2009-11 to 2012-04: The Second Leg — US Housing Markets Grind Lower

On November 1, 2009, the US 10-City Composite index closed at 157.58. By any quick read, the worst had passed — the initial subprime crash was behind, and government stimulus was flowing. But traders who looked closer saw a market that had not found a floor. It was grinding lower, and would do so for thirty more months.

The macro context for this extended leg down is well documented: unemployment remained elevated above 9%, millions of homeowners were underwater on their mortgages, and bank lending standards tightened to levels not seen in decades. These three forces created a persistent headwind that prevented any meaningful recovery in housing values, even as the broader financial system had stabilized.

The mechanism of continued decline was deceptively simple: demand was structurally suppressed while the supply pipeline — foreclosed properties entering the market — continued to expand. The Federal Reserve held rates near zero, but creditworthy buyers were scarce and speculative demand had evaporated. This created a slow-motion clearing process with no visible catalyst for reversal.

City-level performance diverged dramatically. Atlanta housing fell 22.7%, the steepest decline in this cluster, with a maximum drawdown that reached 25.0% — meaning a participant holding from the worst entry point would have endured a quarter of their capital being erased before any recovery. Minneapolis fell 11.1% with a 16.6% drawdown, and Seattle dropped 9.9% with a 13.2% drawdown. Meanwhile, San Francisco fell only 4.7% with a 13.0% drawdown, and Boston recorded the mildest outcome: a 4.4% decline and an 8.1% drawdown.

The US 20-City Composite Historical Replay: 2009-11 to 2012-04 captures this divergence through 30 monthly bars, with six embedded decision points including two marked as critical inflection moments. The index closed at 146.17 in November 2009 and fell to 135.98 by April 2012 — a 7.0% net decline — but the maximum drawdown of 9.9% reveals the true depth of the trough. Single-month moves never exceeded 1.4%, presenting participants with a grinding rather than crashing market.

Atlanta's experience was the starkest illustration of concentrated local pain. The Atlanta Historical Replay: 2009-11 to 2012-04 shows an index that fell from 109.29 to 84.48 over the same window. The headline 22.7% loss understates the severity: a 25.0% maximum drawdown meant that even a mid-sequence entry could result in devastating losses before any bounce. The largest single-month move reached 5.9%, a sudden snap that could breach stop-losses set too loosely. By contrast, the Boston Historical Replay: 2009-11 to 2012-04 presents a market that fell from 153.98 to 147.16 — only 4.4% — with an 8.1% maximum drawdown and a largest monthly move of 2.6%.

For markets, this period established a template for how housing bottoms form: not with capitulation candles but with an extended sideways-to-lower grind that exhausts both bulls and bears. The data shows that even in the worst-performing cities, there was no single dramatic flush — only the relentless compounding of excess supply. The 10-City Composite lost 4.5% net, yet the max drawdown was 6.8%; the gap between the two is the hidden cost of uncertainty.

For financial institutions, the episode exposed weaknesses in how risk models handled the transition from acute crisis to chronic stagnation. Standard deviation-based VaR metrics, calibrated on pre-crisis volatility, failed to capture the persistent tail risk of a market that grinds sideways in loss territory. Post-2009, stress-testing frameworks across mortgage-backed securities desks and real estate investment trusts began incorporating scenario analysis across multi-year drawdown paths rather than single-period volatility estimates.

For decision trainers working today, the central lesson is about managing positions through the second leg of a crisis when headlines say the worst is over. Atlanta's 25.0% drawdown was not a black swan — it was a predictable outcome of buying too early in a grinding market where supply continued to mount. The US 20-City Composite replay illustrates a different lesson: a headline decline of only 7.0% still involved a 9.9% maximum drawdown, meaning that even in a relatively mild market, the gap between entry point and trough can surprise unwary participants. The discipline to set drawdown thresholds before entering, and to treat sideways grinding as an active warning signal rather than a reason for complacency, is the capability this historical window trains most directly.