "inoperative": false, erroneous, misleading, inaccurate--and withdrawn.

But who has a better analysis of the likely impact on interest rates--and then on investment and growth--of the Bush Administration's 2004 Budget? Are the policies that the Bush Administration's wants to put into place this year more like a one-shot $200 billion one-year deficit that then comes to an end next year? Or are they more like a permanent $200 billion increase in the annual deficit?

Well, let's take a look at the difference between the deficits (and surpluses!) that the Bush Administration projected last January and those that it is projecting this January. These differences are due (in small part) to changes in economic and technical forecasts, and (in large part) to proposed changes in policies:

The effect of the policy (and forecast) changes between 2003 and 2004 has been to widen future deficits (and eliminate future surpluses) by an amount that averages 2.7% of GDP--roughly $300 billion dollars a year--in each of the next seventeen years (and thereafter as well). The Bush Administration policy changes look much, much, much more like a permanent widening of the deficit than a transitory one-year fiscal stimulus that is then reversed.

So Bill Gale's numbers apply. And Glenn Hubbard's do not. Notice that Glenn is not lying: his numbers are accurate in-the-ballpark estimates for the particular case he is considering: a one-shot one-year increase in the deficit that is then reversed. Note that Hubbard is not saying that the Bush Administration's policy proposals are like "an extra $200 billion of deficit that lasted a year." He is just hoping that readers will assume that the case he considers is supposed to apply to the Bush Administration's policy proposals.

Posted by DeLong at February 05, 2003 04:02 PM | Trackback

Email this entry
Email a link to this entry to:


Your email address:


Message (optional):


Comments

Professor DeLong says >He is just hoping that readers will assume that the case he considers is supposed to apply to the Bush Administration's policy proposals. <

This is not lying? I must have missed a segment in advanced ethics.

High level appointees swear an oath to the Constitution and to the nation. Misleading people into accepting policies that are destructive to the nation's interest amounts to lying. But then, we live in a Culture of Lies, in which lies are printed as one side of the argument, and truth is printed as the other.

Posted by: Charles Utwater II on February 6, 2003 08:11 AM

It would be nice if Delong would quantify the "small part" and the "large part," especially since just yesterday we were told by implication that it was only the "large part" that mattered.

To the extent that in the short term there have been dramatic changes in revenue, it is clear that those changes have not been driven in "large part" by changes in fiscal policy. Something else has happened, and that something else has been dramatic. And changes in technical forecasts, one might think, would have to be dramatic to account for that change.

Posted by: Thomas on February 6, 2003 03:54 PM

Well, the answer is that I don't know. The Bush Administration has not told me what part of the changes in the budget forecast are due to changes in economic assumptions, and what part are due to proposed policies.

I can see that the changes in assumptions about the pace of economic growth are small, and so I know that they play only a small part in the shift in the projected path of the deficit. But how small--1%? 5%? 10%?--I can't tell.

Posted by: Brad DeLong on February 6, 2003 05:06 PM

Paul Krugman has elsewhere suggested--and I have no reason to doubt him on this point--that it may be that given levels of GDP will not produce the same level of tax revenue that they did in the late '90s, other things, including fiscal policy, being equal. If that's true, and if there have been changes to the forecasts for that reason, in addition to changes to the forecasts due to changes in assumptions about economic growth and changes in fiscal policy, then we really have no idea which part is small and which part is large, at least not without knowing what those changes are. Do we?

Have I missed a step? I'm admittedly far afield from my area of expertise.

Posted by: Thomas on February 6, 2003 09:20 PM

I fail to see the relevence of Thomas' question. When the Adminstration is pushing further tax cuts, why wouldn't someone assume that revenue decreases and technical changes are small? Wouldn't the other case end all arguments for further tax cuts or even the continuance of current cuts? If revenue decreases and technical changes are that great (that they are the large part), how are further tax cuts wise? After all, the first round of cuts were supposed to be affordable. Don't the impact of deficits on future growth outlined by Gale and Hubbard remain?

Posted by: Fred on February 11, 2003 07:26 AM
Post a comment
Name:


Email Address:


URL:


Comments:


Remember info?