Jason Furman and Peter Orszag write,
Longstanding evidence (e.g. Krueger and Summers 1988) has documented substantial inter-industry differentials in pay—a mid-level analyst may have the same marginal product wherever he or she works but is paid more at a high-return company than at a low-return company. Newer evidence (Barth et al. 2014 and Song et al. 2015) suggests that much of the rise in earnings inequality represents the increased dispersion of earnings between firms rather than within firms. This is consistent with the combination of a rising dispersion of returns at the firm level and the inter-industry pay differential model, as well as with the notion that firms are wage setters rather than wage takers in a less-than-perfectly-competitive marketplace.
Pointer from Tyler Cowen.
I bristle at the phrase “same marginal product.” Modern workers are not widget-makers, and their value inside an organization is not visible to people outside the organization. Indeed, even within the organization, the value contributed by individual workers cannot be calculated with any precision.
I know someone, call him A, who works in information technology at a firm in a buggy-whip industry. One of his friends, call him B, just took a job at Google. Assume, probably correctly, that the difference between their two compensation packages is a lot wider than the difference in their skills. Some possibilities:
1. This is a disequilibrium situation. Information technology workers currently produce more value at Google than in the buggy-whip industry. In equilibrium, A will move out of the buggy-whip industry and go to work for Google.
2. This is an “efficiency-wage” equilibrium, in which Google pays B slightly more than B’s opportunity cost. This enables Google to be highly selective in who it hires and also to give B an incentive to provide top performance.
I am inclined toward (2). But in either case, the value of B’s work is high relative to B’s wage, which raises the question of why Google does not hire more engineers. Perhaps the value of the next engineer would be lower, because of management limitations at Google.
I think that the key factor here is that the collective management talent assembled at Google is scarce. It generates more value that the collective management talent at the firm in the buggy-whip industry.
What I am suggesting is that the value of a firm depends a great deal on collective management talent. This includes the skills of individual key executives as well as the team chemistry among them.
One of the challenges of maintaining a high-functioning management team is that the “tournament” to get to the top can become corrupt. That is, managers can start to get ahead by undermining other managers rather than by exercising better judgment. As this sort of corruption becomes widespread, a firm can rapidly deteriorate. For me, this is one of the most interesting phenomena in the sociology of organizations.
It’s not so simple. ‘Buggy Whip’ companies sometimes have to pay well for IT talent precisely because the work is not overtly sexy, and the smallness of the IT departments may provide a lot of latitude for action (though it varies a lot from company to company). Google, on the other hand, now has something of a reputation for hiring really smart people…and then giving lot of them boring tasks:
http://www.businessinsider.com/google-employees-confess-the-worst-things-about-working-at-google-2013-11
At this point, most of Google’s revenues come from technologies (search, maps, Android) that are quite mature and are experiencing only slow, incremental change. Part of the whole point of the Google -> Alphabet thing was for Page and Brin to be able to turn over the management of the boring part of the company (Google proper) and spend their own time on the fun stuff (e.g. self-driving cars). One of the answers to the question of ‘Why doesn’t Google hire more engineers?’ is ‘Google doesn’t have enough compelling work for them to do as it is — and good engineers care a lot about having interesting work’.
You can make buggy whips anywhere. But, to earn excess returns in tech, it is really helpful to be in Silicon Valley. There is a severe housing shortage in Silicon Valley. This serves to limit access to the highly valuable professional network. This creates economic rents, which are shared by the landlords, workers, and firms.
I just took a line management job where my direct reports are software devs. My employer is one of the biggest modern tech companies. The offer was more than 50% higher than I expected and I expected 50% more than I was making when I stopped two years ago to focus on my phd. Ie: the pay is phenomenal and I’m in the Midwest, not silicon valley.
I would like to believe I will be hugely productive but more importantly, the pace and expectations are incredibly intense. i have never worked so hard before. Vastly tougher than the airline where I used to work. I am going to earn every penny of it. It’s brutal.
Of course… I love it. It’s what I have always wanted. To compare me to the managers that stayed at the airline and to say we have the same marginal value is asinine. The very nature of the work doesn’t compare.
Gentlemen,
If Google is the awesome brave new future, why is it clueless about employing more people and growing even faster ? The problem is either (a) their brilliant new ideas about harnessing technology to solve meaningful problems in society do not translate into real dollars, or (b) their brilliant ideas about harnessing technology do not find a reasonable use of a sub-standard human with lower IQ, lower university degree & lower coding mojo.
In either case of (a) and (b), Google may be an awesome brave new future, but that future does not have any space for most of human beings. So they are back to working in the buggy whip industry. So how long do we have to tolerate the buggy-whip industry ? The answer is so long as we do not fix the problems (a) or (b) with Google.
I cannot speak about google, but I did spend a very long time inside Microsoft, a period which ended 10 years ago.
There was always a very real shortage of people who appeared capable of doing key jobs. The idea that “B gets lots more than A, more than their real differences in output” did not apply. There were lots A’s who could not do the jobs I was hiring manager for, and very few B’s who could.
The world may have changed, but I would like to suggest that even as IT becomes a much more mature, mainline, part of everyday experience field, there are still important edge effects.
Google may be hiring really smart people to do boring things, IT is full of boring tasks. But one can presume that google is hiring really smart people (at high wage rates) to do this because less-than-really-smart people botch these jobs.
It may be like surgery – a lot of it appears to be pretty repetitive and not innovative or exciting, but only really skilled teams of people can do it well.
As noted elsewhere in this blog, very few people are widget makers anymore, their output (and potential output) can be very hard to assess, but to me at least the differences appear to be very real and full of stair steps in capacity and output.