Dispatches

A dated prediction about the August 7 jobs report, checked against the coverage that followed — and corrected where it was too blunt.

Truth, arranged strategically, can mislead more effectively than falsehood.

Predictive framing isn't a hunch. It's a method, and it has a test.

A prediction, filed before the September summit, about how media coverage of Xi Jinping's visit to Washington will split into two contradictory frames — dealmaker and predator — depending on which audience the outlet serves.

Western analysts spent thirty years building sophisticated models of China's economy — and kept getting it wrong. The reason is a single sentence: when the Party says GDP must be 5%, local officials deliver 5%, even if they have to build a highway to nowhere. Markets respond to price signals. China responds to political signals. For three decades those two things pointed in the same direction — and the models looked right. Now the inputs that made the system work are exhausted, and the gap between the political signal and economic reality is widening in ways no GDP figure can capture. Understanding that gap is the prerequisite for understanding everything else about where China is headed.

In August 2023, China's National Bureau of Statistics stopped publishing youth unemployment data. No announcement, no methodology change — the series simply disappeared after hitting a record 21.3%. Every serious analyst's estimate went up that day. This is the counterintuitive logic of analyzing information-managed economies: silence is not the absence of data. It is the system's highest-cost signal, revealing exactly what the management most needs to suppress. Reading silence — which topics go quiet, which series enter "methodology review," which academic subjects become sensitive — is the beginning of serious China analysis, and requires no privileged access.

My grandfather said it the way people say things they have tested against experience and found true: perception supersedes reality. He was not describing a cognitive flaw but an operating condition — how institutions actually function under social judgment and political incentive. The G7 issues communiques while its enforcement capacity erodes. NATO produces spending pledges insufficient to close its credibility gap. The Fed signals cuts then holds. Each is the managed appearance of a capability that is absent, diminished, or misrepresented, sustained because the audience rewards appearance and honest accounting costs more in the short run. Until the perception stops working — and the adjustment is rarely gradual.