David Moss wrote a good article in Harvard Magazine about systemic risk and regulation; it’s based on an earlier working paper of his. The problem statement is not particularly original, but very clearly put: Depression-era regulation brought an end to recurring financial crises because deposit insurance was combined with strict prudential regulation to guard against moral hazard. Half a century of stability, however, was undermined by a philosophy that regulation was not only unnecessary but harmful in financial markets, at precisely the same time that financial institutions were becoming dramatically larger in proportion to the economy as a whole. For example, as Moss points out, “the assets of the nation’s security brokers and dealers increased from $45 billion (1.6 percent of gross domestic product) in 1980 to $262 billion (4.5 percent of GDP) in 1990 to more than $3 trillion (22 percent of GDP) in 2007.”
The problem today, in Moss’s words, is that “implicit guarantees are particularly dangerous because they are typically open-ended, not always tightly linked to careful risk monitoring (regulation), and almost impossible to eliminate once in place.” The solutions he outlines basically boil down to renewing that tradeoff, so that government guarantees are explicit and financial institutions pay for them through more stringent regulation and cash.