Ever since the Madoff affair became public I’ve been thinking about Harley Granville-Barker’s play The Voysey Inheritance, which has many parallels with Madoff. (A family investment-management firm uses its clients’ money as its own, and uses incoming funds to make interest payments to existing clients, but is finally undone by redemption requests; there’s even an indelible scene where the father confesses all to his son, and explains his secret: his ill-gotten wealth made him trustworthy in the eyes of his clients.)
The Voysey ineritance, just like Madoff’s money-management operation, was ultimately a confidence game; it’s easy to see why the play was so attractive to David Mamet.
But really all finance is a confidence game, and just as Madoff dwarfs Voysey, so do total stock-market losses (in the tens of trillions of dollars) dwarf Madoff’s. So long as investors had faith in Voysey, or Madoff, or stocks, everything worked fine. It was only when they tried to take their money out in quantity that things imploded.
Edward Hadas calls this "the noble lie":
The noble lie is the foundation on which all banking is built — the ability of a bank’s depositors and borrowers both to consider the same funds as their own. It’s a lie because no bank, no matter how well capitalised, can always let each depositor cash every account…
The fiction of potentially unlimited withdrawal is not limited to bank accounts. Holders of the more advanced financial instruments — bonds, shares and derivatives — cannot all sell at once. If more than a few try, the market price drops sharply. If there is a stampede for the exit, the market disappears entirely.
"Stampede for the exit", of course, is also known under its more wonkish name: "global deleveraging". If everybody’s selling and nobody wants to buy, what you thought of as wealth — that number at the bottom of your brokerage statement, whether you have an honest broker or not — can evaporate with astonishing speed.
Which I think is the answer to Brad DeLong’s question of how on earth $20 trillion of wealth has been lost, when total loan defaults are only on the order of $2 trillion. The defaults sparked a deleveraging, and the deleveraging destroyed the confidence which was keeping asset prices aloft.
For example: if there’s just one person willing to pay $950 for my Google stock, then that’s how much it’s worth. But if that marginal buyer goes away, and there’s a general sentiment that people would rather have cash than Google stock, the price can fall precipitously without much if any news. DeLong tries to model the preference for cash over Google stock as a rise in such things as "liquidity discount" and "risk discount", but that’s not how it works in the real world: few people ever bought Google stock because they did the math and decided that the risk-adjusted present value of future dividends was $950. (Especially since Google doesn’t pay a dividend.)
Sure, DCF jockeys exist, but they don’t tend to be price-setters. Brad DeLong, who’s done a lot of original research on the equity risk premium, is basically in that camp: he buys stocks because he has faith in his own analysis. But most of us aren’t that clever: we just buy stocks out of some combination of greed and fear (that we won’t have enough money to live on, decades hence, if we don’t invest our money in ways which make it grow substantially).
In times of turmoil, we start worrying less about not having enough money in 20 years, and more about not having enough money in 20 weeks. (What if I lose my job? What if my investments fall further?) So we sell our investments.
Can this be modelled as an increase in the liquidity discount? Maybe, but in that case Brad has already answered his own question:
As long as we love our children as ourselves (and most of us do) and as long as we have access to and can credibly pledge collateral for financial transactions (and we can) the magnitude of the liquidity discount should be roughly equal to the technologically and organizationally driven rate of labor productivity growth divided by the intertemporal elasticity of substitution. The technologically and organizationally driven rate of labor productivity growth is a fairly steady 2 percent per year. The intertemporal elasticity of substitution is in the range from 1/2 to 1.
The intertemporal elasticity of substitution might normally be close to 1, but it sure isn’t there right now — not for those of us without tenure, anyway. It came down a lot in the summer of 2007, when the commercial paper market first started seizing up, and it’s even lower now. Ask any hedge fund manager dealing with massive redemptions — it might not be rational to buy with a long time horizon and then suddenly decide to liquidate, but this is not a rational market, and it hasn’t been for some time.
There’s also a strong feedback loop here. In normal markets, the more that prices fall, the more people want to buy. In financial markets, during a time of crisis, the more that prices fall, the more people want to sell. The intertemporal elasticity of substitution isn’t a constant: it’s a variable, which falls along with the market. And the people who don’t adjust their discount rate fast enough to new realities end up, like Bill Miller, getting crushed.
Especially if you’re buying financials and other confidence stocks, you need a lot of other people to be buying them too, otherwise they have a tendency to go to zero. More generally, whenever lenders lose confidence in a company and refuse to refinance its debt, shareholders are likely to find themselves severely diluted at best, and quite possibly wiped out entirely.
It’s entirely reasonable to draw a distinction between Mr Voysey and Bernie Madoff — the men with the ignoble lies — and the more noble lies underlying the stock market. But deleveraging is no respecter of nobility. Which is why all of us are now suffering, not just those who invested with Bernie.