Expect A Circus

Thomas Friedman | Workplace Hostility | Diversity Training Doesn’t Work | Don't Fear The Reaper

If you hire clowns, you should expect a circus.

| Thomas Friedman

…

I’m not a fan of Thomas Friedman, but this quote — referring to the state of our government — is pretty dead on. DOGE was horrific, deporting law-abiding legal residents unconstitutional, and now immolating world trade is incomprehensible.

What every happened to getting egg prices down?

---

---

Workplace Hostility

Seen on Bluesky via Florian Ederer (@florianederer.bsky.social): According to Manuela Collis and Clémentine Van Effenterre,

People are willing to forgo 15-30% of their wage to avoid hostile work environments.

Women report a stronger distaste for exclusive workplaces and environments with sexual harassment and value hybrid work twice as much in the presence of sexual harassment.

The working paper’s pdf is here.

From the report [emphasis mine]:

Our analysis reveals several important patterns in the demand for hostility-free workplaces. First, willingness-to-pay estimates are relatively stable across the wage distribution and do not increase with workers’ outside options, suggesting that workplace quality may be an important driver of labor force participation and retention decisions. Second, vertical hostility (from management) appears to impose greater costs on women than horizontal hostility (from peers). Third, we find evidence that individuals systematically underestimate others’ valuation of hostility-free environments, pointing to potential coordination failures in achieving better workplace cultures.

We further document important interactions between workplace hostility and alternative work arrangements. Women value hybrid work twice as much in the presence of sexual harassment suggesting that remote work may serve as a substitute for avoiding hostile environments. To explore the equilibrium implications of these patterns, we develop a model of compensating differentials where firms offer jobs that vary in both observable amenities (remote work) and unobservable ones (risk of hostility). The model demonstrates that gender differences in perceived risk of workplace hostility are a significant driver of both the remote work pay penalty and office workers’ rents, with effects comparable to major technological shocks facilitating remote work provision.

The economics suggest that people (women in particular) are willing to exchange higher pay for less hostile workplaces, while those who are less concerned about hostility benefit, since companies with high degrees of hostility wind up paying more for their workers, across the board.

---

---

Diversity Training Doesn’t Work

From 2013

In Does Diversity Training Work the Way It’s Supposed To?, Adam Grant and a group of collaborators looked into that, and?

No, it doesn’t.

We found very little evidence that diversity training affected the behavior of men or white employees overall — the two groups who typically hold the most power in organizations and are often the primary targets of these interventions.

There were some other surprises, but the obvious approach — train those most likely to be biased against minorities and women — simply does not change things. The authors suggest a variety of half measures and the recommendation to experiment, but the stark results of the research are fairly negative.

---

Don’t Fear The Reaper

In Why Americans fear the AI future, Noah Smith asks some good questions but the answers are less good.

Americans have spent much of the last four decades watching technology fail to produce the kind of rapid, broad-based income growth that it had in the past. That made them think of their economy as more of a zero-sum proposition, and it made them fear that any new technology will simply create more winners and more losers.

And, of course, they might lose their jobs altogether.

But he makes the case that 'normies' will benefit most from LLMs and their ilk. Still, people don't believe it.

Why doesn’t the general public believe in this good case yet? I suspect that one main reason is that nothing like this optimistic scenario has happened during the last 40 years of IT innovation.

Exactly. Corporations have siphoned off the benefits of technologies and used it to increase inequality.

If inequality hadn’t risen since 1975, the typical American would make $59,000 instead of $37,000. For most Americans, a $22,000 boost in income would be life-transforming, and would put a much different spin on the last half century of economic growth.

And even as their incomes grew at this modest pace, Americans could see the upper slice of the country becoming much richer, much faster. They could see the lavish homes owned by business owners and highly-paid knowledge workers, the fancy private schools where they educated their children, the servants they could afford to clean their houses and sculpt their lawns and take care of their children so they could go out on the town. They could see the fabulous wealth that technological progress in the IT industry was creating for America, and they knew that they were only getting a tiny sliver of it.

In other words, the era of IT innovation — of the computer revolution and the internet, of social media and smartphones — felt close to a zero-sum game for lots of people. Not quite a zero-sum game, but close to it. And in a zero-sum game, the only way you get richer is if your neighbor gets poorer. That’s a world where pessimism makes sense, even for the winners — if you’re on the top rung of a zero-sum society, you know it’s only because you climbed over someone else’s back to get there. The people making $500k at Google can look around and see that they’re pulling far ahead of the average American. And they know that if their human capital is ever suddenly devalued, they could be knocked right back into the slow-growth world of the middle class. No wonder they’re afraid of AI too.

Even those that benefit are afraid too.

Woulda, coulda, shoulda

We should have spent the last half century actively fighting against inequality. Higher wages, wage boards, and stronger unions for local service-sector workers could have turned checkout clerk jobs and barista jobs into good middle-class jobs. Universal cash benefits could have mechanically redistributed the wealth being created by the long IT boom. Better public transit and more public spaces could have created more of a sense of common destiny and shared experience between rich and poor. These measures might not have prevented techno-pessimism — Germany has good social protections, and Germans fear AI even more than Americans — but it might have managed to preserve some of the national optimism that the U.S. inherited from its youth.

But we didn't, and so the populace on the whole has become disaffected: they don't trust the economic system, and the elites at the top that guide it (and benefit the most), to make an egalitarian future. And so nearly every change proposed is questioned as yet-another-way-to-be-screwed-over.

…

A dozen concluding paragraphs and an argument from Sangreet Paul Choudary below the paywall.