Sunday, April 6, 2014
Say I’m the CEO of a company with toxic waste buried in our back yard. One day on my way into work, I spot John Travolta climbing over our fence holding a chemistry kit and a spade. After I press the button that releases the hounds, my next step should be…calling my broker? Until John files A Civil Action, I have a window in which I can sell my stock with no market discount.
That has to be a problem for deterrence theories of tort law, doesn’t it? Read on, friend.
Typically, the goal of an ex post price instrument is to force rationally forward-looking actors to internalize currently the expected future price of any externalities they produce. Even before the costs are paid, expected liabilities should be capitalized into share price or reflected in insurance premia, meaning that current shareholders bear the burden of future payouts. But the manager allowed to trade on non-public liability information in effect holds a put option, making her insensitive to future cost. Deterrence doesn’t work if the folks who call the shots can, ahem, bury their heads in the sand.
At first glance it seems like a puzzle why shareholders wouldn’t contract around this problem. After all, they’ll be the ones left holding the bag of waste. But shareholders don’t actually internalize all the expected future costs. Their exposure is the value of their stock, no more. Indeed, since shareholders can self-insure by diversifying their investment portfolio, it’s a bit of mystery why publicly-traded firms buy insurance at all. One answer may be precisely that it is the manager, with a large portion of her wealth tied up in the firm’s stock and future ability to pay her salary, who is risk-averse and so wants to sign the insurance contract.
In short, just as Carlton & Fischel (1983) argue, shareholders actually want managers to trade on inside information to the extent that such trading increases managers’ preference for risk taking to something more like the shareholders’ own. That was an efficient contract when the only risk was whether the manager’s choice of investments would pay off. But when the manager is taking risks that lead to outside harms, the rule no longer looks quite so appealing to society as a whole. So at a minimum, regulators should want to prohibit trading on non-public information about regulatory action or tort liability. (Or perhaps, as Macey (2007) sort of suggests, allow managers to trade, but only if they sell short, turning them into whistleblowers. This seems like a good plan only if the manager is trading on information about a disaster caused by someone else in the firm…)
Or maybe fixing insider trading rules is just a patch on a sinking ship. Let’s take a step back. Ex post liability rules may fail to fully deter the owners of firms wherever expected costs exceed equity capital. The only way for deterrence to operate at something close to full capacity is to impose individual penalties on managers. Firm-level liability accomplishes this by threatening the manager’s store of personal wealth tied up in the firm and, thanks to her risk aversion, perhaps by triggering the purchase of insurance that internalizes future costs for shareholders. But this is a Rube Goldberg device whose actual impact may vary depending on how the manager’s pay is structured, and perhaps on her behavioral response to that structure. Wouldn’t it make more sense for all deterrence to be aimed directly at the manager? And if victim compensation is also important, perhaps to mandate liability insurance?
Or have I got that all wrong?