More fiscal implications of a rising capital-share of income

Paul Krugman has been talking capital-bias and rising profit-shares recently. I was going to write about the implications of this for Social Security and Medicare, but he got there first. Below, I put some (very rough) numbers on how much a rising capital-share of income impacts the current financing of these programs.

The broad issue in a nutshell is that a rising share of overall income in recent years (even decades) has been accruing to owners of capital rather than to workers (or, if you like, accruing to owners of physical and financial capital rather than to owners of human capital). I might immodestly note that there are substantial sections on this topic in both the wages and the incomes chapter of The State of Working America, 12th Edition.

For now, I’ll just talk about some of the interesting tidbits from State of Working America and then sketch out one implication of these rising capital-shares for current fiscal policy debates.

First, the rise in the capital-share (again, the share of overall income claimed by owners of financial capital) really does seem to be happening. In State of Working America, we generally focus lots of attention on the corporate sector of the economy—a sector that accounts for about three-quarters of all private activity. For technical reasons, looking at trends within the corporate sector gives the clearest picture as to whether or not there really has been a substantial shift away from labor and toward capital owners. So has there been such a shift? Definitely. The figure below marks the capital income share at each business cycle peak since 1979.

Update: Already violated my own rule-of-thumb in saying you should look at the corporate sector, but illustrating the point with a figure that doesn’t do this. Here’s what the capital shares in the corporate sector look like since 1959—looking at business cycle peaks. I include 1995 because the labor market boom in the late 1990s provides a temporary reversal of the longer trend, and wanted to show just how sui generis that period was.

Second, this rise in the share of overall corporate-sector income claimed by capital owners (the rise in the capital-share) has not been driven by a rising capital-intensity of corporate sector production. In fact, the capital to output ratio has been flat or slightly falling over most of the 1979 to 2007 period, and this includes the extraordinarily rapid rise in measured investment in the late 1990s during the information and communications technology boom. So, capital owners are not contributing more inputs into the productive process as a share than they used to, but they are reaping a larger share of the output. What does this mean? That the return to each input they contribute has risen. In the jargon, the rising capital-share of income has been driven by a rising profit rate—how much profit capital-owners get per dollar invested, or, more precisely, profits divided by a measure of the corporate sector capital stock.

This figure illustrates the non-role of a rising capital to output ratio with a (too) simple exercise: it holds the profit rate (again, capital income divided by the corporate capital stock) constant at its 1979 level and shows what the capital income share would have been given (just) changes in capital inputs as a share of overall production. The answer is clear: The growth of the capital stock relative to output has contributed very little to the rise in the capital income share in recent decades, and it is the profit rate which has driven this. As Krugman notes, this rise in the profit rate could be a signal of rising corporate power, either driven by growing corporate concentration, or by a continued improvement in their bargaining position vis-à-vis workers driven by anti-worker policies, or some combination of both.

Third, when you include capital gains, the shift away from labor incomes towards capital incomes plays a significant role in driving the rise in overall income inequality. Basically, between 1979 and 2007, this labor-to-capital shift explains about a third of the overall rise in the share of total income claimed by the top 1 percent. Given that this overall rise in inequality has been enormous, one-third isn’t peanuts.

Lastly, since one is not allowed to write about economics these days without weaving in something fiscal-related, it’s important to note that this shift from labor to capital incomes has non-trivial impacts on the revenue streams that fund the key social insurance programs—especially Social Security.

It is well-known by now that inequality within the distribution of annual earnings has eroded the taxable base of Social Security. At the time of the 1983 reforms, the “cap” on annual earnings that were taxed was set to capture 90 percent of all economy-wide earnings, and to rise with the rate of average wage growth. But since wage-growth above the cap was so much faster than the average in subsequent decades, more and more of overall earnings are spilling over the cap; by 2011, only 84.2 percent of economy-wide earnings were subject to the Social Security payroll tax in 2011. This rise in earnings inequality basically robbed the Social Security Trust Fund of more than $650 billion in assets by 2011, and restoring the cap to again cover 90 percent of earnings would do away with about one-quarter to one-third of the projected 75-year shortfall. Widespread recognition of this problem has led many to advocate either readjusting or even outright abolishing the cap on Social Security’s taxable base.

What’s less well-known is that the shift from labor to capital incomes just within the corporate sector has probably had an effect on Social Security’s finances about 50 percent as large as this effect stemming from wages spilling over the cap, as income that shifts from labor to capital owners is no longer in Social Security’s tax base. Staying within the corporate sector, we can calculate how much economic output would have ended up as labor compensation instead of profits had the capital share of corporate sector output remained at its 1979 level going forward. Assume that 90 percent of this increased compensation came in the form of wages, and that 90 percent of these wages were subject to the Social Security tax, and it turns out that the Social Security Trust Fund would have roughly $365 billion more in assets by 2011 (and the Medicare Part A Trust Fund would have about $100 billion more).

And in just the latest full-year—2011—Social Security lost about $45 billion in revenue that it would have had if the Social Security tax applied to 90 percent of all earnings (as it did in 1983), and another $47 billion that would have been collected as payroll taxes had labor compensation’s share in corporate sector output remained at its 1979 level of 74.4 percent instead of the 68.0 percent that actually characterized that year. Lastly, it should be noted that this only includes potential taxable labor compensation lost in the corporate sector due to rising profit-rates; the economy-wide numbers would be even larger.

In 1993, the wage “cap” on Medicare contributions was removed. And in 2013, as part of the Affordable Care Act (health reform), the Medicare contribution will apply to all income, not just wages. In short, both dimensions of inequality that could potentially erode Medicare’s tax-base have now been removed by legislation, yet Social Security contributions are continually undermined by rising inequality. If reforms to the social insurance programs are demanded, it seems that restoring this eroding Social Security wage base would not be the worst place to start.

But yet again, all of this shows that the broader economic context has great ramifications for how we choose to make fiscal policy decisions, and The State of Working America, 12th Edition can greatly help in understanding these details.