⚖️ US Wealth & Income Distribution
🗺️ US Income Distribution by Metro click a city marker → opens that metro's ZIP income map in a new window
New York Los Angeles Chicago Dallas Houston Miami Washington Atlanta Philadelphia Phoenix Boston Riverside San Francisco Detroit Seattle Minneapolis Tampa San Diego Denver Orlando Charlotte Baltimore St. Louis San Antonio Austin Portland Sacramento Pittsburgh Las Vegas Cincinnati Kansas City Columbus Indianapolis Cleveland Nashville San Jose Virginia Beach Jacksonville Providence Milwaukee Raleigh Oklahoma City Louisville/Jefferson County Richmond Memphis Salt Lake City Birmingham Fresno Grand Rapids
Methodology

Signal pipeline. Each series is first reduced to a unitless value, then passed through an ARIMA(1,1,0) model (one difference to detrend, one autoregressive lag). The forecast residual — the part of the latest move the model did not predict — is standardized by its rolling volatility into a normal deviate (the σ on each card), ranked against its own full history as an empirical percentile, and confirmed by a 3-period same-sign streak before a green / red / amber color is committed.

De-dollarization. Dollar magnitudes are shown primarily as ratios to the real economy — debt/GDP, trade balance/GDP, loans/GDP, net worth/GDP, medical cost/GDP. Expressing a stock or flow in units of the economy that produces it makes the reading robust to dollar debasement: both the numerator and the denominator are priced in the same unit, so a rise in the price level cancels out of the ratio. Pragmatic de-dollarization, not revealed truth.

Reading the cards. The badge is the current σ/percentile; the trace is the de-dollarized level over time. An amber-pulsing trace means the long-term slope and the near-term ARIMA reading disagree — a genuine ambiguity, shown honestly rather than forced one way.