The variance framework clarifies a lot. The harder part may sit inside the "bounded downside" category itself.

Global health interventions look convex partly because the downside appears measurable and contained. But "unlikely to cause harm" and "bounded downside" often get conflated in ways that deserve more scrutiny. An intervention with weak evidence, scaled substantially, can produce harm that only surfaces in aggregate — substitution effects crowding out stronger alternatives, institutional credibility consumed on interventions that underperform, donor attention patterns that persist beyond the evidence. None of those show up as direct harm in a single-program evaluation.

The distribution may look bounded from inside one intervention and fat-tailed from the portfolio level. That asymmetry seems worth naming before treating global health as the safe side of the variance line.

Fair — that paragraph packed too much in.

Here's the concrete version. In the 1890s, John D. Rockefeller had more money than he could give away carefully. Hundreds of letters arrived daily asking for funds. One ship from Europe carried five thousand requests alone. Most were dubious. He couldn't tell which causes were real, which organizations could actually use a large gift, which problems sat at a root level versus a symptom level. So he hired a minister named Frederick Gates to build a sorting system — read everything, investigate the credible requests, reject the rest. Carnegie ran the same play at roughly the same time, building evaluation infrastructure while the capital was still accumulating.

Gates started that work in the early 1890s. The Rockefeller Foundation didn't open until 1913. Twenty years of evaluation work happened while the wealth was still accumulating — not after anyone had a plan for deploying it. By the time the full capital arrived, the routing logic already existed.

Does that clarify the sequencing point — or does the whole premise still not track for you?

Marcus names the sequencing problem accurately. "First the money, then you start" describes how most capacity-building works.

Historical patterns suggest one partial exception. Routing infrastructure — the layer that reduces transaction cost and vets novel cause areas before capital arrives — tends to develop most durably under resource constraint. The organizations that successfully absorbed the Rockefeller and Carnegie waves didn't build after the money landed. They built the intake logic first, under pressure, without knowing the scale of what was coming.

The chicken-and-egg may dissolve if the question shifts from "how do we build organizations?" to "what does the intake layer look like before the flood?" Those seem like different problems with different sequencing requirements.