Two points: 1. The boom-bust-boom was a feature of many of the housing markets in the 2000s-2020s, both across US regions and internationally. 2. The fact that there was a rebound does not disprove that valuations were extremely stretched in 2005-06. Most bubbles are based on optimism that has some fundamental kernel of truth but gets massively exaggerated (see Kindleberger...). The key issue in the 2000s boom was not just the bubble in house prices but the massive expansion in household leverage.

I've really been enjoying @DavidBeckworth @Macro_Musings podcast lately (we are in a golden age of econ podcasts!). A recent episode with former CEA chair Tyler Goodspeed takes on a fascinating question: Are recessions driven by random shocks (bad luck) or the product of inherently cyclical boom-bust phenomena? Goodspeed's new book argues it's the former. The episode is good fun and very interesting. mercatus.org/macro-musings/t… Two thoughts I wanted to share in reaction, given that this is a topic close to my heart: 1) Somewhat ironically, even though we call short-run fluctuations business *cycles*, the random shocks view is close to the mainstream approach in academic macro. A long tradition in time series macro going back to Granger's work in the 60s looks at the properties of macroeconomic time series (spectral densities!) and finds that there is not much evidence of stable, regular cycles in macro series. Modern DSGE macro models expansions/recessions as the product of relatively persistent shocks, rather than endogenous cycles. The view that business cycles are the product of endogenous boom-bust dynamics is slightly more on the periphery of macro than people might think! 2) That said, I would push back a bit on the strongest version of the random shocks view. I agree that a lot of variation at business cycle frequencies comes from random shocks. But there is quite a bit of evidence that the endogenous boom-bust story has some truth, at least during specific episodes featuring credit and financial booms preceding financial crisis recession. There is quite a bit of evidence that rapid expansions in credit, especially to certain risky sectors like real estate, coupled with booming prices of real estate and other assets, are a strong predictor of future downturns and financial crises. Financial crises are not unpredictable "bolts from the blue." While the exact timing of a crisis is very hard to predict, there are certain vulnerabilities that make financial crises and deep recessions more likely. These dynamics don't hold mechanically during every business cycle, but they do matter for understanding major downturns involving financial instability. A few examples off the top of my head include: 2008 in the US, Spain, Ireland, etc; the Japanese financial crisis after the 1980s boom; the crises and deep recessions in the Nordic economies of late 1980s; emerging market crises like Mexico 1994 and Thailand 1997; or historical crises like Australia’s housing boom and bust in the late 1880s, the panic of 1893, and perhaps the US Great Depression.

For researchers interested in U.S. banking and finance, we are sharing a new data resource! It contains information on bank balance sheets, bank runs, and bank failures from the 19th century to the present:

Dataset #2: Historical Call Reports. Provides bank-level balance sheets for all national banks from 1863 to 1941, digitized from the Annual Report of the Comptroller of the Currency. Includes over 370,000 bank-year observations for over 14,000 banks. finhist.com/historical-call/…