Massive jobs revision, Kamala wealth tax, polls vs prediction markets, end of race-based admissions

The BLS revised nonfarm payrolls down by 818,000 jobs for the year ending March 2024, and total downward restatements over ~12–14 months may exceed 1.2 million jobs. The besties argue this pattern of one‑direction (always downward) revisions undermines trust in headline reports and means market participants have been trading billions on data that later prove inaccurate. They call for modernized, possibly crowdsourced or payroll‑system–based measurement to make U.S. labor stats as real‑time and reliable as financial markets require.

With unemployment drifting above the Fed’s 4% target and inflation near a “2‑handle,” Friedberg cites market-implied odds of 75–100 bps of cuts by year‑end. Chamath suggests the jobs revision materially increases the justification for a 50 bps cut in September. Sacks warns the economy is already being “narrowly kept out of recession” by ~US$2T annual deficits, arguing this is an unsustainable way to maintain a soft landing.

MIT’s first race‑blind class shows Asian‑American enrollment increasing from 41% to 47%, with declines in Black and Latino shares, igniting debate about fairness. Chamath argues admissions should target students obsessed with MIT’s core disciplines (physics, engineering, hard sciences) rather than applicants seeking a credential. Sacks and Friedberg endorse a color‑blind meritocracy that may still weigh socioeconomic hardship, while warning that past quota‑like engineering hurt talented Asian students and that race is a poor proxy for disadvantage.

The panel criticizes “top school” fetishization, contending many public tech schools (e.g., Virginia Tech, NC State, Waterloo, Drexel, Iowa State) produce hungrier, less ideological, more technically strong graduates. They champion co‑op and apprenticeship models that integrate real work with study, and cite GitHub, real project portfolios, and performance in the job as better signals than brand‑name diplomas. In their own companies and funds, they deliberately recruit from non‑elite schools and train talent internally.

Friedberg emphasizes that both polls and betting markets give probability distributions that are constantly updated as events and information change—they’re not binary guarantees. Nate Silver’s models weight pollsters by historical accuracy and run thousands of simulations; prediction markets reflect bettors with skin in the game but can be thin and manipulable in low‑liquidity races like VP picks. Sacks uses markets more to track sentiment shifts (e.g., Polymarket swinging back to Trump even during the DNC) than to treat them as deterministic forecasts.