Automation, Differential Hiring Preferences, and the CEO/Worker Wage Ratio

With all due respect to David Atkins, the following bit of analysis is just plain wrong:

While not all of these CEOs are of American companies, a great many of them are. And it hasn't mattered. If the argument about competitive wages and tight bottom lines for American companies were true, then the top of the pay scale should see some tightening as well. Sure, it's not quite as easy to outsource a CEO as it is to outsource an assembly line worker, but it's not that much more difficult. If competition is really that tight, then CEOs and other executives should feel the pinch as well in a global competitive environment.

Well... no. Or, at least, not automatically. The pay rates for factory workers and CEOs, like any other price, are a function of supply and demand. David's thesis would be correct if globalization had an equal effect on supply and demand for both classes of workers, but it turns out that's not necessarily the case. On average, across the globe, it should be a wash (I think), but if you confine your analysis to just one country (America), as David has done, it turns out that globalization may bid up one price while lowering the other.

Consider the labor pools to which American manufacturers had access pre- and post-globalization. In the "pre-" era they had access to the indigenous workforce, whereas in the "post-" era they also had access to large supplys of labor in Latin America, Asia, and elsewhere. From the perspective of an American company labor supply went up, probably by a lot. What about demand for American workers? Globalization, in theory at least, increased the ability of foreign manufacturers to access American labor pools, but the decline of American manufacturing strongly suggests that more manufacturing jobs moved abroad than were created by foreign investment. Thus, we can confidently say that globalization did little or nothing to increase the absolute demand for American line workers. Increased supply and stagnant demand implies a fall in the average worker wage.

Now, what about CEOs? In the global economy American CEOs are competing against all the other smart people from other countries. But what if, when it comes to CEOs, companies have strong differential preferences with regard to nationality? An American company that wouldn't blink an eye at outsourcing manufacturing to China might nevertheless balk at hiring someone of Chinese descent as CEO. Conversely, what if American CEOs are, for whatever misguided reason, seen as having more cachet internationally? If the first condition holds true (and I suspect that it does, being nothing more that pro-American bigotry) then CEOs may be largely protected from competition at home while still enjoying increased demand abroad. If both conditions hold true the effect is magnitude further. Thus globalization may plausibly increase demand for American CEOs more than it increases supply, leading to an increased wage.

But there's another confounding variable as well. Over the period in question (1980 - 2012) automation continued to put downward pressure on manufacturing wages while placing no similar pressure on CEOs' pay. Indeed, the opposite may be true: If CEO pay is tied to profitability, and automation increases profitability, then the adoption of automation may cause CEO pay to rise. If globalization had a equal effect on both classes of workers we'd still expect to see an increase in the CEO/worker wage ratio by virtue of this affect alone.

So David's thesis, while initially plausible, fails to take into account that globalization may affect demand for different classes of workers differently. Additionally, the job duties and compensation packages for workers and CEOs are vastly different as a result of which other factors, such as the increased adoption of automation, which might lead to increases in the CEO/worker wage ratio even in the face of downward wage pressure due to globalization.