Monday, November 15, 2010

Productivity Growth: Why Japan's Will Remain Low

This is the initial sketch of a line of argumentation. I will return to it periodically over the next several weeks.

Japan's baby boomers are set to retire as they hit age 65 -- unlike Europe and the US, 77% of men below that age remain in the labor force (and 50% of men age 65-69). [see the post below on women's LF participation for the distaff side of the story] That will have many side effects, from the labor force (boosting the demand for younger workers) to capital markets (decreasing private savings with the potential to increase interest rates and exchange rates). These are perhaps best analyzed by focusing on the demand side. (I think that is best done with the assistance of a DSGE model to help highlight the multiplicity of channels for interaction, even if we should not rely on such models as more than loosely suggestive, they're simply to complex to be sure what assumptions matter, and in all cases force offsetting adjustments that are both too strong and too smooth to be credible because to solve them requires that they convergence to a steady state.)

Alongside such shifts are long-run supply-side trends. The obvious one is the shrinking size of the potential labor force, only partially offset by the ongoing increase in female labor force participation. Investment surely won't cease, and the likely rise in capital per worker means that GDP per worker will continue to rise. The net of the two will be a small number, perhaps 0. What happens to productivity is thus crucial to aggregate growth and important for per capita and per worker growth.

Among the better-off end of members of the OECD (the rich countries' club), that has converged strongly around 2% pa. Why should Japan be different?

I argue here that it will be significantly lower for two principle reasons. The first is that young workers will be a smaller share of the labor force, and that most of the gain in human capital comes at younger demographics. Part of that is the standard diminishing returns argument. However, I think that it is also difficult for those who are older to change. That is because at higher incomes the opportunity cost of learning (or retraining), which I believe to be a time-intensive process, is also higher. Can old dogs learn new tricks? -- I like to think so. But will they? -- I see lots of reasons why they would choose not to.

A second reason is that at the firm level higher productivity is realized only in modest part by organizations changing their structures and product lines in an organic manner. Instead the bulk of gains are via exit of those low in productivity and entry of new firms (or business units) that are high in productivity. Now in a dynamic, growing economy the two are only loosely linked: with growing demand, you don't have to have exit in order to have entry. The old order can simply fade into obscurity [wanted: an appropriate aphorism]. Not so in Japan today.

Let me use real estate and retailing as examples. In 2007 the local express train station in the Tokyo suburb where I lived continued to have a camera shop; so did a shopping street (shotengai 商店街) where I had lived 25 years before. But by then film cameras had been relegated to a small coterie of hobbyists; these two stores were less than busy. Why did they not exit? -- in a slowly growing economy (with a falling population) the opportunity cost of doing so is high. In a normal world, as we're used to thinking of things, you could turn your store to some other line of business, or sell it to someone who wanted an office or even house close to a station. In the Japan of 2010, such business sites have almost zero market value.

Now in this particular example the size of the shops is small, no bigger than the office in which I sit while I write this. Even in a world of small-scale retailers they are on the small end of the spectrum. Casual empiricism suggests however that the same issue affects retailing and office space as a whole, even if not as strongly. In a market that is moving about as fast as its aged residents (25% over age 65), the benefits of exit are sharply diminished. That however impedes the "churn" of new for old firms, particularly if the new need to invest in new structures and new locations and new types of capital. Inevitably they would be marginally better -- but the word "marginally" is the operative one. When the old refuse to retire, the young have less room to maneuver.

What I have sketched is an exercise in pure logic, not (yet) backed by data or models other than my informal prose one. The logic can surely be extended to other margins of adjustment, including capital investment. Of course many factors affect productivity independent of the above. My own sense is that they reinforce rather than offset the opportunity cost story above. One possibility is that the rise in contingent labor (part-time and contract work) among the young is a prelude to a "lost generation" who will not find positions as the "boomers" retire and so will not benefit from the to-date-normal process of human capital accumulation with tenure. Ditto the shift in schools, with the 5.5 day school-week giving way to "yutori" (feel-good!) education that will not position Japan's youth for a lifetime of learning. My sense is that health care has also entered an era of diminished returns, so that the needs of the boomers in that area will detract from the ability of the economy to provide goods and services to those who are younger, who will literally be a generation working to support their elders as much as themselves. But offsetting this are Japan's foreign assets, which will allow the country to run a (modest) trade deficit forever. To put it in an archaic manner, the Japanese economy in its senescence can clip coupons.

Let me end this initial draft with my own sense that, however speculative the line of argumentation, I will in fact be able to tie it to empirical indicators that will undergird its validity. But what of magnitude? My hunch is that Japan will be looking at productivity growth in the range of 1.0%-1.5%, that is, up to a full percentage point lower than other OECD countries. If other factors repress investment -- crowding out as falling savings meets still-large deficits -- then this implies that Japan will see average GDP growth decline to under 1% pa. Now today's young will face higher tax rates to pay for their parents retirement (surely a more equitable alternative than forcing children to provide directly for their parents, but that's a different issue...). So per worker output will not grow, per worker income will fall. Japan won't cease to be prosperous, assuming that the politics of governing this process don't go severely awry.

Ha Japan's sun already set?
As an economist, that's too bad, because Japan is in the vanguard of the aging world, and these sorts of issues cry out for analysis. Unfortunately economics as a field is however not immune to market forces, and with Japan out of the limelight it will be very difficult to marshal the resources needed to do that sort of work.

Mike Smitka