Friday, December 4, 2009

Growthmanship: The Rock and the Hard Place

President Obama:
Now, if we can't grow our economy, then it is going to be that much harder for us to reduce the deficit. The single most important thing we could do right now for deficit reduction is to spark strong economic growth, which means that people who've got jobs are paying taxes and businesses that are making profits have taxes -- are paying taxes. That's the most important thing we can do.
Yes, what Obama says is correct, as far as it goes. Let us not forget, though, that the "9-point-something trillion- dollar" debt (not "deficit") got racked up on the premise that this year's deficit would stimulate sufficient economic growth to enable "siphoning off" enough of the growth dividend in tax revenues to repay the deficit. The siphoning off phrase comes from NSC-68. We're living in the perpetual growth Utopia conceived by Leon Keyserling: debt requires growth, growth requires deficits, deficits accumulate debt, debt requires growth... The only thing "siphoned off" in this vicious circle is the option of taking a portion of productivity gains as increased leisure.

As Eugene McCarthy and Bill McGaughey pointed out in their 1989 book, Nonfinancial Economics: The Case for Shorter Hours of Work,
"The main reason that leisure is in disrepute among Treasury Department officials is that they can't tax it. A proposal such as the shorter workweek, which would redistribute the burden of work and its income more evenly, would reduce the tax collector's take from a given volume of economic activity. Therefore it cannot be."
Few people recall that the "voodoo" of supply-side economics was supposed to be that it would induce (or compel?) people to WORK MORE HOURS. This was clearly stated by Paul Craig Roberts, Reagan's Assistant Secretary of the Treasury for economic policy in a Wall Street Journal oped published on March 19, 1981.
"For the supply-side policy to work, taxpayers don't have to respond to lower marginal tax rates by giving up vacations, going on a double shift and saving all their income. When you have a work force of more than 100 million people, small individual responses result in a large aggregate effect. If the average number of hours worked per week rises from 35 to 35.5, GNP rises by $24 billion."
Well, folks, supply-side economics "worked" in the sense that it ratcheted up the working hours. It didn't work in the sense of generating a revenue dividend large enough to pay for the lower marginal rates. Oh, well. C'est la vie, eh?

Given the 'choice' between the guns-and-butter swindle of Keyserling and NSC-68 and the Reagonomics supply-side swindle, I'd opt for none of the above. There IS an alternative. But it requires valuing people more over money and time over tax revenues. So it won't happen. Or it will only happen over the dead bodies of most economists. "In any highly developed discipline," McCarthy and McGaughey wrote,
"there is a tendency to become so specialized and refined that its respected practitioners appear to lose common sense. If medieval philosophers counted the number of angels that could dance on the head of a pin, some contemporary economists deal in equally strange and fictitious concepts. To many of them, it would seem, money is reality, while leisure is an empty spot in time devoid of wealth-producing activities. Although U.S. economists in the postwar period have paid much attention to the techniques of financial manipulation by which governments might keep the economy on a prosperous course, they have paid far too little attention to the way people actually work and live. That approach has produced its own set of problems, more than a few of which appear now to be coming home to roost."


wordy said...

In my view the discussion of the worktime issue has several flaws:
a) it tends to be data- and metrics-averse. E.g.: actually we have gone through several decades of worktime reduction - very possibly the sharpest worktime reduction in history. In 1965 or thereabouts the workweek lasted 38.8 hours. Since then it shrank almost linearly to about 33.2 hours. So it's not like nothing has happened on that front. OTOH, there appears to be a lot of infrastructure work that needs to be done in the U.S. - at least to my European eyes. The American College of Engineers says the U.S. needs to build, rebuild or maintain 5 trillion dollars worth of infrastructure. Let's reduce that by 20% on the assumption that there is some special pleading involved, and you still have an enormous amount of projects that exceeds any stimulus proposals by far. It's not clear why people should not perform productive work when there is such work still undone. It's also easier for trade unions to mobilize workers and negotiate worktime reductions when the employment level is high.
b) It looks to me as if the relationship of means and ends didn't get quite enough respect. One thing is clear: worktime reduction is wealth redistribution - upwards in the NAIRU-theorem-guided variant of involuntary unemployment of a segment of the workforce, downwards in the case of a general increase in leisure time. And while it's possible to fantasize about revolution as a means of change, the crucial issue is the abolition of the "money rent" that gets extracted quasi-automatically once the inflation-constraint of economics is reformulated as a government budget constraint that enables the financial sector to help itself to larger profits. (Also see my previous comment on Dean Baker's unhelpful categorization of a "speculation tax" as a contribution towards reducing the government deficit.) Ultimately worktime reduction is an end in terms of the choices of individuals choosing more leisure over high workloads and a means towards reaching higher employment levels, while higher state spending is a means towards satisfying the infrastructure, service and recreation needs of the citizenry as a whole and therefore an end in terms of choosing the form of action most likely to achieve these common goods, i.e., the "public purpose".
c) Please don't indulge in non-sequiturs. Two of the more famous ones are that a gold standard or public ownership of the means of production bring about peace. Well, no. Neither does worktime reduction per se. Nor does it automatically eliminate waste and pollution. We shouldn't even risk to make it seem like we thought it did.

Sandwichman said...

Wordy: "Please don't indulge in non-sequiturs."

You mean like that one? Pot/kettle.

Sandwichman said...

"In my view the discussion of the worktime issue has several flaws..."

Then, please remove me from your view, wordy. You obviously have nothing worthwhile to contribute to contribute to mine.

wordy said...

See, technically "Please don't indulge in non-sequiturs" doesn't imply that I accuse you of having done so already. I don't know why you feel a need to react like you did (i.e., playing the "double-bind" card).
Basically I am interested in finding criteria for the ordering of the relative importance of several major issues. This appears to piss you off. Now I made the mistake to kind of insinuate that you might come across as seeming to claim that you had discovered the holy grail. So let me choose another wording: you appear to insist that the worktime issue provides the one distinction in the political debate that makes the crucial difference and thus reduces the significance of other issues. O.k., let's try a thought experiment and go back to the Black-bill era: at the beginning of the Great Depression there was not only a sharp deflation and very high unemployment. Even average life expectancy fell - kind of like in Yeltsin's Russia, though perhaps only half as bad. In the course of the New Deal, this trend was reversed - kind of like in Putin's Russia. Question: was this due to New Deal policies? Could it be that this reversal would not have happened as quickly in a full-employment Black-bill world without expanded social security and other policies that were enacted as a part of the New Deal and had the consequence of producing deficits and growth?

The point isn't even that I believe that these are questions with all-too-obvious answers. The point is that these questions can be asked and that they show why - if you really want to convince a majority of the feasibility of your concepts - you have to pay attention to some of the points I raise. I don't know anything about how much money went towards medical services before and after the crash that started the Depression. I am also aware of the connection between alcoholism and unemployment, e.g. Then again: as I pointed out, the workweek did, in fact, become significantly shorter over the last few decades. If there was no such reduction in the workweek in the 1950s - I only have data starting in the early 60s -, then your argument has a big problem, because the 1950s were provably much more egalitarian than later decades (Many say the "Great Sort" began in the 70s, but I suspect that you can make a good case that the crucial trend-reversal year was 1965).
There are, of course, alternatives to thinking about this last point. But why evade it? Imagine you had an opponent who was going to claim that you were asking people to choose between shortening the workweek and increasing life expectancy (or perhaps increasing it faster). As I said, I don't have all the statistical data available, but reacting to possible counterarguments by ending the debate might backfire.

And the Peter Weiss excerpt was brilliant. Also, I'd like to hear what others say.

wordy said...
This comment has been removed by a blog administrator.
wordy said...

Four more points:
1) I do think that we could switch to a 35-hour-week without adverse consequences now.
2) I believe that such a switch will only happen as a result of a significant rise in employment. It will be an effect rather than the initial cause of less unemployment (though it would, of course, contribute to a further reduction in unemployment as long as it is not counteracted by "starve the beast"-policies that would reverse the economic cycle).
3) The initial reversal of the current situation, though, will not be effected by means of worktime reduction ceteris paribus.
4) The Black bill envisaged economic sanctions against employers who wouldn't introduce shorter hours (i.e., excluding them from federal, state and municipal business). Dean's proposal offers economic rewards to employers instead (in the form of tax relief). That's a totally different scheme - one that will inevitably be perceived as introducing competition between financing leisure on the one and public investment and social services on the other hand. Once deflation subsides, this perception of competing uses of public funds will even be an accurate reflection of reality. The conclusion would be that the sanction mechanism is preferable - especially at a point in time when an increase in state spending via public works programs and infrastructure improvements would, in fact, mean that businesses would have to pass up possibilities for expansion in order to stick to their old work schedule whereas a tax relief-scheme is an open invitation to gaming the system (firing, reorganizing and rehiring) and would therefore have no advantage over expanded public spending in terms of efficiency and "getting the money to where it's needed".

Sandwichman said...


I assumed your second comment was just a duplicate so I removed it.

Sure, technically, you didn't accuse me of a non-sequitur, in which case you bringing up the issue of non-sequiturs was itself a non-sequitur. That was my point.

I suppose your comment about being data and metrics averse also had nothing to do with me or anything I've written on this blog. I hope it's not too ungracious of me to point out that your data example was fundamentally flawed -- the demographic and sectoral changes in the period from 1965 to 2009 completely overwhelm any 'secular trend' that one might want to infer from the Average Weekly Hours data. "The Workweek" didn't become significantly shorter. Women with children massively entered the workforce, which is not so much a "shorter work week" as a second shift for mothers. Also, a lot more part-time McJobs (without benefits and at lower pay) were created in the service sector. And hey! You know what? Those numbers you cited are ESTABLISHMENT data, not household data. So if I cobble together two part-time, low wage jobs for a total of 50 hours a week to make ends meet and you work a full time job at 40 hours a week, our "Average Weekly Hours" are not 45 hours (90/2) but 30 hours a week (90/3)! Even if you go to the lengths of downloading raw BLS household survey files and extracting the hours data from them (which I have) you end up with some dilemmas. What do you do about the folks who report working one hour a week and those working 110 hours a week? How accurate are the self-reported data?

It's not that I'm data and metrics averse, I am just all too familiar with how readily people grasp at flimsy data straws, imagining them to be the holy grail.

Instead of reciting a bunch of numbers I might be accused of cherry picking, I will give a reference that contains a good technical discussion of the numbers: Barry Bluestone and Stephen Rose, "The enigma of working time trends." It's in Working Time: International trends, theory and policy perspectives, edited by Lonnie Golden and Deborah Figart. Disclosure: I also have a chapter in the book. Warning: instead of one "tell-tale stat" Bluestone and Rose discuss in 16 pages the insights and interpretive pitfalls of a range of sources. Their conclusion is that "the long-term trend of translating economic growth at least partially into shorter hours has been reversed."

I hope you'll excuse my impatience, wordy, but when someone pulls a "gotcha!" with a statistic that they clearly don't comprehend, I sort of feel that "dialogue" is not what they're really looking for.

On your third comment: my position has long been that the "best" time, economically, to reduce the hours of work is when there is relatively full employment and stable growth. Unfortunately, that also happens to be the time that everyone seems to thinks the boom will last forever and preventing future unemployment is the furthest thing from their minds. During a recession, it seems to me that work time reduction has to be combined with some kind of supplementary program, otherwise the belated "work sharing" program takes the blame for what is actually the consequence of a lack of policy foresight. Perhaps the ideal would be if the propaganda campaign, conducted during the recession when people are looking for remedies, had some lasting educational value.