The new arch capitalist is … your grandma

Posted on November 3, 2015 in Debates – Full Comment
November 2, 2015.   Stephen Gordon

There aren’t any fixed constants in economics, and most of the statistical regularities identified by economists disappear when they are used as a fulcrum for policy. For example, the seemingly robust relationship between inflation and the rate of growth of the money supply disappeared once central banks tried to use the link to control inflation.

But one stylized fact did seem stable. In a 1957 study, economist Nicholas Kaldor noted that the shares of national income that went to labour and to capital were roughly constant over time. In other words, expressed as a share of GDP, total wages and salaries paid to workers stayed more or less the same. In the short and medium term, this ratio would increase during economic downturns, because even if labour income generally falls during recessions, investment income drops even more. And as the economy recovers, profits grow faster than wages, bringing workers’ shares back down. But over the long term, movements induced by the business cycle would look like fluctuations around a constant labour share.

Because more people are able to earn wages that allow workers to save for their retirement, the median pure capitalist is now a retiree, not a plutocrat
This pattern persisted in the decades that followed Kaldor’s original 1957 study and economists grew to rely on it, to the point of having constant labour shares built into most models of the economy. There was remarkably little discussion behind this assumption; the labour share of income was the dog that did not bark during the profound structural changes of the second half of the last century. Economists are usually preoccupied with explanations for how and why things change, and spend rather less time wondering why they stay the same.

To be sure, there was — and is — much discussion about the effects of technical change and globalization on how national income is distributed. For instance, workers in sectors that benefited from new export opportunities were better off, while those in sectors exposed to foreign competition did worse. Similarly, for some workers, new technologies meant higher productivity and incomes; others saw their jobs disappear as more tasks became automatized. But these gains and losses among workers cancelled each other out, leaving the total labour share unchanged.

Recent U.S. data are forcing economists to revisit their thinking. Ever since the spike during the U.S. recession in 2001 (remember, the share goes up in recessions), labour’s share of income has declined significantly. This decline is to be expected during an expansion, but the U.S. labour share did not jump up during the 2008-9 recession and has remained flat ever since.

This decline is discernible in Canadian data, but it is much less apparent than it is south of the border. For one thing, the current labour share is similar to what it was 15 or even 30 years ago, and it did follow the usual pattern of rebounding during the recession. But if you were to draw a line through the Canadian data over the past 35 years, you’d see a downward trend.

It is customary to greet almost any new development in the economy with alarm, and the decline in the labour share is no exception. The decline has occurred in conjunction with an increase in the share of income going to the top end of the income distribution and many have gone on to conclude that the shift of income from labour to capital is at least partly to blame for increasing income inequality. Old habits die hard: capitalists are still equated with high incomes and workers are still equated with low incomes.

To be fair, those habits were ingrained by centuries of economic history. Up until very recently, the only way to generate high incomes was to own a lot of capital and wages were so low that it was almost impossible for workers to save much. In a world where wages are barely above subsistence levels and in which only the rich owned capital, a shift of income from labour to capital will exacerbate income inequality.

But that’s not the world we live in now. The high earners in the U.S. and Canada are, for the most part, working for a living: their surging incomes are driven by the salaries they earn, not the assets they hold. If there is a link between profits and the high salaries earned by certain corporate executives, the story directly contradicts the narrative in which the shift of income from labour to capital increases income concentration. If corporate executives are using their position to siphon off profits before they can be distributed to shareholders, this reduces the share of total income going to the owners of capital. The problem would be a labour share that is too high, not one that is too low.

In any case, capitalists aren’t who they use to be. Because more people are able to earn wages that allow workers to save for their retirement, the median pure capitalist — that is, someone who does not work and who derives her income from her asset holdings — is now a retiree, not a plutocrat. Plutocrats still exist, but any policy designed to suppress investment income is going to generate significant collateral damage among the elderly. And who’s to say that the increased labour income won’t simply be appropriated by the one per cent?

While the decline in the labour share of income may still be a puzzle that requires an explanation, I’m not yet convinced it’s a problem that needs solving. Demographics seem to be the key for understanding why the labour share is falling, and for why the decline should be welcomed. In an aging population, a constant labour share of income means the steady immiseration of retirees.

< >

Tags: , , ,

This entry was posted on Tuesday, November 3rd, 2015 at 1:31 pm and is filed under Debates. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.

Leave a Reply