December 12, 2002
What Goes Around, Comes Around

Paul Krugman praises the first (1948) edition of Paul Samuelson's Economics textbook. In Krugman's view, a number of topics dropped from textbooks since as outdated and outmoded are now important once again. He's right:


Vintage 1948: Synchronicity! There I was, rereading Trent Lott's statement about how much better things would have been if Strom Thurmond had won the 1948 election, when I looked at the bookshelf right above my computer - and there it was: Samuelson's Economics: The Original 1948 Edition (available in a new printing from McGraw-Hill, and highly recommended if you're interested in the history of thought in economics.)

Lately old-fashioned Keynesian macro has been on my mind, for a couple of reasons. We (me and my co-author, who also happens to be my wife) just finished a draft of the fiscal policy chapter of our textbook in progress; I also spent last week discussing deflation and liquidity traps in WWS 524. And it has struck me repeatedly how relevant the old stuff seems in the early 21st-century world.

It wasn't that way for most of my career. By the time I got my Ph.D., the monetarists had lost some big battles - even by 1977 it was clear that you couldn't stabilize the economy by maintaining a steady growth rate for M3.14159, or whatever aggregate was fashionable that week - but there was a consensus among economists that monetary policy could and should be the tool of choice for stabilization. Fiscal policy was too slow, too awkward, and anyway would just lead to crowding out.

More broadly, demand-side questions were no longer interesting. Yes, of course, you needed to get the timing of monetary policy right, worry about inside and outside lags, blah blah, but that was for hired analysts at the Fed or the investment banks; it wasn't the kind of thing an ambitious academic should spend time on. Aggregate demand was basically a solved problem, or so we thought. Aggregate supply - why nominal variables seemed to have real effects, whether it was really necessary to pay a high price for disinflation, and so on - was where the intellectual action was. Later still, the focus shifted to endogenous growth and all that.

Did I even read Samuelson 1948 until recently? I think I glanced at it, but it seemed impossibly musty. The way it dismissed the importance of monetary policy seemed ludicrous. "[S]short-term securities do not today yield much more of a return than cash", the book declared  - referring to a bizarre world in which overnight interest rates were barely above 1 percent. Ridiculous, no? And then there was skepticism about the effectiveness of monetary policy. "If banks and the public are indifferent between gilt-edged bonds - whose yields are already very low - and idle cash, then the Reserve authories may not even succeed in  .... bidding down the interest rate ... an expansionary monetary policy may not lower effective interest rates but may simply spend itself in making everybody more liquid" [Samuelson's italics]  And then the book gave primacy to demand-side fiscal policy as a tool of economic management, with hardly any discussion of crowding out. How quaint!

OK, alert and well-informed readers know where this is going. Here we are in late 2002. The overnight rate is 0.02 in Japan, 1.25 in the U.S. In Japan we've had an object lesson in why Milton Friedman's monetary analysis of the Great Depression was so misleading: M2 or whatever is an endogenous variable, and when you're in liquidity trap territory conventional monetary policy can't target broad aggregates. Crowding out doesn't seem like such a big issue when you're at or close to the zero bound. Meanwhile, in the euro zone individual nations have no monetary policy, so stability-pact limitations on fiscal flexibility loom pretty large.

Oh, and sticky prices don't seem to be an obstacle to economic recovery right now. On the contrary, we're pretty happy not to see deflation developing rapidly.

In short, we are now in an economic environment that resembles the one Samuelson wrote about in 1948, and suddenly his economic analysis doesn't look all that old-fashioned anymore.

I don't mean to say that macro has learned nothing since 1948. Of course we've learned a lot. But there's not much in the current policy debate that rises above the level in Samuelson's 1948 edition, and quite a lot that falls below. For example, it's clear that the ruling party thinks that a cut in spending, matched by an equal cut in taxes, is expansionary ...

Posted by DeLong at December 12, 2002 11:15 AM | Trackback

Email this entry
Email a link to this entry to:


Your email address:


Message (optional):


Comments

>>For example, it's clear that the ruling party thinks that a cut in spending, matched by an equal cut in taxes, is expansionary ...<<

I assume Paul Krugman meant a cut in social program spending. By definition, in GOP rhetorics, there can't be no pork-barrel MILITARY spending, however.

Personally, I have always had a problem with assuming that all kinds of public spending results in the same multiplier effect. That's only true in a representative agent model, with production of a homogenous good, and a uniform production technology.

With heterogenous agents with different marginal propensity to consume, however, where the virtuous circle of spending begins can matter a lot, it seems to me. The same remark obviously holds for tax cuts... And an other level of complexity is introduced when we start of think of the implications for the multiplier of, oh-so-out-of fashion, Input-Output matrix coefficients.

I/O analysis was dismissed because it had (in its most primitive versions) to assume constant technical coefficients of production. However, as far as short- to medium-term demand management is concerned, I don't perceive this to be a convincing critique. Is it time to go back to it to figure out what kind of goverment spending and/or tax cuts would most efficiently help us avoid deflation?

Posted by: Jean-Philippe Stijns on December 12, 2002 12:08 PM

I always felt better about plain old Keynesian economics than the rational-expectations nonsense that was all the rage in our day in grad school.

Let's agree that a liquidity trap is a potential issue (although long-term corporate bond rates are a long way from zero).

Let's agree that the balanced-budget multiplier is not negative.

Can we also agree that to a first approximation expected future tax increases are not stimulative?

Posted by: Arnold Kling on December 12, 2002 03:16 PM

>>Can we also agree that to a first approximation expected future tax increases are not stimulative?<<

If you believe in rational expectations, certainly not :)

>>Let's agree that the balanced-budget multiplier is not negative.<<

Aggreed, but what the Bush team does is the opposite of the good-old Keynesian balance-budget strategy: the Bush administration CUTS taxes and LOWERS (social) spending. Balanced budget multipliers (when we want them to help) are supposed to apply to a case where you RAISE taxes to EXPAND spending...

>>Let's agree that a liquidity trap is a potential issue (although long-term corporate bond rates are a long way from zero).<<

Could it be due to a hosts of risk premia? That could still imply limited willingness for banks to lend to corporations... How are corporate bond rates doing in Japan e.g.?

Posted by: Jean-Philippe Stijns on December 12, 2002 10:35 PM


Allow me to beg in advance for indulgence of my profound ignorance of economics. From my untutored point of view, it seems that the "modern" approach could be continued if one were able to offer negative interest rates. Am I correct in this? If so would there be theoretical advantages or disadvantages to such an approach as opposed to more traditionally Keynesian methods?

Ray

Posted by: Raymond Sim on December 13, 2002 09:22 AM


Allow me to beg in advance for indulgence of my profound ignorance of economics. From my untutored point of view, it seems that the "modern" approach could be continued if one were able to offer negative interest rates. Am I correct in this? If so would there be theoretical advantages or disadvantages to such an approach as opposed to more traditionally Keynesian methods?

Ray

Posted by: Raymond Sim on December 13, 2002 09:22 AM

"I assume Paul Krugman meant a cut in social program spending. By definition, in GOP rhetorics, there can't be no pork-barrel MILITARY spending, however." -- I was wondering the same thing. Overall, Bush is increasing spending, right? (Not sure.)

"Personally, I have always had a problem with assuming that all kinds of public spending results in the same multiplier effect." As I recall, the multiplier is less for military spending, but don't know that for sure (either).

"The same remark obviously holds for tax cuts..." I don't understand why the component of tax cuts going to the extremely wealthy is expansionary. Wouldn't it largely be a case of the wealthy paying taxes, versus the wealthy buying Treasury bonds?

Best,

Posted by: on December 13, 2002 10:33 AM
Post a comment
Name:


Email Address:


URL:


Comments:


Remember info?