Daniel Jordan Davidson observes that I don’t properly distinguish between ‘productivity’ and ‘product’ in this post: the former refers to the output that can be produced per unit input, while the latter refers to total output produced. Thank you to Danny for pointing this out to me; I’ve left the post as-is because I don’t think it affects the substance of my argument, but it is an important terminological distinction.
§1.
A few years ago, it was all the rage to talk about how ‘meritocracy’ was an illusory ideal. There were two dimensions to this. The boring one was just that Western capitalism isn’t completely meritocratic (think of discourse around the US college admissions scandal); this is boring because nobody ever claimed that it was, just that (a) meritocracy is an ideal worth striving for and (b) Western capitalist nations are closer to it than anyone else.
The second, more interesting dimension was the people who accepted (b) as true, but took issue with (a). A common theme was that the term ‘meritocracy’ was originally coined by Michael Young to describe a dystopia—it was supposed to be a negative, an ideal that we should reject, and yet the West has ended up embracing it. People wrote books attacking the ideal of meritocracy, notably Daniel Markovits’ The Meritocracy Trap and Michael Sandel’s The Tyranny of Merit.
I am late to this party; the discourse has already moved on, and mentioning meritocracy is pretty blasé. But I think the ideas here are interesting. I agree with some of the arguments that people make against (a), when they stick to actually attacking meritocracy. But their arguments become a lot weaker when they try to substantiate them with examples from actual Western capitalism; and this is because (b) is not true. Western capitalist countries are not particularly close to meritocracy, and their policies were never actually meant to be meritocratic. They are something closer to ‘productocracies’; the distinction between merit and productivity is crucial for understanding what’s going on here.1
§2.
Let’s start by defining some terms. ‘Merit’ refers to a person’s talents, abilities, and deservingness: interviewees for a job can be described as showing a lot of merit. A meritocracy is a society in which people are rewarded (with wealth and often also power) in proportion to their merit. ‘Productivity’, as its etymology would suggest, is a measure of how much someone produces: A is more productive than B when A produces more goods and services (or more valuable goods and services) than B.
For our purposes, there are two important differences between merit and productivity. The first of these is what kind of concepts they are. ‘Merit’ is a thick ethical concept: its descriptive meaning is bound up together with ethical commitments. The use of the term ‘merit’ implies, at least some of the time, that those who have merit should be rewarded on the basis of it, or at least that they deserve to be recognised for their talents. (Although today we don’t think that those with fewer talents should be thrown into the outer darkness, where there will be weeping and gnashing of teeth.) ‘Productivity’, by contrast, is a technical term from the academic discipline of economics. It doesn’t have any ethical implications, at least not in the direct way that ‘merit’ does.
The second difference is that merit is an intrinsic property (it describes you independent of the situation you’re in), whereas productivity is extrinsic (it depends on how you relate to your situation). Merit is about what kind of person you are, what skills and qualities you have. These things also play a role in productivity; but how productive you are is also a matter of who you’re working with, what tools you have, how the firm you’re working for is structured, what the laws of your country are, etc. etc. Another way of putting this is that merit is one of the inputs into the process of labour, while productivity measures the output.
§3.
The relevant economic result for our discussion is that, in a competitive market, workers are paid their marginal product, the amount of additional productivity they bring to their company. If a new worker would increase the company’s productivity more than the amount the company has to pay them in wages, then the company will be incentivised to hire them; if the increased productivity is less than wages, then the company will have a strong incentive not to hire them. And as such, the equilibrium state is one where workers’ wages are equal to their productivity on the margin. So the amount workers are paid, or whether they are hired at all, is a function of their productivity.2
I don’t mean to imply that the model of a competitive market is a good description of the labour markets of Western capitalist nations; this would take us into a technical discussion of economics, and anyway it is irrelevant for my argument. All I need is the premise that this model has been one of the central ideals that shapes Western capitalism, and Western policymakers have been deeply influenced by it when changing laws and regulations. As such, this model tells us what the West has been aiming for (at least some of the time), whether or not it’s actually got there.
And the important point is that the West has not been aiming at merit, because merit and productivity can come apart. An example might help here. Consider a startup founder, Max, who’s looking to make a new hire. There are two candidates: Sarah is a well-qualified, talented, excited worker with a CV as long as your arm; Alison does not have the same talent or qualifications, but had previously worked with Max on a different project and built up a rapport with them. Sarah has more merit, because she’s intrinsically more suited to the job. But the rapport between the Alison and Max is important: it might help Alison hit the ground running and get used to the job quicker (crucial in the risky early days of a startup), and it will probably make the two of them more comfortable with bouncing ideas off each other and help Max come up with solutions to their problems through the exchange of ideas. For these reasons, it might be the case that Alison is more productive than Sarah. So Max might hire Alison even though Sarah has more merit.3
Consider another example. All of us have come across people who are just a bit awkward to be around: they’re not abusive or serial liars or violent or anything like that, they’re just lacking in a few crucial social skills. They might be a bit abrasive, or unable to pick up on jokes, or too quick to feel victimised, or talk too much about themselves, or just obsessive in an unpleasant way. Some of these people are incredibly well-qualified and meritorious (we all know a high-flying maths student like this). But they can often make work an unpleasant place for other people, dragging down their productivity. And in some circumstances, this can have a net negative effect on productivity on the margin. So firms will be hesitant to hire someone who demonstrates poor social skills in an interview, even if they also demonstrate incredible merit.
The general point is that the ideal of a competitive market rewards people based on productivity, not merit. Merit is an ethical idea, that tracks what kind of person you are and what you deserve. But the market isn’t trying to give people what they deserve, and in a lot of circumstances it doesn’t do so. Jeff Bezos does not have several billion times more merit than you or me—that would be an ethical judgment, and an absurd one at that; but he has produced goods and services that the market dictates are worth that much. Productivity is an extrinsic property, and so depends on features of the world that aren’t captured by the intrinsic property of merit. And if Western capitalism has been aiming for a market-based ‘productocracy’, then it has not been motivated by the ideal of meritocracy.
It will probably be objected that merit is correlated with productivity. Sure, there are other factors, but whether you have merit isn’t irrelevant to how much you produce. This is true: if people are rewarded according to their marginal productivity, then rewards will weakly correlate with merit. But this isn’t a new feature of modern capitalism—it’s entirely possible to have an aristocratic society where merit is weakly or even strongly correlated with being an aristocrat, because (for example) training and education are only offered to the aristocracy.4 Yet we’d never call such a society meritocratic—meritocracy was supposed to be the opposite of aristocracy! So this weak correlation doesn’t actually show that Western nations are more meritocratic than other societies, or that meritocracy has been more of a driving ideal.
§4.
So why do so many people think the West is meritocratic? Why was meritocracy such a common target for those looking to object to inequality or ‘neoliberalism’ or whatever?5 I think the answer is in two parts.
The first is just that the distinction here is quite subtle and easy to miss. I’ve had discussions about this before where, even after laying out the distinction as clearly as I can, smart people have struggled to understand how someone could be productive without it being a result of their merit. Indeed, some people have tried to resolve these disagreements by redefining ‘merit’ so that it just means productivity.
This would, of course, make competitive markets into meritocracies by definitional fiat. But in the process, it would mean we could no longer account for the two important differences between the concepts, that merit is intrinsic while productivity is extrinsic, and that merit is ethically thick while productivity is thin. We should keep the concepts separate; but this will leave open the possibility of confusion. So the first reason why people think the ideal of meritocracy is ascendant is because they might not be able to fully grasp the distinction between a competitive market (which actually is an ascendant and influential ideal) and a meritocracy.
But the second is that, because ‘productivity’ is an ethically thin concept, there’s no visceral ethical understanding of why it might be good to reward workers based on their productivity.6 By contrast, saying that people are rewarded based on their merit sounds like people are getting what they deserve, are being rewarded for their character and hard work. It is thus convenient for politicians to use the rhetoric of meritocracy when pushing through policies that are really about increasing productivity, and often that is what has happened.
Now, as Quentin Skinner teaches us, the language that politicians use to defend their policies has effects on the shape those policies take; so the fact that the language of ‘meritocracy’ has been used has shaped the policies that have gotten proposed. This has had its most important effects in education, which is also the area that a lot of examples are taken from in debates over meritocracy. But this is not the same as policies having actually been driven by meritocratic ideas from the start. The language of ‘meritocracy’ might have had indirect effects on policy in some areas, but it has not been its main driver.
§5.
It might seem that the upshot of this discussion is anti-market: the market fails to respect people’s merit, rewarding them on the basis of morally irrelevant extrinsic factors, and we should move towards a more just, more meritocratic system. But I think this is completely wrong.
First, I think Sandel and Markovits (and others who make similar arguments) correctly identify that a meritocratic society would have deep ethical problems of its own. Markovits, for example, correctly argues that a meritocratic society would have deep inequality that would widen over time. He is wrong to identify our society as as a meritocracy (e.g., while we have widening inequality, it is caused primarily by the political economy of housing), but he is right about the problems that would emerge if it were.
And second, this line of argument takes an overly moralising approach to understanding political economy. One of Sandel’s central points is that merit itself isn’t morally ‘pure’: even though merit is an intrinsic property, whether or not you had the opportunity to gain merit depends on extrinsic factors (like how much money your parents went, where you went to school, how lucky you were in job applications, etc.). The attempt to reward people solely on the basis of ‘morally relevant’ factors leads not to meritocracy but to luck egalitarianism, an absurd position which, in attempting to factor out all morally irrelevant factors, is left with the libertarian free choice of a Kantian self as the only basis for making decisions about reward and desert.7
Given, of course, that there are no autonomous Kantian agents with libertarian free will running around, if we try to make a sharp distinction between morally relevant and morally irrelevant factors we will inevitably fail. We have to take a more sensible and realistic perspective, and be more willing to be flexible about what considerations we count as relevant. And among these considerations will be the ways in which markets coordinate action, incentivise efficient use of resources, communicate local knowledge and reliably incorporate partial information, and provide a basis for reliably increasing living standards. When we abandon the idea that there’s a sharp distinction between ‘morally important’ and ‘morally irrelevant’ factors, we have to pay attention to these central facts.
I don’t think we can, or should, throw out the ideas of desert, reward, or merit, and replace them solely with the impersonal hand of the market: Isaiah Berlin rightly insisted that the great strength of a liberal society is that it is pluralistic and incorporates a wide variety of values. But a meritocracy would be just as ethically monistic as the harshest anarcho-capitalist market society; I wouldn’t want to live in one, and I’m deeply glad that in spite of widespread misapprehensions, I actually don’t.
Readers familiar with Hayek will recognise that much of this post is Hayekian in spirit; I claim no particular originality. If you enjoy this post, I highly recommend looking at Hayek’s ideas about merit, which can be found in chapter 6 of The Constitution of Liberty.
Yes, economists, I’m aware this is far too simplistic. I think this paragraph captures just enough of the core features of the model for the purposes of my argument, but if any readers want to dig a bit more into this result, pretty much any introductory economics textbook will contain a discussion of the marginal product of labour.
This link between personal relationships and productivity is one of the reasons why ‘it’s not what you know, it’s who you know’.
For more on this argument, see Bernard Williams’ ‘The Idea of Equality’.
The irony that people trying to object to ‘neoliberal capitalism’ end up just reiterating arguments originally made by Hayek should not be lost on the reader.
Moreso when the distinction is made between average productivity and marginal productivity; many people who do grasp the distinction between productivity and merit are nonetheless confused as to why people care about marginal productivity rather than average productivity.
I make no apology for calling luck egalitarianism absurd.