Factor-Price Equalization

Related to my previous post on the decline in labor arbitrage in the tech market, it's worth studying the data supporting factor price equalization in this age of greater factor mobility.

Per Arnold Kling:​

Again, I want to suggest that there is a connection between this trend and the stagnation of median incomes in the United States, and even to the decade-long drop-off in employment here. New patterns of trade are developing that are reducing the advantage that a person enjoys merely for being located in the United States. There still are advantages, as evidenced by the excess supply of people who wish to immigrate herte. However, the Great Factor Price Equalization is underway, thanks to the fall of Communism, the rise of the Internet, and sporadic progress in institutional development in the emerging-market countries.

Some of the most compelling evidence for Factor Price Equalization comes from Evan Soltas.​ In the early 1970's, something changed in the U.S. Up until then, "real wages and compensation had increased roughly one-for-one with productivity; since then, productivity has soared with comparatively no change in real earnings".

Taking the U.S. apparel market as a case study, Soltas goes to show convincingly that almost all apparel manufacturing fled to Asia, with corresponding spikes in trade to GDP, exports, and wage increases in manufacturing in Asian markets, all of which factor-price equalization would predict. The charts over at Soltas' post are damning.

Here's further evidence from the Conversable Economist.​ A World Bank report calls our current age, from about 2000 onward, the Third Age of Financial Globalization, and charts on both investment and savings show convergence between high income and developing countries.

Something like this is happening in technology, though it will be quite some time before wages between tech workers in Asia and the Bay Area come anywhere close to being equivalent.​ Even with the existing disparity, though, Chinese developers working for American companies are seeing their wages outpace those of their in-country peers at a dramatic pace.

Within the U.S., it's not a coincidence that so many companies in tech offer similar benefits and perks now: a liberal vacation policy, free food and drink, your choice of computing equipment, the ability work from home when desired, a lax dress code.

When skill converges, luck becomes more important. When the offer numbers and perks converge, getting an edge on recruiting and retention depends on other factors.

Fundamental attribution error and tech mercenaries

In social psychology, the fundamental attribution error (also known as correspondence bias or attribution effect) describes the tendency to overestimate the effect of disposition or personality and underestimate the effect of the situation in explaining social behavior. The fundamental attribution error is most visible when people explain the behavior of others. It does not explain interpretations of one's own behavior—where situational factors are more easily recognized and can thus be taken into consideration. This discrepancy between attributions for one's own behavior and for that of others is known as the actor–observer bias.

From Wikipedia. I've always heard fundamental attribution error used to describe people. But more and more, I believe it's applicable to companies as well.

​That's not to say it isn't also still a problem when applied to people in technology. For example, a common belief is that tech workers in Silicon Valley are more mercenary than in other tech markets like Seattle or New York.

When I moved up to the Bay Area from Los Angeles in mid 2011, I was curious to experience this supposedly venal culture firsthand.

Having worked in all three of those markets over many years now (and also having been at Hulu for three and a half years in Los Angeles), I'm not so sure the people are more mercenary in Silicon Valley. Much of the short tenures in the Bay Area may be explainable by ​the environment rather than some intrinsic ruthlessness on the part of the people living in the area.

For one thing, the number of tech companies in Silicon Valley just dwarfs that in any of those other tech markets. Sure, fewer people left Amazon than I would have expected , but if you wanted to stay in Seattle, where else would you go? To Microsoft? Real Networks?​

The number of startups in the Bay Area also exceeds that of any other tech market by a huge margin. Since startups have such a high failure rate, inevitably it drags down job tenures.​

If you control for those two factors, would the Bay Area still rate as having a more mercenary culture? Someone with access to more data would seem to be able to answer this question quite easily (maybe LinkedIn has enough data to run such an analysis).​ My hypothesis, just based on my personal experience, is that the Bay Area's supposedly mercenary culture is just a technology version of the fundamental attribution error.

If you're operating outside of the Bay Area and feeling fairly secure with your workforce, though, beware. The world is changing, and the fight is coming to your doorstep. For one thing, LinkedIn and other such services have made it easier and easier for other companies all over the place to reach your employees with enticing offers. If you don't think every one of your employees is receiving multiple offers a week, if not per day, from recruiters and headhunters and LinkedIn, you may be living in the 90's.

Think you're safe because your employees don't want to relocate? More and more companies are going to where the people are, opening satellite offices in any market with a good base of talent. Shoppers in many states may not be the only ones lamenting the fact that they now must pay sales tax on Amazon purchases. Competitors are also feeling the hit as Amazon now has free reign to open offices in those states and staff them with abandon. There are three major Amazon offices in the Bay Area already, and they're recruiting aggressively. But the same thing is happening in Seattle, Amazon's home court; Facebook and Google, among others, have opened offices there. 

This is not to mention the fact that some of the best employees can choose to work from wherever they want. It's rare, but not as much as it once was.​ Almost every company I've worked at in my career now has some employees who work by themselves out of some random place. They're good enough, and the demand for their skills so far outstrips supply, that they spend most of their work year in a remote destination of their choice, maybe a home office in their hometown in North Dakota.

All of this is good news for employees, who now have more options as the liquidity and efficiency of the labor market surges. For employers, it's difficult to say whether it's a positive or negative thing. If you treat it as a zero-sum game versus employees, it must by definition be a negative if employees are gaining.

Within the tech sector, though, one might hypothesize that companies that can offer more cash benefit from being able to compete for employees anywhere. Startups who need compete on the uncertain benefits of stock options more than cash now must contend with the tech giants, if even if the startups locate in more remote markets. However, this might just filter out those who who are risk averse and who'd flee at the first sign of adversity.

I began this post as an examination of the myth of the Bay Area mercenary culture, but the larger theme may really be the reduction of labor arbitrage in the tech industry.​ By no means has it disappeared entirely, I am not advising you start your next startup in Kansas City. If Jeff Bezos were starting Amazon.com in today's environment, though, it's not a slam dunk that he'd still choose Seattle.

Speed of employer learning

While some students will be able to go to college only if they receive financial aid and others have the resources to go wherever they want, most fall into a middle group that has to answer this question: Do they try to pay for a college that gave them little financial aid, even if it requires borrowing money or using up their savings, because it is perceived to be better, or do they opt for a less prestigious college that offered a merit scholarship and would require little, if any borrowing? It’s not an easy decision.

“It’s not just the sticker price and the net costs,” said Sarah Turner, professor of economics and education at the University of Virginia. She added, “How likely is it that you will get into medical school or law school or have some other opportunities” if you choose the more prestigious college?

That’s the rational argument. In these decisions, though, emotion often wins out, and it can lead to the slippery slope of excessive borrowing.

From a NYTimes article on measuring college prestige vs. cost of enrollment. Do parents have adequate data at hand to help make such an evaluation when choosing where to send their children to college? Where is the true dividing line between an elite school, where the connections generate meaningful returns, and an expensive but non-elite university, where you're just wasting money?​

There is value beyond just your post-graduation earning power to going to an expensive, non-elite institution, but the party that will bear the debt should have more information up front to decide whether the trade-off is one they want to make.​

One of the complexities of making such a decision is that a student often has no idea what they want to do after graduation and thus must make the decision without clarity on their likely earnings right after school. The difference between becoming a novelist  and going to Wall Street makes a huge difference on the model. To wit, going to an expensive institution if you plan on becoming a chef someday really seems like a terrible idea.

In a recent post at Marginal Revolution, Tyler Cowen noted that GMU graduates now earn higher average salaries than UVA grads, despite the fact that UVA is a more exclusive school by all conventional criteria.

The signaling model, in its simplest, most stripped down form, assumes that employers cannot judge the marginal products of individual new hires but instead pay them according to their credentials.  Yet here we have a case where employers seem quite willing to make a judgment about marginal product and indeed that is a judgment which contradicts data on exclusivity of academic origins.  Once you postulate that employers are willing to make estimates of individual marginal products which differ from the rankings that might be given by “raw ability,” the signaling model is  less applicable.  I don’t want to claim that the wages converge exactly on marginal products, but the credentials clearly are just one factor of many.  Employer judgments of expected marginal products are not dominated by credentials, and you can imagine that after having a worker for a year or two the credentials are even less important as a means of judging prospective marginal product.

Another way to put this point is that the speed of employer learning is in fact fairly rapid, and some of it happens before the job even starts.

There is an analogous depreciation of the signalling value of having a high profile company on your resume. So many folks have now worked at major technology companies like Amazon, Apple, Facebook, Google, and Twitter that the ​value of that signal on a resume has been diluted.

​The speed of employer learning doesn't apply just to people just out of school.

The mystery of Mona Lisa

If we examine the Mona Lisa face, zone by zone, the reason for its mysteriousness becomes clear: there are different emotions expressed in different facial zones.

Her mouth, as everyone has noticed, has a slight smile.

Her eyes are a little sad.

Her forehead is blank and unexpressive.

​From an analysis of the Mona Lisa from the perspective of the social psychology of facial expressions (see Malcolm Gladwell's profile of Ekman and the field he originated, the study of micro expressions).

I wonder if Dan Brown can turn this into a page-turner.

Lobbying: a great (the greatest?) investment

In a striking infographic, the United Republic shows why lobbying is so pervasive: it's an unbelievably effective form of spending (for now I'm linking to the NYTimes hosted version of the infographic as most of United Republic's pages, including the infographic, are 404'ing on me).

​As the NYTimes article notes, the ROI on lobbying dwarfs that of any investment an ordinary citizen might hope to capture.

According to statistics United Republic assembled, the prescription drug industry spent $116 million lobbying for legislation to prevent Medicare from bargaining down drug prices — legislation that enabled drug companies to make an additional $90 billion annually. That amounts to an extraordinary 77,500 percent return on investment. Oil companies, in turn, had a return on investment of 5,900 percent, and multinational companies, 22,000 percent.

​You're not feeling as hot about those shares of Apple or Google you've held for a few years now, are you?

In fact, the ROI on lobbying is so astronomically high that Tyler Cowen wonders why politician's don't demand larger bribes from lobbyists, or why companies don't spend more on lobbying.​ This disparity between the cost of lobbying and its returns is known as Tullock paradox. It's ironic, isn't it, to ask why government isn't more corrupt than it already is?

In Government's End, Jonathan Rauch predicted this would be the logical wall any democratic government would run into: demosclerosis, or the inability of a democratic government to deal with our deep problems because motivated lobbyists spend billions fighting for and maintaining the status quo. In such a situation, only marginal incremental change is possible.​

If you've ever worked in a large corporation you may also recognize that inertia that comes from entrenched groups defending their turf.​

It would be wonderful if we could simplify our tax code, but the prevalence of lobbying makes it unlikely. So many of the odd tax loopholes and shelters and rules are there specifically because some narrow interest group fought to get them into the tax code.

In fact, my variant of the Tullock paradox is why corporations like Apple still have to shelter foreign income from domestic taxes at all. You'd think they'd have lobbied their way to ways to get that income back home without the IRS laying their hands on much of it at all.​