In Asset Markets, “Big Players” Induce Herding Since Traders Trade with Other People's Money

The trouble with discretionary market interventions is not that they come from government but rather that government is a Big Player. Bigness does not a Big Player make, but it helps. To be a Big Player it is not enough to be able to influence the market, to be largely immune from competitive pressures, or even to meet both conditions. A Big Player is anyone who habitually exercises discretionary power to influence the market while himself remaining wholly or largely immune from the discipline of profit and loss. Finance ministers and central bankers are paradigmatic Big Players, but only when exercising discretionary power. Private actors, too, may be Big Players. Each seller in oligopoly is one, which is what makes modeling oligopoly so difficult.

Our claim may seem unreasonable. Surely the authorities exercising discretion have an objective function or reaction function. Cannot such a function be set forth with just as much (or little) confidence as the equations describing underlying conditions of demand and supply? No: the concept of objective or reaction function cannot be put to good use.

In exercising discretion, a central bank commits itself to nothing but what it deems best in view of information available at the time of each decision, interpreting and weighing each bit of information as it then chooses. The variables in the bank’s reaction function and the weights given to each, as well as the function’s mathematical form, may thus change from decision to decision. Were changes in the function ruled out, the bank would be following a rule instead of exercising discretion. The econometric problem of identifying a changeable reaction function is necessarily insuperable, and assuming its existence is useless for guiding expectations.

We may put the matter in another way. A decisionmaker employing discretion learns over time. ‘‘Real’’ learning implies more than the accretion of information; it implies a change in knowledge and in how news is ‘‘framed,’’ ‘‘filtered,’’ or interpreted. Real people are always somewhat surprising. Just as we would not expect to model our loved ones mathematically, so we should not expect to succeed in modeling central-bank presidents. The behavior of Big Players, to paraphrase Keynes, affords no scientific basis whatever on which to form any calculable reaction function.We simply do not know (Keynes, 1973).

In asset markets, Big Players induce herding. Many of the traders in an asset market are trading with other people’s money (Scharfstein and Stein, 1990). Good and bad trades affect their reputations as hired hands. Reputation typically hinges on relative performance. If most S&Ls are failing, including yours, you are not considered a bad banker, just the unlucky victim of an industry crisis. When you go along with the common wisdom and things go bad, you can share the blame with everyone else and leave your relative performance still looking pretty good. If things go well instead, so much the better. The really dangerous thing is to defy common wisdom and lose; your reputation is shot and you must look for a new line of work (Scharfstein and Stein, 1990).

—Roger Koppl and Leland B. Yeager, “Big Players and Herding in Asset Markets: The Case of the Russian Ruble,” Explorations in Economic History 33, no. 3 (July 1996): 368-369).