Steve Kates wrote:
>
> I think there should be a Godwin's Law for Economics. Whoever brings
> empirical results into a theoretical discussion automatically loses.
>
> It should not be thought that I stepped back very far when I agreed
> that the failure of the stimulus is not obvious. It's obvious that
> it's not obvious, since this will remain an open and never ending
> debate for as long as economists exist.
>
> But so far as the economic policy side is concerned, there is no
> waiting around for academic economists to decide which way is up. With
> the sequestration in the US and other similar actions across the world
> by those who are trying to manage their economies, this is a debate,
> that for the time being anyway, is resolved. No country in the world,
> with the possible exception of the US, would try to stimulate their
> economies through additional levels of public spending. The recessions
> are not over. Economic conditions are worse than in 2008. But
> increases in public spending are off the table everywhere. If we can't
> even agree on that, then what can anyone ever say that can be a
> foundation for further discussion.
>
>
I think for any law to be useful, there has to be a mechanism for its
enforcement. That is why I despair at Steve's suggestion. Who will
enforce Godwin's Law for Economics? I also think data or empirical
results can be useful in a "theoretical" discussion. After all, aren't
theories supposed to be evaluated with evidence to ascertain their
reliability? I believe a more useful approach to dealing with
"empirical results" used to affirm a certain belief system is rather to
examine the nature of the data used to estimate the results as well as
the methodology employed in constructing the functional form or
estimating equation. On that basis, it is easy (for me, at least) to
dismiss the meaningfulness of estimated government expenditure
multipliers as a basis for belief in Keynesianism, particularly fiscal
stimulus.
Keynes's multiplier argument is founded upon three fundamental
assumptions that turn out to be false: (1) that savings are not spent
but are a withdrawal from the expenditure stream, (2) that government
(and business) expenditures don't depend upon income or savings (even
for a closed economy), and (3) that consumption spending takes a
unidirectional form, like running a relay race -- A's consumption
becomes B's income, then B's consumption becomes C's income, and so on.
Now if one corrects assumption (1) to realize that savings fund business
investments as well as government budget deficits, and (2) that
government spending has to be financed by taxes (paid out of income) and
there cannot be any measured consumption expenditures without any
current production and sales--so-called "autonomous consumption" for the
economy as a whole, then the expenditure multiplier has to be equal to
infinity. But there is also nothing left to multiplier it by. That's
why the government expenditure multiplier EFFECT is zero.
No amount of fooling around with functional forms negates the above
conclusion. There is thus no point, as far as I'm concerned, arguing
with someone who insists on basing their belief in Keynesianism on
estimated multipliers. I published the "mythology of the Keynesian
multiplier" in the _American Journal of Economics and Sociology_ in 2001
and I repeat the point in footnote 20, p. 87, of my modern Ricardian
equivalence article in the _Journal of the History of Economic Thought_
(March 2013). How else can I hope to persuade a non-repentant Keynesian
(who also claims to be a historian of economic thought) of the folly of
such belief? If one introduces central bank new money creation into the
argument, then we would have an explosive multiplier effect on real
income (output and employment) nowhere observed on earth! As Murray
Rothbard once observed, regarding the silliness of the Keynesian
multiplier argument, all government needs to do to create prosperity for
ever is just to find just 1 dollar to spend.
Indeed, I think such "studies" as publicized by Alesina without getting
to the heart of the Keynesian mythology don't serve a very useful
purpose. They are rather a distraction. Aggregate data are generated
by a multitude of factors (or impulses, the favorite language of the
econometric estimators). Without carefully identifying them and
isolating their respective impacts on observed data, no estimation tells
a useful story about the economy. This is what we learn from
econometrics. And this is also why someone once wrote about the two
things he wouldn't like to see in their preparation: sausages and
econometric estimation, the latter because many unsavory things can be
done to generate the end result!
James Ahiakpor
> On 17 November 2013 00:53, Alan G Isaac <[log in to unmask]
> <mailto:[log in to unmask]>> wrote:
>
> On 11/16/2013 8:36 AM, Alan G Isaac quoted:
>
> "The range of the spending multiplier estimated using
> these various approaches is from .4 to 1.5, with some
> estimates even lower than .4 and some estimates larger
> than 1.5. However, most fall in the .4 to 1.5 range."
>
>
>
> If I may offer just one more quote from some people who care about
> the evidence.
> Jordą, Ņscar and Alan M. Taylor, 2013,
> "The Time for Austerity: Estimating the Average Treatment Effect
> of Fiscal Policy"
> http://www.nber.org/papers/w19414
>
> "[W]e have a measure of the multiplier that
> explicitly accounts for failures of identification
> due to observable controls. Our estimates ...
> suggest even larger impacts than the IMF study when
> the state of the economy worsens. ... It appears
> that Keynes was right after all."
>
> As Steve now allows, it is *not* obvious that the fiscal responses
> to the Great Recession invalidate Keynesian claims about the
> role of aggregate demand. Not in the least.
>
> Cheers,
> Alan Isaac
>
>
>
--
James C.W. Ahiakpor, Ph.D.
Professor
Department of Economics
California State University, East Bay
Hayward, CA 94542
(510) 885-3137 Work
(510) 885-7175 Fax (Not Private)
|
