The Capital Fundamentalists Ride Again

Michael Mandel

In a recent post on the savings rate, James Hamilton of Econbrowser writes

In the long run, a key determinant of national income is national wealth, which is only acquired over time through national saving.

I have no beef with Professor Hamilton, who is a fine economist (and also teaches at the University of California, San Diego, one of my favorite departments).

However, that word 'only' is exceedingly misleading and leads to bad policy.

National wealth arises out of three main sources: Investment in physical capital, investment in human capital, and investment in intellectual capital--that is, the development of new technologies and new ways of doing things.

Investment in physical capital--machines, buildings, etc--corresponds to savings in the national income accounts (what I mean by 'corresponds' is that investment should equal savings, when foreign money flows are taken into account).

However, most spending on education--human capital--is bizarrely counted as consumption (!) in the national income accounts. Most government and academic spending on research and development is also counted as consumption, while most business spending on R&D is counted as an expense of doing business, not as investment or savings. In other words, increases in education and R&D spending can actually reduce the national savings rate, as currently measured, even while they add to national wealth.

More generally, technology-based improvements in productivity can boost productivity growth and national wealth without requiring an increase in the national savings rate.

Here's two nice charts which demonstrate this. The first chart plots the ten year average of national savings against the ten-year productivity growth rate. National savings falls steadily over this 34 year period, but productivity growth first goes down and then accelerates back up to the previous level (the chart runs out to 2004).

Now take a look at this chart. I calculate real net national wealth as the net worth of households, minus the federal debt, adjusted for inflation. I plot the ten-year growth rate of real net national wealth against the national savings rate. Lo and behold, wealth growth accelerates without a commensurate gain in national savings.

This is important stuff. The link between national savings and productivity growth, and national savings and wealth growth, is much weaker than most people realize.

One more table for fun. It turns out that national savings doesn't even have a clear link to capital investment. Take a look here.

Capital investment vs national savings
decade ending national savings rate* capital growth rate**
1970 20.8% 4.5%
1980 19.6% 4.6%
1990 18.1% 4.3%
2000 16.5% 4.3%
2002 16.4% 4.5%
*10-year average of national savings rate
*10-year growth rate of capital services for the nonfarm business sector,
taken from the BLS multifactor productivity statistics

The left column is of course the national savings rate. The right hand column is the growth rate of capital services, which can be thought of as a measure of the growth rate of the capital stock.

And look at that! The savings rate drops over time, but capital growth stays more or less the same.

I'm done now--and ready for the response of the capital fundamentalists.

Before it's here, it's on the Bloomberg Terminal.