Japan, Demographics and National Productivity

Noah Smith echoes Paul Krugman?s suggestion that ?when properly accounting for demographics? Japan had an impressive run in the 2000s.

In particular, they both refer to the following chart

?

Noah continues

Japan had one lost decade, not two.

Now, here?s the puzzle. What caused the Japanese growth speedup of 2000-07?

Remember, during that entire period, Japan was stuck in a liquidity trap.

Properly accounting for demographics is always tricky since in theory folks can alter retirement ages, fertility rates, work-school balance, etc.

But here is one important thing of note, labor force participation by women not only differs systematically between countries and over time, it also differs systematically overtime, between countries. Here is what I mean.

Here are the female participation rates for Japan and the US

The US female labor force participation, is rising over time, and for most of recent history higher than in Japan. Importantly, however, the relative increase over time shows a pattern not easily visible in the graph above.

The relative participation of US women peaks around 2002 or so and then begins to decline. If we restrict our time period to the interval that Noah and Paul looked at, and invert our ratio we get this:

which is strikingly similar to the red line here

A natural interpretation is that a relative increase in the fraction of women participating in the workforce contributed to a relative increase in GDP per working age adult.

This is tricky of course because the decision to participate in the workforce is itself driven by economic considerations. However, its plausible that its driven by cultural considerations as well.

For example, if Japanese households held on to the single-male-wage-earner model for longer than US households then the aging of the Japanese workforce would not have affected the female Japanese labor force nearly as much. Simply put, early cohorts contained few working women who could age out of the workforce.

Thus the total demographic effect of aging is attenuated by the absence of female workers in the older cohorts. So, adjusting for age-demographic effects actually induces gender demographic effects.

That having been said, one could argue that the only reason that the US posted the growth gains that it did was because of the huge rise in female labor force participation from 1970 ? 1995. Its been been a lost half-century for everyone. I think I heard that argument somewhere else actually.

Yet, that having been said, one could argue that the increase in female labor force participation hit the US ? and other developed countries ? institutional and capital structure by surprise. The resulting productivity slowdown from 1970 ? 1995 occurred because the capital structure was not prepared to absorb the quantity of workers entering the workforce.

This thesis might be easier absorb if you consider the interrelatedness of household and market production. A shift from one to the other means that unexpectedly: the roads are to narrow for the fraction of commuters, the houses are too far away from the offices, there aren?t enough restaurants, there aren?t enough daycares.

In general home life is now more complex and capital intensive than optimal and work life is simpler and more labor intensive than optimal. Folks don?t get enough done while they are at work and have too much to do when they get home.

That doesn?t sound to far off the common complaints that folks offered up during the late 20th century.

Source: http://www.forbes.com/sites/modeledbehavior/2013/02/11/japan-demographics-and-national-productivity/

lou dobbs rock salt david letterman march of dimes james randi wargames blake griffin dunk