1.

My colleagues Jonathan Ponder and Moeko Porter’s “Women on Boards and Beyond” is out, and as always, deserves a closer look. The headline: Women held 28.3% of board seats globally as of October 2025, up a percentage point from the year before, with nearly half of companies now hitting the 30% threshold. The EU’s June 30 deadline for 40% representation is concentrating minds in affected markets, even as some U.S. companies have quietly stepped back from framing board composition through a diversity lens.

The report goes well beyond headcounts. Women held a higher share of committee seats than overall board seats, with the strongest gains on nomination committees in emerging markets. Controlled companies lagged the most on female representation. And in a finding that may surprise, the information technology sector ranked lowest in the share of companies with at least 30% women on their boards, yet recorded the highest percentage of female board chairs (13.2%, up from 11.9%) and the highest share of female CFOs (26.8%, up from 23.2% ). The pipeline is not always where you’d expect it.

Read here.

2.

Why are women, on average, more skeptical of AI than men? That’s the question authors Sophie Borwein, Beatrice Magistro, R. Michael Alvarez, Bart Bonikowski and Peter J. Loewen set out to answer in “Explaining women’s skepticism toward artificial intelligence: The role of risk orientation and risk exposure,” drawing on a survey of roughly 3,000 respondents in the U.S. and Canada.

Their answer: It’s less about technology literacy and more about risk. Women tend to exhibit greater sensitivity to risk generally, and face greater exposure to AI-related risks specifically, including job displacement in occupations where women are concentrated. Using text analysis of open-ended responses, the authors find that women across education levels expressed more uncertainty about AI’s benefits and raised concerns about its downsides more frequently than men. The policy implication the authors draw is that AI governance needs to account for these differences, and that bringing more women into AI-related fields isn’t just an equity argument, but a design one.

Read here.

3.

Making the case for using more machine-driven decisions than less: Use of algorithms to value single-family homes could help reduce racial bias in the U.S. housing market. That’s the finding of Lyndsey Raymond, a postdoctoral fellow at Microsoft Research who will join the MIT faculty this summer, in “The Market Effects of Algorithms,” a working paper that traces what happened to house prices as housing records shifted from paper to digital between 2009 and 2021.

The digitization opened markets to a new class of algorithmic investors, those using automated acquisition engines, predictive valuation tools or dedicated data-science teams to identify and price properties at scale. Raymond finds that their entry compressed valuation gaps that had persisted for decades under human-based appraisal, increasing the average sale price of minority-owned homes by 5% and reducing racial disparities in home prices by 45%. It’s a counterintuitive result at a moment when algorithmic bias is more often cited as a problem to be solved than a solution to one, and a useful reminder that the direction of bias depends heavily on what you’re replacing.

Read here.

4.

If you’re tracking the growing impact of extreme weather events and other physical risk on property insurance, “Perspectives on Insurance and Climate Risk,” a literature review by Parinitha Sastry of Wharton and Ishita Sen of Harvard is an excellent place to start. Forthcoming in the Journal of Finance, the paper surveys nearly 190 papers on how rising physical risk shapes supply and demand for homeowners insurance, and how those disruptions ripple in housing and mortgage markets.

The authors are candid about the field’s gaps: Property insurance is less studied than health or life insurance, partly for want of good data. That’s changing. But the economic pressures aren’t waiting for the research to catch up. Insurers are pricing risk annually with models built for longer horizons (facing capital constraints that worsen after the very disasters that trigger claims) and operating in a regulatory environment, state by state, that often prevents them from charging actuarially fair premiums. For investors in financial institutions or real estate, the paper offers a useful map of where the bodies are buried.

Read here.

5.

With the U.S. Supreme Court’s ruling on the Trump administration’s tariffs returning trade policy to the front page, there may be no better moment to read “The Incidence of Tariffs: Rates and Reality,” by Gita Gopinath of Harvard and Brent Neiman of Chicago. The paper examines what actually happened to prices, sourcing and the dollar after the 2018-2019 tariffs and the 2025 round.

The tariffs announced in April 2025 looked alarming, with a trade-weighted statutory rate peaking near 27%, but importers paid roughly half that (about 14%), thanks to shipment lags, exemptions, USMCA compliance and enforcement gaps. What was paid still bit: Pass-through to U.S. import prices ran at roughly 94% in 2025, meaning American importers, not foreign exporters, absorbed most of the cost. China’s share of U.S. goods imports collapsed from 22% to 8% between 2017 and September 2025, and the tariffs translated into what the authors call a “production tariff” equivalent to a cost increase of more than 1 percentage point across U.S. manufacturing. One puzzle: Standard theory predicts tariffs should strengthen the dollar, as in 2018-2019. This time it fell, which my colleague Ashley Lester has attributed to policy uncertainty compounding trade disruption. The Supreme Court ruling adds another variable to watch.

Read here.