Someone asked me “so did Goldman deceive ACA and investors?” I don’t know. Am I being coy or hiding the ball? No, I think that fraud cases like any litigation should revolve around the facts, not deductive logic and theory. And we are just starting to get facts.
Saying investors should have known these were risky securities and declaring “bubble” after the fact, doesn’t speak to whether there was fraud in this particular case. Just as noting that Goldman had many potential conflicts from playing every angle of a transaction doesn’t mean that there was deception.
It is a little strange to be an academic and call for focusing on facts not theory, particularly since I love me my theory. Theory may not be able to answer whether there was fraud in a particular case, but I think it can help us think about overall patterns in regulation and regulatory history. I wrote a 2006 law review article: “The Next Epidemic: Bubbles and the Growth and Decay of Securities Regulation” that talks about issues I am still grappling with. Some of the themes in the article:
- An attempt to explain the political economy of the pattern of “deregulation” (which means more than just the repeal of statutes, but looser interpretations of law, laxer enforcement, even government stimulation of markets) during market booms followed by re-regulation after the bust.
- An analysis of how securities law compliance – particularly with antifraud rules -- starts to decay during market booms.
This latter point is worth considering as we move from discussing individual litigation back to rethinking regulation. Post-hoc litigation strikes me as a poor substitute for regulation. One reason is that litigation is inherently pro-cyclical – that is, it only bites after the crash has occurred. Anti-fraud claims have little bite during market booms for any number of reasons, including it is hard to uncover fraud when prices are rising, there may not be a cause of action until prices drop, and regulatory budgets are stretched thin.
The deterrence value of fraud litgation after the fact is lessened by the time value of money, the fact that many individuals with responsibility will escape the net (for example by jumping a firm's ship in time) and the fact that expected liability (damages + other losses) * (probability of getting caught) may be a lot less than the gains.
We could just ratchet up the punishment or lower the standards of culpability to compensate -- "shoot them all" -- but that isn't exactly due process and it likely isn't effective either.
What's the upshot? To best prevent "the next time" - let's keep our eyes on rethinking regulation.
Financial Crisis, Legal Scholarship, Legal Theory, Securities | Bookmark
TrackBack URL for this entry:
https://www.typepad.com/services/trackback/6a00d8345157d569e20133eccc6722970b
Links to weblogs that reference "Just the Facts” vs. Cycles of Fraud:

Sun | Mon | Tue | Wed | Thu | Fri | Sat |
---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | ||
6 | 7 | 8 | 9 | 10 | 11 | 12 |
13 | 14 | 15 | 16 | 17 | 18 | 19 |
20 | 21 | 22 | 23 | 24 | 25 | 26 |
27 | 28 | 29 | 30 | 31 |
