Like The Hub?
Join our community.
Join

Sean Speer: Welcome to the age of overeducated underachievers

Commentary

I admire and respect New York Times columnist David Brooks. His political analysis—including his underrated thinking on “national greatness” conservatism in the late 1990s—and his more personal writing—including his 2015 book The Road to Character—have had a meaningful effect on me. He’s someone that I’ve regularly turned to make sense of culture, ideas, and politics. 

So I wasn’t surprised that his recent column on the role that the “winners” of the modern economy have inadvertently played in creating the economic and social conditions for Trumpian populism has received such widespread reaction. 

It’s a thoughtful piece of analysis about big changes in our economy and society, how certain people and groups have navigated these changes better than others, and how they’ve come to create an interconnected system of education, employment, institutional leadership, familial and social relations, and cultural norms and signals that perpetuates their advantages. As Brooks writes: 

The ideal that we’re all in this together was replaced with the reality that the educated class lives in a world up here and everybody else is forced into a world down there. Members of our class are always publicly speaking out for the marginalized, but somehow we always end up building systems that serve ourselves.

His column is an important and necessary reckoning with the factors that have caused large shares of Western societies to become anxious about the present and pessimistic about the future. It’s far more constructive than many in his circles who are satisfied to dismiss the concerns of their fellow citizens as mere expressions of backwardness, ignorance, or racism. Brooks deserves tremendous credit for his empathy and self-introspection. 

Much of his analysis is persuasive. It’s clear that the shift from a goods-producing economy to a knowledge economy has preferenced those with certain credentials and skills as well as particular regions and communities. The trickle-down effects, including on marriage and families, broader social relationships, and of course politics have been profound. 

Yet as important and insightful as the essay is, it suffers from one major flaw. It assumes that “meritocracy” and “credentialism” are essentially synonyms. They are not. 

Although it’s of course true that many, perhaps most, of the winners in the modern economy are highly educated, there are still some who have succeeded without educational credentials and an even bigger share that are failing in spite of them. 

The latter group—let’s call them “overeducated underachievers”—strikes me as crucial for understanding growing pessimism about the future, the rise of populism, and the anger and frustration that we’re increasingly seeing manifest itself in our politics. 

The rise of the working-class university graduate 

Last year, I co-authored a paper for the Cardus Institute on Canada’s modern working class. We defined working class as those in occupations that typically don’t require post-secondary credentials. One of our most surprising findings is that 53 percent of those in working-class jobs actually have post-secondary certificates, diplomas, or degrees (see Figure 1). This group excludes full-time students so their “underemployment” cannot be explained as merely a transitory step in their careers. 

Graphic credit: Janice Nelson.

There are no doubt various factors at play, including individual preferences, foreign credential issues, a geographic skills mismatch, or even employer discrimination. But a major contributing factor is the interplay between the rise of credentialism—the idea that one’s ability or intelligence is measured by their academic credentials—and what is sometimes referred to as “credential inflation”—the economy-wide trend of rising expectations with respect to educational credentials required for a job. 

British thinker David Goodhart argues that these economic and social currents have caused us to “overshoot” when it comes to education and training. We now appear to have a large-scale overeducation problem in advanced economies including Canada. 

Consider for instance a growing body of research on the rise of these “overeducated underachievers” and their labour market experiences and outcomes. The numbers are quite striking. A major 2014 study for instance found that in the United States about 37 percent of four-year college graduates tend to have more education than their jobs require. Goodhart’s more recent work on the United Kingdom shows that more than a third of university graduates are in non-graduate employment more than five years after graduating. Previous research by the Parliamentary Budget Office finds similar outcomes for Canada. 

Although these overeducated workers tend to earn more than less educated workers in the same occupation, they earn a lot less than similarly educated workers in occupations that match their credentials. They also tend to remain overeducated over the long run. One consequence is the wage penalty for their overeducation similarly tends to persist. Another is that these workers report having low levels of job satisfaction as well as self-reported bouts with anxiety, depression, and other mental health challenges. 

Those affected have fallen between the cracks of Brooks’ conception of the meritocracy and the real-life experience with modern credentialism. They’re the ones who “did everything they were expected to”—including obtaining a university or college credential—and yet haven’t realized the payoff. In Canada, for instance, they’re earning on average 41 percent less than those in non-working-class jobs and, depending on where they live, struggling to afford rent and other basic costs. The promise of the so-called “democratization” of educational access has failed to fully materialize for them and their families. 

There’s a strong case in fact that while there have been significant benefits to expanded access to post-secondary education, they’ve come with these underestimated costs that we’re only now starting to understand. Cultural norms and public policies in favour of what Goodhart calls “peak head” has devalued non-cognitive skills, eroded academic standards, and contributed to credential inflation in the job market. Put bluntly: there are people with advanced degrees who shouldn’t have them and the market has had to adjust to account for them. 

These people themselves cannot be blamed for these developments. They’ve responded to the same elite signals and social norms about the utility of post-secondary credentials that are implicit in Brooks’s essay. They got good enough high-school grades to get accepted into university or college programs which they ultimately completed. They may even have subsequently done an advanced or professional degree. Yet they’ve still been overlooked or rejected by the meritocracy. 

What’s behind the rise of modern credentialism?

It’s interesting to think about what has caused the educated leadership class—including business, cultural, and political leaders—to cultivate such powerful social expectations. What is behind the rise of modern credentialism? 

A major factor undoubtedly is a progressive view about education, intellectual knowledge, and human progress. There’s an inherent assumption that expanding education is the key to unlocking a version of utopia in which an expanding “cognitive class” can overcome the class struggles of the old goods-producing economy. It also captures the inclusive ideal of meritocratic thinking and the role of education as the “great leveler” in society. The “university for all” crowd is a good example of this ideological tendency. 

Another may be the cause of self-selection bias. If everyone around key decision-making tables in society has post-secondary credentials, it’s not a huge surprise that they’d cultivate social norms related to education that would tilt in favour of their own backgrounds and experiences. Duke University professor Nicholas Carnes’ work on the so-called “white-collar government” (which refers to the narrowing of educational backgrounds and professional experiences of members of the U.S. Congress) has documented how the growing homogeneity of the political class influences broader policymaking including with respect to education. 

Goodhart argues that it’s not merely bias. It also reflects in his words “an indiscriminate spirit of not wanting to kick away the ladder on the part of people who have had valuable university experience themselves.” Yet the problem with this type of thinking is that it imposes one’s own attributes and preferences on the rest of society. It may be well-intended but it’s also narcissistic. It counterintuitively undermines pluralism in the name of inclusion. Real inclusion would strive to enable people to pursue their own interests and maximize their own strengths rather than presuming the right path for them. 

My own hypothesis is that the massive cultural and political emphasis on post-secondary education in recent decades is itself a sign that the leadership class doesn’t know what to do. A combination of factors—including globalization, public policy, and technological change—has transformed the modern economy from an “economy of things” to an “economy of thoughts.” The resulting “skills-biased technological change” is reshaping market demand, employment opportunities, and earnings based on certain credentials and skills. 

Policymakers have struggled to keep up with or even make full sense of these developments. In the absence of a clear understanding of the long-run implications for the economy and society, an expansion of post-secondary education has become the default response. A growing population of overeducated underachievers is its byproduct. 

A bourgeois revolution: The real risk of political disruption

The gap between the promise of credentialism and its disappointing realities for rising numbers is something that we must confront before it engulfs our politics. For all of the attention paid to the threat posed by working-class populists, the real risk of political disruption may actually come from the rise of white-collar populists. 

The latter has good reasons to be agitated. They were led to believe that educational credentials were a path to social mobility and membership in the meritocratic class. Yet they instead find themselves in lower-paying and more precarious work than their typically uneducated parents. The real revolutionaries in other words may not be the proletariat but the disaffected and unfulfilled bourgeois. 

University of Connecticut scientist Peter Turchin has been warning about this risk for more than a decade. In a 2010 article for the science journal Nature, he predicted the rise of today’s political instability due in part to what he called “elite overproduction.” 

The basic idea is that our societies are training and producing more highly-educated individuals than there are elite positions in business, government, or other key institutions. This supply-demand gap threatens the rise of counter-elite movements that may eventually aim to overthrow the political order. He points to historic examples like the French Revolution, the Russian Revolution, and even the American Civil War as evidence of his theory. 

Turchin has returned to this theme in a new book entitled, End Times: Elites, Counter-Elites, and the Path of Political Disintegration. In it, he uses a metaphor of musical chairs to convey his argument: 

I use a metaphor of playing a game of musical chairs, but instead of removing chairs one at a time, you keep adding more and more players. And so as they’re twice as many, three times as many, players as there are chairs, you can imagine how much chaos would ensue. This is a good metaphor for our societies because elite aspirants typically are energetic, ambitious, well-educated, good at organizing, and therefore when they’re frustrated in getting the positions that they expect, many of them turn to trying to infect, overthrow, the unjust social order as they perceive it.

For these reasons, he warns that “disgruntled elite-wannabes are far more threatening to societal stability than disgruntled workers.” Their growing ranks therefore ought to be the subject of far greater political and policy attention.  

Which brings us back to Brooks’s essay. Although he’s right to look inward at the role that educated elites have played in contributing to the rise of contemporary populism, his conclusions aren’t quite correct. The chief problem isn’t less-educated classes or meritocracy itself. It’s the flawed culture of credentialism and its diminishing returns on educational credentials for legions of overeducated underachievers that may ultimately prove to be his class’s gravest mistake.

If you don’t like working-class populists, you’re going to hate the white-collar populists. Unless, of course, you are one.

Malcolm Jolley: The key to tasting wine? Staying in the moment

Commentary

I don’t remember how I acquired my copy of Adam McHugh’s wine memoir, Blood From A Stone. It was published last fall, and it’s possible his publisher sent it to me, or maybe their Canadian distributor. I used to review food and wine books frequently and was on lists to receive review copies, or at least catalogues of new books from which to request one.

I suspect I bought it myself, though. Karen MacNeil, the California-based wine writer, reviewed it glowingly on her blog in January. I have a vague, though probably not false, memory of ordering it on a whim after reading MacNeil’s post. January was a crazy-not-in-a-good-way month, in what has turned out to be a crazy year in my house and with my family. I must have been distracted by events and forgot about the book in the pile at the corner of my desk.

I found Blood From A Stone just in time for a couple of weeks’ holiday in July, and the timing couldn’t have been better. After the trying winter and spring, I was ready for it. The book is about how McHugh changed his life from being a hospice chaplain to being part of the wine trade in Central California, coming to terms with loss, and re-imagining one’s self and circumstances.

Much of the book is about wine, generally: science, history, culture, all of it. If McHugh’s reader is already a wine nerd, then he tells some familiar stories in a novel and amusing way. If McHugh’s reader is not already a wine nerd, then he provides an excellent introduction to the subject. Though it sometimes touches on dark subjects, it’s a fun and witty read.

One of the things that Blood From A Stone is really about is change, welcome or otherwise. Through the book, McHugh makes connections between events in his life and the processes, whether chemical or cultural, that make wine. 

Wine is also really about change. The winemaker turns sweet grapes into dry alcoholic wine. Or at least they guide the natural process of change. The same might be said of any cooked food or alimentary product. But what separates a bottle of good red wine from a bottle of Coca-Cola is that it is, at least notionally, alive and prone to change as it sits in the cellar or opened on the counter.

Another California wine writer, Alder Yarrow, published a kind of interview last week on his website, Vinography, with the French bio-dynamic winemaker pioneer, Nicolas Joly. In it, Joly suggests the true test of a well-made wine is time after the bottle has been opened and the wine exposed to oxygen.

He tells Yarrow: “Here’s is the test of truth. You follow the wine. You drink a glass, another glass, and then you wait and have another glass three days later. In summer, all true wines will improve for 8 or 9 days.”

This seems extreme, though having met the man, I have no doubt that he does it regularly and that his Loire Valley wines, made with organic Chenin blanc grapes grown on land that has been under vine for centuries, are up to the challenge. Less dramatically, I have returned from a week away to find a perfectly good half bottle of wine waiting in my fridge. Whether it had improved or not, I couldn’t say.

Most wines today don’t need to be opened well in advance of being served or to be decanted to encourage more contact with oxygen in the air. Modern, temperature-controlled wineries make stable wines, which don’t require heavy tannins, from grape skins, stems, or new oak barrels, to keep them from spoiling.

Producers still make wines that are meant to age, but today they don’t have to, and the great majority of wine is meant to be drunk within a year or two of being bottled. This summer I am regularly pouring a white wine from the North of Italy: the 2021 Zenato San Benedetto Lugana. Lugana is a DOC designation for white wine that holds the distinction of being stretched between two political regions: Lombardy and Veneto on the south shore of Lake Garda.

The Zenato Lugana is just under $20 in Ontario, and there is enough of it sent here that it is relatively well distributed across the provincial liquor retail monopoly. Apart from the twin virtues of being well-priced and fairly easy to find, the Zenato Lugana is also a fruit-forward crisp sipper with some weight in the mouth. It’s not terribly complicated, but it’s a pleasing refreshment that’s easy to sip as an aperitif.

But here’s the problem: I am not sure exactly what it tastes like. When I first tried it, I made a note that I was receiving tangerine. The second time, I was convinced it was peach. By the third, it was grapefruit and apricot. Now, when I open a bottle, I wonder which version will end up in my glass.

Taste is, of course, subjective. I write my tasting notes independently, but then I like to compare them against other wine writers I respect. They’re never exactly the same, but often similar, especially if I have resorted to vague descriptions like “red fruit”. And sometimes they are as different as tangerines and peaches. When that happens, I’ll look up a third and a fourth review, only to be further confounded by a lack of sensory consensus.

Wine is, of course, itself subjective. Any number of variables, from temperature to oxygen exposure or the shape of the glass a wine is poured into, will affect taste. Many (like me) believe the weather can change how a wine tastes and distinguish between “fruit days” and “root days.” The best explanation for this phenomenon that I have heard is that day’s barometric pressure affects the volatility at the surface of the glass of wine, which changes how many aromatic molecules will reach the taster’s nostrils.

In the end stone fruit notes from the Verdicchio grapes that make the Luagana are winning over the citrus ones. But I like that that could change with the next bottle. It might also change when the 2022 vintage replaces the 2021. Wine like most good things is about being in the moment and never really knowing what’s coming next.