Correlation assumptions in CDO design collapsed when housing markets proved linked.
Evidence
The idea is that if these different real-estate markets are already uncorrelated, then cutting them up and recombining them should make them super uncorrelated. If the house in Baltimore defaults, the equity tranche holders are wiped out, but that loss is isolated from the holders of the loans on the condominiums and the Upper East Side penthouses. Safety combined with greater returns (at least for those holding the risk) led to an explosion of demand for these securities as US housing prices nearly doubled between 1997 and 2008. They were no longer a simple equity hedge. They became objects of investors’ desire in their own right. — Mark Blyth, Austerity, 91
Categorization
- XREF: The uncorrelation assumption failing is the mechanism that later forces the bailout/too-big-to-fail problem the chapter title flags. — connects to: Security design boosted cross-market correlation and turned local losses into systemic collapse. — Mark Blyth, Austerity: The History of a Dangerous…, 96; Housing price declines triggered the 2008 financial crisis — Nick Srnicek, Platform Capitalism, 31; Diversification across geographic mortgage pools created rather than removed correlation. — Mark Blyth, Austerity: The History of a Dangerous…, 97 (+2 more)
Broader Connections
- Security design boosted cross-market correlation and turned local losses into systemic collapse.
#corroborates(0.95) — Mark Blyth, Austerity, 96 - Diversification across geographic mortgage pools created rather than removed correlation.
#expands(0.90) — Mark Blyth, Austerity, 97 - Housing price declines triggered the 2008 financial crisis
#qualifies(0.75) — Nick Srnicek, Platform Capitalism, 31 - Claiming that correlated market crises are statistically impossible is plainly false
#corroborates(0.75) — Mark Blyth, Austerity, 104 - LTCM’s near-collapse was commonly blamed on blind faith in financial models
#exemplifies(0.75) — Donald MacKenzie, An Engine, Not a Camera, 152