Complicating labor markets

Over dinner tonight, I had an interesting interdisciplinary conversation with a sociologist who studies organizational behavior. I thought I’d share just one main portion of the conversation, because it has a takeaway that I think serves as both a compelling and an accessible critique of some classic economic theory. So here goes.

The sociologist I spoke with told me about some of her research on “leadership” in business contexts. I told her that I frankly thought most “leadership studies” were bullshit, motivated primarily by people who had to justify their own high pay and lofty positions without any reference to a tangible skillset. I also told her that I thought the field suffered from a lack of cynical descriptive realism — i.e., if we actually looked statistically and dispassionately at what defined “leaders” in large organizations (senior executives), we would come up with more traits like “over 6’2,” “related to and on friendly terms with lots of other rich people,” and “high levels of hormone x,” & etc. and fewer teachable character traits and maxims fit for best-selling books. She laughed and agreed, but said that there was some actual interesting descriptive social scientific research being done on leaders that wasn’t just mindless boosterism or maxim-hunting.

For example, she said, there’s been a lot of interesting research to pin down what makes star upper-middle level managers in organizations work very well. Think of the person who’s not quite in the c-suite, but whom employees might identify as a motive force behind the actual work that gets done among the mid-level engineers or whatever. Like a very exceptionally good project-manager. What’s so great about these people? What do they have right? Well, one interesting takeaway from the research is that these people don’t, themselves, seem to have anything — or at least it’s not something they have that makes them so exceptional. And the reason we know this is that it’s a very common occurrence to find that, when these people are hired away to other firms, they do not perform exceptionally well in those other contexts. And this indicates that the success of these upper-middle level managers hinged on their embedment within their particular social and organizational environment. So this evidence — that star upper-middle managers often fail when moved to different organization contexts — suggests that probably they were effective leaders not just because of anything inhering in their character itself, but because they’d built up networks and linked chains of trust, credibility, and authority, through a long (and, probably, at times, somewhat lucky and random) cumulative process.

Let me try to illustrate this idea with a story. Suppose Abigail’s been at The Company for 10 years. In year 1, she won the trust and admiration of her co-worker Benjamin through her hard work and personal kindness. In year 5, she got a promotion, which Benjamin viewed as entirely fair, given her hard work. She then became Benjamin’s boss, but they remained friends, and she even helped him through a tough time. Eventually, in year 10, the two of them both moved up another step, and Benjamin got his own hire — now Clarence, an engineer, is working for him. Clarence saw from his first day on the job how highly Benjamim thought of Abigail, and so he immediately started to admire her as well. From day one, he saw how the two seem mutually invested in each other and their shared projects, and so he learned to feel the same way. More, Benjamin, having long enjoyed a good relationship as Abigail’s subordinate, manages Clarence in the same harmonious way that Abigail managed him. Altogether, Abigail, Benjamin, and Clarence make a really great, harmonious team at The Company. And it’s not just about their collective organizational functioning. Clarence is even more creative on his individual projects as an engineer than he was in his previous job, because, now, the smooth and trusting way that he and Benjamin and Abigail all seem to flow together makes him feel rewarded, confident, and at ease — all stimulating his creativity.

So you can see how, in this story, this very functional team isn’t really dependent upon Abigail’s charisma or character traits, or whatever, but on the surprising ways in which her good deeds toward, and hence her good reputation with, Benjamin diffused throughout her hierarchy. She becomes a trusted leader and authority figure not simply by the logic of her own innate abilities, but by the logic of accumulating social network factors that take on a life and momentum of their own.

Now suppose that Abigail gets hired away to another firm, called The Other Company. The Other Company had been considering an internal hire, named Delilah, but eventually passed over the internal candidate for Abigail. So from the moment she shows up, Abigail feels some waves of resentment and hostility from her new immediate subordinate, Delilah, which diffuse down to the team beneath Delilah, many of whom also feel disengaged from their work by their feeling that they have not being given sufficient recognition by senior management, & etc. A vicious cycle of dysfunction, alienation, disengagement, and declining credibility commences, and, at this point, there’s nothing Abigail can really do to stop it.

In the real world these same basic kinds of processes take place through much more complex interactions among much larger networks, influenced by many more factors. But hopefully my story illustrates the basic idea. To put this more simply, it is a mistake to think as if “there are people who are leaders, and we can put them into a context in order that they may lead.” Rather the research suggests that is more accurate to say, “social contexts, themselves, create leaderswho are like nodes in a network that largely has a logic of its own.” In this conceptual vein, it doesn’t even analytically make sense to talk about a leader per se independent of a context. Leadership doesn’t inhere in a person her/himself; it inheres in networks in social and organizational contexts, and just who happens to fit in the right node is probably influenced by a lot of random, unpredictable, and surprising things.


This all strikes me as true, and was perhaps obvious to my readers who are less immersed in classical economic theory than I. But the idea stuck out to me as so marvelous and interesting precisely because it is so contrary to classical economic models of labor markets. I wrote about the basics of the economic theory of labor markets in my post on the theory of the minimum wage. But here’s another brief review: Imagine a competitive labor market in which workers of different skill levels can use capital and technology owned by firms to produce profitable goods. In this competitive market, your wage must approximate your “marginal productivity,” (i.e., the actual value-add of what you produce for the firm that employs you), because if your firm were paying you less than that, its competitors would stand to profit by hiring you away, paying you up to just as much as your marginal productivity. In economic thinking about labor markets, then, your compensation gets set not by any abstract notions of desert, or anything else, but in the same way that the price of any other asset gets set — by the intersection of the demand for and supply of the good. This theory comforts us that we don’t need to get caught up in tricky moral, ethical, or political debates about what various workers deserve, because these questions are answered by competitive markets, with your wage set equal to your marginal product.

But according to the ideas we’ve discussed above, your “marginal productivity” is not a fixed quality of you as a worker, but varies from one social and organizational context to another. And this seriously screws with the economic model of labor markets, in which workers’ wages are kept at the level of their marginal productivity by their ability to take that productivity elsewhere. Think about it this way: Suppose Abigail has a value-add (a marginal productivity) of $200,000 a year to The Company; but suppose that The Company and its competitor The Other Company have both read this blog post and so they know that she would only be worth $50,000 a year in the context of The Other Company. In this case, if these facts are all known, then The Company could get away with paying her only $51,000 a year — depriving her of 3/4 of her value-add. (And this suggests that also in a lot of other less extreme cases, the same basic mechanisms can keep workers from enjoying the full fruits of their productivity.)

It might be more realistic, though, to consider a world in which none of these facts is really known. That is, The Company isn’t sure how to quantify Abigail’s value-add, and The Other Company doesn’t know anything about Abigail either, nor about how she would perform in their work environment. So not only is it not the case that “the value of Abigail’s labor” is a fixed definite quality; it’s also not the case that there’s really any bidding for her labor going on. So what, then, determines Abigail’s wage and compensation in the real world? My conclusion is that the answer must, inevitably, be, “people, and their ideas.” These ideas include abstract ideas about desert, the needs of the employee (people get raises when they have families, even though this detracts from their productivity), etc., etc. Economists might have trouble with labor markets working this way, because it’s next to impossible to precisely model, and hence precisely understand, a world in which some prices get set by “people’s vague, imprecise, and highly biased ideas about what seems about proper,” but an ordinary person would note that this is obviously the way the real world works.


This, by the way, reminds me that I had  a conversation with my girlfriend this past summer, over a two-martini, one-cigar lunch (she’s a babe), in which she basically tried to articulate the ideas I’ve been talking through in this post, and I pushed back, and insistently mansplained the classical theory. So, sorry babe. Public apology here.


This all seems like an important insight to me, but I want to make clear that it’s not immediately obvious what the big takeaways from this line of thinking are. A pet peeve of mine is that many people seem to think that any good criticism of classic market-based economic reasoning immediately makes the case for more government regulation of markets — and they often don’t make the basis of that inference completely clear.

For example, here, one might reasonably argue that the fact that there does not exist a “competitive market” for the value that upper-middle managers create in their local organizational networks, such that these managers are not protected by their ability to move with their feet and so can get paid less than their “marginal product,” strengthens the case for legal regulation of payscales — e.g., forbidding firms to cut the pay of employees with 10+ years’ experience. At the same time, our reasoning here might also reasonably suggest that we would want to make it much easier for firms to fire people — because management’s desire to fire them is probably evidence that they did not have a high “marginal product” in that particular organizational context, and so we as a society should want them to leave and end up in a firm where they’d be better matched and more productive. And even these two arguments are very much on the level of abstract theory, without regard to the difficulties of legal implementation.

To me the major takeaway is that we have by no means reached a world in which markets are so well-functioning that we can now all stand back and allow the logic of competitive markets to make all of our decisions for us. We still need to depend on people willingly choosing to make ethical and moral decisions even when markets and governments aren’t forcing them to. If there is an upper-middle manager who has accumulated enough authority and trust that she is worth $300,000 a year to my company, but her productivity is so locally-embedded that other firms will only pay her $60,000 a year, I could get away with paying her 1/5th of her value-add. But I shouldn’t do it, because that would be the wrong thing to do. And she depends on that. That factor — morality and interpersonal trust — still needs to be there for our economy to function.