I came across an interesting graph (and some commentary which will prove to be instructive) on Matt Yglesias’s blog:
(Image link goes to original post)
This is another one of those graphs that doesn’t give you a full y axis, so those precipitous drops aren’t quite as precipitous as they appear. In any case, what we see here is that people get laid off — and less stuff gets made* — during recessions. This isn’t a particularly stunning revelation, but as with many trends it’s kind of nifty to see it laid out in graph form.
Yglesias, of course, uses these data to justify Keynesian intervention:
This, incidentally, illustrates why people who say that it’s not possible for deficit spending to stimulate the economy because “the money has to come from somewhere” are mistaken. There are, right now, people not working who could be working. And there is productive capital standing idle that could be put to use.
Once I put a leash on my Rothbardian tendencies, I realized that he’s right — with a few qualifications and assumptions that turn out to be rather important. If the government spends money to give Bob a job, it doesn’t need to take the money directly and immediately from Alice — it can instead either borrow the money directly (increasing debt) or lower central banking rates and do so indirectly. (Both of these increase inflation, which acts as a tax on both Alice and Bob by reducing the value of their money. TANSTAAFL and all that.) This won’t necessarily reduce Alice’s productivity by the same amount that it increases Bob’s — Alice might save more and spend less as the national debt gets more and more terrifyingly large, but giving Bob a job doesn’t put Alice out of one.
The major assumption here is that when the government spends money on Bob, he increases productivity — by which I suppose we mean creates value — faster than inflation, rent-seeking, and other agents of friction and entropy destroy it**. This is the famous Keynesian multiplier, discussed somewhat further in this post on Marginal Revolution. Keynes built a hypothetical in which the multiplier was around 10; more recent estimates put it at 1.3 or so, but as long as it’s greater than one we should have a net benefit.
At least, we should have a net benefit for the short term. This is the major qualification I mentioned: deficit-financed stimulus spending is very much a short-term, time-varying thing. Tyler Cowen explains:
Say you have a debt-financed increase in government spending. You can get some dollars out of low-velocity pools into high-velocity pools on the first round of redistributing the spending flow. Do not expect complete crowding out and so nominal aggregate demand can increase, thus boosting output and employment. But the second and third round effects of the redistributed money are usually a wash and the boost to velocity dwindles. Why should it stay in a high-velocity sector of the economy?
So the short-term effect of Keynesian spending may be increased short-term productivity, depending on what the economy looks like when you do it. Bob gets a job building interstates or coal-fired powerplants. But in the long term, we’re left with a lot of debt and a thundering herd of public institutions which might once have been useful for passing out the stimulus bucks but now justify their existence only through rent-seeking and relentless lobbying. What happens then?
Yglesias may have found the answer to that one, too:
At the same time, call me crazy but isn’t there a long-term downward trend in this data series? Why would that be?
At a wild guess, I’d say it’s because more and more capital is tied up in debt service and porcine government.
——
* It occurs to me that whether “less stuff gets made” depends on precisely which capacity is lying around unused. In the sense that fewer widgets are being built from widget-components, I can take “less capacity used = less stuff gets made = less value produced” pretty much at face value. On the other hand, not all capacity is equal: some companies (Google, say) create a lot of value for every dollar they take in, and others (Chrysler and AIG) actually destroy value. In that sense, we might be better off if some “capacity” lay around “unutilized”.
** This is disturbingly analogous to buying an investment and expecting it to go up. We’ve seen a shit-ton of finger-wagging over heavily-leveraged banks; why not apply the same language to governments?
