Control over skill realization: A response to Fix in RWER (2019)

[Jeremy Zerbe]

Blair Fix’s “An Evolutionary Theory of Resource Distribution” (Real-World Economics Review, 2019) provides a compelling, wide-ranging theory of social power and its effects. Near the end of the article, however, as he rightly assails the assumptions of human capital theorists about the relation of “hidden skills” to hierarchical positionality, I believe he hits a wrinkle in his richly laid-out tapestry of human competition and collaboration.

This wrinkle is not one that undoes his thesis. Rather, when smoothed out, I believe it lends even more credence to his characterization of the power ethos so central to our modern corporations. In the following response, I propose to do some of that smoothing out, picking up from his critique of the “unobserved heterogeneity issue” (Acemoglu and Autor, 2011).

—

Is a growth or loss of skills reflected directly in an individual’s professional promotion or demotion (and consequent rise in or loss of income)? No, of course not. As Fix establishes, the timings just don’t work out. Skills take time to develop while promotions happen instantly. Skills deteriorate gradually while demotions often take place without warning.

But this is an oversimplification of the relationship between skill-change and hierarchy-change. After all, isn’t a promotion the realization of pre-developed skills? Isn’t demotion an “unrealization” of skills, deteriorated or not, that the firm no longer deems useful?

To be promoted, an individual generally attempts to grow skills that can be realized through promotion and the expansion of their professional power. Until promotion, many of these skills go unrealized because there are no outlets to realize them. Call these outlets “professional responsibilities.” These responsibilities may be directly related to an individual’s ability to wield power over (or inspire obedience in) subordinates. They may also be related to the individual’s mastery over aspects of their established professional discipline, or to aspects of disciplines other than their own.

This is why, in some firms, titles like “lead” exist, for individuals who focus on growing their depth of current skills rather than the breadth necessary to become a “manager” or “director.” Fittingly, pay increases are generally smaller for promotion into a “lead” role than they are for promotion into a “manager” role, which explicitly expands an individual’s control over subordinates. When being demoted, it is reasonable to surmise that these skills are being deemed unnecessary to the utility of the firm. Through the decrease in hierarchical position and income, the utilization of these skills (embodied in the individual) is being unrealized—usually forcibly so. Unlike promotion, individuals rarely seek demotion.

That all of these decisions (of promotion and demotion; of increases and decreases in income; of not just the realization but the quantification and even the very evaluation of professional skills) are made by those up the hierarchy about those down the hierarchy is evidence enough of hierarchical power distribution in a firm—evidence ultimately borne out in Fix’s empirical research into the distribution of income within the firm. It doesn’t need the damning timing “mismatch” that he identifies.

This is the essential difference between the unfree corporation and the free association. A corporation might provide mechanisms (beginning with the initial job interview, and continuing through one-on-one check-ins with a manager, annual performance reviews, etc.) for the individual to advocate for their skills and, therefore, hierarchical position and income. However, the ultimate decisions about that individual’s skills and position are made on the terms established by the corporation—or more specifically, the corporation’s hierarchy. In a free association, an individual is treated as the ultimate authority on their skills, and other associated individuals (whether working above, below or alongside the subject) act as the legitimizing agents of their skills, responsibilities, position and income through a process generally referred to as democracy.

Even if the “hidden skills” that human capital theorists must believe in to explain the positions and movements within a given hierarchy were to exist, these skills would not account for an individual’s relative lack of input about the skills with which they themselves are most intimately familiar and control over how those skills are utilized. The belief in these “hidden skills” is merely capitulation to the logic of the corporation’s unaccountable hierarchical structure: a belief that those above can and do truly understand, assess and compensate those skills simply by the nature of their position in the hierarchy.

If that were the case, these “hidden skills” wouldn’t be hidden at all. They’d be easily explicable, measurable and even testable. Not only could we potentially see how often a manager is right or wrong in their assessments of subordinates, we would likely see this reflected in the manager’s own position in the firm’s hierarchy.

But this is not how managers are assessed. While a constant revolving door of subordinates may constitute a dark mark on a manager’s record, it is by no means make or break. Just like the successes of a manager’s subordinates are defined by their obedience to the manager, a manager’s success is defined by their obedience to the individuals even further up the hierarchy. If the company or industry culture is such that turnover rate is deemphasized, a manager’s “hidden skill” at assessing subordinates is also deemphasized. Think of fast food service, telemarketing and other industries that not only ignore a manager’s inability to successfully assess and retain talent but rather develop systems that make constant turnover less disruptive.

So if these hidden skills don’t exist (or at least are explicable as observable skills), where does this leave us? Back with Fix’s concept of the modern “power ethos” in which people are empowered to act according to their social influence. That influence allows them to control when the skills of subordinates are realized and unrealized, regardless of when the subordinates actually develop those skills. Decisions about the realization of skills are made according to the mission of the firm, or more specifically, the logic of the corporate hierarchy. If those skills are “not needed” by the firm, then the individual will not be promoted and their professional responsibilities will not be expanded—regardless of how useful the skills may be agnostically.

It’s in this hierarchical decision-making process (and not simply the promotion or demotion of individuals within a hierarchy, or even their increased or decreased income) that the true control inherent in the power ethos is made apparent—a process that readily and regularly eschews useful skills and their potentialities in service to the “needs” of the firm, dictated by those at the very top of the hierarchy with the least information about the firm’s members but the most power over them.

Topic

[Blair Fix]

A quick note, Jeremy, to say that your comments are well received. They highlight many of the problems of understanding causation in a social environment, and are worth a detailed reply. I hope to have one by next week.

[Blair Fix]

Thank-you, Jeremy, for you well-thought-out response to my work. I accept most of your arguments, so I’ll use this response to outline some related ideas.

As I’ve started to popularize my research on hierarchy, a common response (by those who are skeptical of it) is that hierarchical rank simply indicates skill. People with more skill are rewarded with a promotion, where they then earn more income. I’ve had some time to think about it, and I’ve realized that this thinking ties into deep-seated ideas about meritocracy.

To get started, let’s formalize this meritocracy reasoning using a causal diagram:

skills ⟶ hierarchical rank ⟶ income

Skills cause (greater) hierarchical rank, which then causes (greater) income. So on it’s own, skill does not cause income. Instead, skill must be ‘realized’ though a promotion.

I agree that my ‘timing’ argument does not dispel this meritocracy model. Skills can build slowly and then be rewarded suddenly with a promotion. As long as ‘skills’ are the criteria for promotion, people can claim that ‘skill caused income’.

In a sense, they are right. If we take the ‘meritocratic’ corporate hierarchy for granted, then gaining ‘skill’ is the path to earning more money. Or at least, gaining the credentials that people think indicate skill is the path to earning more money. Hence the need to get an MBA to get into management. Never mind what you actually learned in your MBA …

For many people, this thinking is enough to satisfy their curiosity. It tells them "if you want more money, do x." But from a scientific standpoint, it’s unsatisfying. It doesn’t tell us why skill (however defined) can only be transformed into income when it is ‘realized’ by a promotion. Which brings us back to hierarchy.

Here you raise an excellent point:

… the ultimate decisions about that individual’s skills and position are made on the terms established by the corporation—or more specifically, the corporation’s hierarchy.

As soon as we think this way, the idea of ‘skills causing income’ gets hopelessly muddied. I’ve lamented on my blog that many social scientists don’t understand causation, and I think this is worth restating. Many social scientists are trained in the tradition of methodological individualism, in which a ‘cause’ must ultimately be attributed to a trait of the individual. That ‘causes’ can be social does not compute.

As you point out, it is the hierarchy that decides which traits get rewarded and which do not. In a feudal society, the hierarchy rewards birthright. In a ‘meritocratic’ society, the hierarchy rewards ‘merit’. But as you point out (and as I’ve explored here), not every trait is rewarded with a corporate promotion. Only those skills that are deemed valuable to the corporation are rewarded.

The problem is that once we admit that the social environment is itself a cause of income, then things get really complicated. What caused the social environment? Where did the belief in birthright come from? Where did the corporate belief in ‘merit’ come from?

Thinking about these questions is where the methodology of economics completely fails us. Take the current obsession with randomized experiments. Imagine we take one group of people and give them a Harvard education. Take another group and give them a community-college education. Which group earns more income? The Harvard group, obviously. Conclusive proof, the randomistas would say, that skill causes income.

Now imagine trying to test the effect of the social environment. Now you have to take a group of people and put them into different societies. One group goes to a hunter-gather society with no hierarchy. Another group goes to a feudal society where rank is awarded by birthright. And a final group goes to a corporate ‘meritocracy’. It’s a fun experiment to think about, but one that is impossible to implement.

Still, I think about this experiment a lot because it is useful for understanding causation. In his Book of Why, Judea Pearl defines ‘cause’ in counterfactual terms. If x causes y, then taking away x removes the effect y.

With this definition in mind, let’s go back to our meritocracy model:

skills ⟶ hierarchical rank ⟶ income

Understanding the causal role of hierarchy rank in determining income requires taking away hierarchy. In other words, in a world without hierarchy, you can observe what statisticians call the ‘direct’ effect of skill on income:

skills ⟶ income

The funny thing is that societies that lack hierarchy also lack any significant form of inequality. So the direct effect of skill on income is … nil. Not really surprising. The reward to any human trait is mediated by our social environment … so it’s ‘direct effect’ is nil.

I’ll close by returning to the idea of ‘hidden skill’. My research shows that no personal trait (with available data) affects income as strongly as hierarchical rank. The reason, I believe, is quite simple: there are many paths to hierarchical advancement. Maybe you inherited a business from your parents. Presto — rank by birthright. Maybe you live in a racist society. Presto — rank by race. Maybe you live in a society that worships ‘intelligence’. Presto — rank by IQ score.

Many different traits are rewarded with advancement. And that’s why no individual trait predicts income as strongly as hierarchical rank. That’s because no individual trait perfectly predicts hierarchical advancement.

But imagine that there was such a skill. Although unobservable, this skill perfectly predicts hierarchical rank, and this ‘explains’ the returns to hierarchical rank.

I’ll call this the ‘dark matter’ argument. In cosmology, ‘dark matter’ is inserted anywhere that the Newtonian theory of gravity fails. In our theory of income distribution, ‘hidden skills’ are our dark matter. We insert them where ever observable skills don’t explain income (as well as hierarchical rank does).

You’re quite right that my argument about a ‘timing mismatch’ doesn’t dispel this argument. I think that the strongest critique is that the hidden skill hypothesis simply isn’t science. It’s irrefutable … hence garbage.

That said, I think we both agree that we shouldn’t dismiss ‘skills’ as a pathway to hierarchical advancement. I would love to see research about how the beliefs and attitudes of corporate brass affect the traits that are (or are not) rewarded in the corporate hierarchy. We need a modern-day Veblen to write a ‘Theory of the corporate class’.

[Jeremy Zerbe]

Thanks for your response, Blair. I think your walkthrough of the causality between skills and hierarchy here is clarifying. I was a little concerned that my focus on the “realness” of skill acquisition might be taken as a defense of human capital theory (which is mechanistic, intangible, unquantifiable and, frankly, bunk), so I’m happy that our understandings dovetail here.

I agree that a “Theory of the Corporate Class” would be a welcome addition to our line of inquiry. I think it would likely pick up from Veblen’s analysis of the lower leisure class whose non-industrious work exists to maintain the leisure, expand the control and, ultimately, spend the money of the upper class. As I’ve been reading his Theory of the Leisure Class, that’s how I’ve been thinking about the modern “business” work that those of us in corporate jobs do in service to owners of those corporations. It’s not quite a 1-to-1 comparison but it’s been illustrative to me. After all, aren’t our corporate overlords the new gentry and their brands the livery of servility?

[Dominic White]

Blair Fix wrote:

That said, I think we both agree that we shouldn’t dismiss ‘skills’ as a pathway to hierarchical advancement. I would love to see research about how the beliefs and attitudes of corporate brass affect the traits that are (or are not) rewarded in the corporate hierarchy. We need a modern-day Veblen to write a ‘Theory of the corporate class’.

Moral Mazes is quite an illustrative example of the basically feudal relations between lower- and higher-ranked corporate workers. Aged a little now and certainly would be curious to see how the dynamics have changed in (e.g.) Amazon, Apple, etc. contrasted with the financial and chemical corporations profiled in the book.

[Jeremy Zerbe]

Thanks for the suggestion, Dominic. This book looks fascinating. The quotes pulled from it in that Wiki article ring particularly true to me, and I’d bet it would make a good partner to Bullshit Jobs which I still haven’t gotten around to.

I noticed that the famously illustrative Peter Principle was linked as a related article, along with one that I’d somehow never heard given a name before: Parkinson’s Law, or the idea that “work expands so as to fill the time available for its completion.” The idea there is simple, but I was interested in the formula provided and how it might relate to the capitalization formula that’s put forth in CasP—in particular because it has to do with the growth of employees and the direction of their efforts.

Since it expands well beyond the bounds of CasP’s quantification (and, like most business “laws,” is veiled in a degree of irony), perhaps there’s nothing there. However, variable x clearly relates to CasP variable E (and the volume of employees that factor into it); variable k seems to coincide with Blair’s studies in hierarchy and income; and variables m and n seem to relate to sabotage in some form.

Worth a thought? Or am I grasping at straws here?

[Jonathan Nitzan]

There is plenty of research on institutionalized waste and inefficiency in capitalism, dating back to the early neo-Marxists and Veblen, among others.

Institutionalized waste and inefficiency are often seen as a ‘trick’ designed to create additional demand to ‘absorb’ the ever-growing surplus output generated by capitalism. They are presented as paradoxical, since the assumption (including of Veblen and the neo-Marxists) is that capitalists try to maximize profit, and that this maximization is generally presumed to require efficiency and abhor waste.

In CasP, the focus is not on maximum profit but differential profit, which often requires inefficiency and waste (and which can easily sustain both insofar as they are universal).

The tongue-in-cheek equation you refer to is written from the perspective of the individual manager (or rather the theorist of the individual manager), not the differential accumulators. And it seems nearly if not totally impossible to operationalize.

[Jeremy Zerbe]

I just wanted to drop back into this thread because I picked up Moral Mazes and have been reading it for the past couple of days. It’s excellent so far and I think it would be a worthwhile read for anyone here, especially Blair. It absolutely does provide a great deal of groundwork for a “Theory of the Corporate Class” and even refers to Veblen in the endnotes.

For a taste, here is a quote from the introduction in which the author Robert Jackall talks about what I’ll call the creorder of bureaucracy:

Bureaucratic work shapes people’s consciousness in decisive ways. Among other things, it regularizes people’s experiences of time and indeed routinizes their lives by engaging them on a daily basis in rational, socially approved, purposive action; it brings them into daily proximity with and subordination to authority, creating in the process upward-looking stances that have decisive social and physiological consequences; it places a premium on a functionally rational, pragmatic habit of mind that seeks specific goals; and it creates subtle measures of prestige and an elaborate status hierarchy that, in addition to fostering an intense competition for status, also makes the rules, procedures, social contexts, and protocol of an organization paramount psychological and behavioral guides.

And then, from the first chapter, as Jackall lays out the historical context that has given rise to bureaucracy:

No major occupation or profession in our society has escaped the process of bureaucratization. They are all—from assembly-line workers to physicians—specialized, standardized, arranged in a hierarchy and coordinated by higher authorities. Moreover, bureaucracy is never simply a technical system of organization. It is also always a system of power, privilege and domination. The bureaucratization of the occupational structure therefore profoundly affects the whole class and status structure, the whole tone and tempo of our society.

Now, these are more general observations on hierarchy than Blair’s research around hierarchical levels and their effect on pay, but as the book continues, Jackall dives deeper into the actual “skills” and “responsibilities” of the upper echelons of the corporate structures of the firms in which he was embedded. What he finds, of course, is that these skills are negligible, the managers themselves nearly interchangeable (by design), and that the defining feature that gets one promoted or sidelined is how “comfortable” they make others above and around them feel. This is a refrain to which Jackall and his subjects constantly return.

[A manager] makes other managers feel comfortable, the crucial virtue in an uncertain world, [by] establish[ing] with others the easy predictable familiarity that comes from sharing taken for granted frameworks about how the world works.

If there really is a “hidden skill” to be identified, it must certainly have to do with exuding this notion of comfort. Though how that would begin to be quantified or even qualified is well beyond me.

All in all, a really fascinating book. While it has definitely aged a bit, it has much to offer on these topics, including institutional waste which has been a particular interest of mine for a while (and about which I may return to this thread again later). My copy is a Twentieth Anniversary Edition with an afterword about the financial crash of 2008-09, so I look forward to Jackall’s update. Thanks again Dominic for the suggestion!

[D. T. Cochrane]

What this thread made me think about was the concept of “skill.”

A point I often hammer at is that innovation is not inherently good. I take this from Castoriadis, who argued the same about creativity. He observed somewhat that both Auschwitz and Beethoven’s 5th are creative outputs. We should not celebrate something simply for being creative. Since innovation is a subset of creation, the same applies.

We should also  be thinking this way about “skill.” Think about the skills that appear to be useful in climbing a corporate hierarchy. Yes, there are technical skills, skills associated with command of knowledge, especially domain specific knowledge. But there are also skills of navigating the personalities in the workplace. This is thoroughly entangled with one’s cultural background and the backgrounds of those that occupy the higher rungs of the hierarchy. People coming from backgrounds different from those of the people that typically occupy higher rungs often cite a feeling that the “culture” of the workplace felt alien. Different types of behaviour will be interpreted in different ways. Women talk about seeing aggressiveness in men be rewarded, while aggressiveness in women would often be punished. Those from outsider backgrounds often have to conform in order to succeed. I recall being told of a study of students entering and exiting med school regarding their perspectives on different topics. The men and women entering med school differed in some significant ways, such as on the importance of gender. On leaving med school, those differences were much more compressed. However, the compression was almost entirely about the women’s perspectives converging on the men’s. I suspect similar dynamics are at work in corporate hierarchies. For a woman, a person of colour, a queer person, to climb the ladder, they have to modify much of their behaviour to converge on what is considered “normal” by the gatekeepers who have also been trained and acquired this skill..

But this can also be very workplace specific, which is an important part of the luck factor. An aggressive woman who lucks into a workplace where the bosses appreciate her aggressivenes may be able to breakthrough the gender glass ceiling there, while if she’d started at a different workplace, she would have failed.

As BF has very capably demonstrated, the reasoning of the higher skill = higher placement in the hierarchy = higher income is entirely circular, since the evidence of being higher skilled is being higher in the hierarchy. But we could also dig into what is claimed to constitute “skill.”

 

[Author]

Replies