The wage premium at large firms

Do very wealthy CEOs yank potential wage gains away from median-like workers?  The excellent Adam Ozimek writes to me:

You say: 6. If the top earners are screwing over their wage earners in the big companies, by pulling in excess wages, options, and perks, we should observe non-stagnant median pay for people who avoid working in firms with fat cat CEOs.  Or we should observe talented lower-tier workers fleeing the big corporations, to keep their wages up.  Yet no evidence for these predictions is given, nor are the predictions considered.  It is likely that the predictions are false.

And in fact isn't this precisely the opposite of what the evidence on the employer size wage-premium tells us? If large firms were better at keeping wages down, then the employer size wage-premium would be negative, since small firms would pay more for comparable workers. Apparently this has been true for a long time, for instance from this paper http://gatton.uky.edu/Faculty/Troske/publish_pap/restat_sizewage_feb99.pdf

The fact that large employers pay higher wages than small employers has long been recognized as an important component of the variation in worker wages.This phenomenon was first documented by Moore (1911) and later confirmed by King (1923), Mellow (1982), Oi (1983), and Brown and Medoff (1989) among others.

That paper also argues:

Davis and Haltiwanger (1991) show that the gap in real hourly wages between production workers in plants with 20 to 49 employees and production workers in plants with more than 5,000 employees increased by 79% between 1963 and 1986, and that the gap for nonproduction workers in these same plants increased by 49% over this period.

So the firm size premium is growing. This seems inconsistent with their story.

That is all connected to my earlier review of the new Hacker and Pierson book.

Comments

Can you really isolate the "firm size" effect though? Larger firms will tend to be more productive and, presumably, more productive firms offer higher wages. If the gap in wages between small and large firms has grown, I would hazard a guess that it reflects an increase in productivity dispersion, rather than purely "wage" dispersion.

If you spread a large number (executive compensation) over a large employee base and thus aggregate salary base, what else would you expect to show other than that there would be no effect on anything? Divide a large number by a very, very large number and what do you get. Would you even expect to show that executive compensation stood out. Or, have you thought about the old scam, take a nickel from everyone, and you make a lot of money.

Frankly, overpaid execs are probably not taking from employees as much as they are from shareholders.

We should outsource executive management to foreign countries--they do much more for much less. I've worked with Fortune 500 executives, and they are overpaid. What they are good at, though, is corporate infighting to rise to the top. Classic internal warfare stuff.

And, some of the best executives I have worked with are high tech mid cap business whose executives actually create something before their company is acquired by a Fortune 500 company. They then go on to create another company. Those guys deserve and get good pay IF things work out. They are typically shunned by the Fortune 500 firms and do not fit well into those organizations before leaving.

Frankly, overpaid execs are probably not taking from employees as much as they are from shareholders.

Of course.

Also,

Davis and Haltiwanger (1991) show that the gap in real hourly wages between production workers in plants with 20 to 49 employees and production workers in plants with more than 5,000 employees increased by 79% between 1963 and 1986, and that the gap for nonproduction workers in these same plants increased by 49% over this period.

Those are spectacular percentages, no doubt, but what was the size of the gap to begin with? A 79% increase in a ten-cent gap doesn't mean much.

Besides, you write,

And in fact isn't this precisely the opposite of what the evidence on the employer size wage-premium tells us? If large firms were better at keeping wages down, then the employer size wage-premium would be negative, since small firms would pay more for comparable workers.

But this doesn't follow from the paper you link to. As I understand it, from a quick skim, the paper investigates possible explanations for the gap that stem from inherent characteristics of large firms that may make their workers more productive - better job matching, better capital, etc. It says nothing about CEO pay. If you were going to use this evidence to refute the "screwing over wage earners" theory of CEO pay, you would have to compare wages after controlling for these factors.

I'd guess this reflects monitoring. Small firms can monitor their employees more easily to assure performance. Large firms cannot so they need to pay for premium talent that does not require as much monitoring. Just my guess.

It seems that larger companies are better positioned to outsource jobs overseas, thus taking the lower paying jobs out of the equation. Wouldn't this skew any analysis?

You use a paper that examines the period between 1963 and 1986 to support your position on the phenomenon of excessive CEO pay , which had barely even begun its rise by 1986 :

http://2.bp.blogspot.com/_6yQkaL8M87U/Rh4p2Tgo_6I/AAAAAAAAAAc/iHcaERMkKQY/s320/ceo+pay.gif

When you have a preconceived notion to support , I guess inapplicable data is better than no data .....

Some of the comments also remind me of the debate in industrial organization economics on whether firm size correlates with concentration or profitability, whether firm size reflect efficiency, better use of internal markets, ability to diversify r&d risk, yada yada yada.

To see firm size and CEO comp being used relative to employee comp to prove anything is striking, given how the IO economists dispute the relationships between firm size and many of the variables I just mentioned.

There is one limiting factor to any correlation: you can't have high exec comp in a company with low sales and few employees; but the converse is not true: you can, and most often do, have high exec comp in firms with high sales or capital. One doesn't cause the other, however.

If this is true. should the government have programs to help small companies and tax them less than large ones.

as a participant in industrial activities in large organisations i may add that at least in the manufacturing activities that i am familiar with the direct labour cost part of total product cost may be lower in larger companies / factories. in heavy equipment manufacturing, the final assembly labour cost is typically in the lower single digits as a percentage of standard cost of sales (and consequently an even smaller fraction of total costs) why the pressure to keep labour costs down may be less than in a smaller company which tend to be less automated and produce less complicated products where the direct labour cost has a greater impact on profitability.

I second Geoffrey. It appears to me that in many cases the firms with very high CEO salaries are those that are steadily reducing their US workforces through outsourcing (in which I'll include offshoring). Firms that do this end up consisting of just the top of the organizational pyramid, and often need more higher-level people to oversee the outsourced work.

Comments for this post are closed