There are five factors for systemic risk.? Here they are:
- Asset size of the institution, including synthetic exposures.
- Degree of leverage of the institution, including synthetic exposures.
- Asset-Liability mismatch, particularly financing long assets with short liabilities (including derivatives and margin agreements — think of AIG, or mortgage REITs on repo).
- Degree to which the institutions owns financial companies equity or debt, or vice-versa, where other financial companies have claims on the institution in question.
- Riskiness of the assets owned by the institution in question.
Contributing to the risks include easy monetary policy, which can lead/has led? to the neglect of risk control.? Personally, if I were a regulator of systemic risk, I would throw my effort at companies that fit factors 1 and 2, and analyze them for the other three factors.
Systemic risk is layered levered credit risk. A lent to B, who lent to C, who lent to D, who financed a bunch of bad mortgages.
#5 is underwriting risk
#4 is connectedness risk
#3 is liquidity risk
#2 is financial risk
#1 is risk to the economy as a whole.
So when I read articles like this, or books about systemic risk by academics that are so bad that I don’t want to review them (set them to work picking fruit, it would be more valuable than what they currently do), I simply say systemic risk is easy.? Look at my five points.? You can eliminate systemic risk by:
- Breaking up the big banks. (1)
- Disallowing banks from owning the equity of other financials and vice-versa. (4)
- Forcing strict asset-liability matching at banks, and? (3)
- Sizing capital to the riskiness of loans made. (2,5)
- Move to double liability on banks — they can’t be limited liability corporations.? Investors and managers must have their net worth on the line for any losses.
This isn’t hard, but the banks will scream.? Let them scream, and let the stocks of the banks fall.? Banks take risks beyond what they ought to because of poor regulation.? They should be regulated well, and have lower returns on equity as a group.
I think that your suggestions would, in the long term, lead to
1. Identical geometric average growth
2. Less variance in growth (i.e., smoother/less acute up swings and downswings)
3. Longer business cycles
I also think that if your banking regulations would be tied to your tax proposal, it would further smooth and lengthen the business cycle (by reducing total leverage in the system and discouraging mal-investment).
The major barrier is, as always, the asymmetrical structure of political rewards. If you make policy with generally bad effects that are either felt latently or felt minimally by many people, there are no negative consequences (in terms of reelection or backlash). If, however, you make good policy that takes away something free for a small group, the backlash is instant and acute. I think a handful of bankers has more political clout than the other 300 million Americans.
A simpler and perhaps more achievable solution could be to:
Split apart the utility function of banks (checking, savings, loans, trust, etc.) from the casino function (derivatives, private equity functions, investment banking, etc.)
Put tighter regulation and FDIC production on the utility part. Perhaps we would need to break up the largest – not sure.
Let the casino function swim neck deep in risk if they wish, just ensure the principals and stock holds take 100% of the risk with no tax funds at risk.
Nope, sorry, it’s not even remotely that simple. For one thing, your list would almost certainly not have identified LTCM ahead of time. (They weren’t very highly leveraged until the very end, when asset prices plummeted.)
You cite “riskiness of the assets owned by the institution,” but who’s deciding which assets are “risky”? And how risky is too risky? Taking risks is what financial institutions do, so at some point you have to draw the line between “acceptably risky” and “too risky.” Where you draw that line isn’t just a minor detail, it’s fundamental to the entire process of identifying systemically important institutions. You can’t just assume that away. It makes you sound like one of those ivory tower academics.
Same goes for your proposal to “force strict asset-liability matching at banks.” Well, what constitutes “strict” asset-liability matching? That’s another core issue that you just ignore/assuming away.
It’s easy to write blog posts about systemic risk, but actually getting down into the nitty-gritty isn’t easy. It’s extremely complicated and nuanced. I’m surprised you’d write such a silly post.
#1: don’t bail out bondholders.
no systemic regulator necessary