Securities markets and government regulators, especially in times of financial crisis like these, eye each other warily, each trying to ascertain what the other knows and what the other is going to do.
The government looks at securities prices as an important source of information about when it should intervene to rescue a bank, like Citigroup, or a major industrial company, like General Motors. Meanwhile, investors are doing all they can to figure out not just the economic prospects of companies, but also the probability that government will intervene in troubled firms. If investors expect the government to step in to save a firm, they might bid up the security’s price to a level not warranted by the fundamentals.
In doing so, the market would undermine the usefulness of the security’s price to regulators as an indicator of whether action is needed. This problem, a “hall-of-mirrors” effect, is examined by Wharton finance professors Itay Goldstein and Philip Bond, along with Edward Simpson Prescott of the Federal Reserve Bank of Richmond, in their paper, “Market-Based Corrective Actions,” forthcoming in the Review of Financial Studies.
Before “regulators intervene in Citibank they want to see that the price is very low. The fact that the price is very low … gives them reason to intervene,” Goldstein said. “If the market expects intervention, the price might go up.” But if that happens, the government might not intervene, and a bank that desperately needs help will not get it. “There is a feedback loop that makes it very difficult to pin down what is going to happen.”
The dynamic can have serious consequences for the market. Prices become divorced from economic fundamentals because of anticipated government intervention, and investors might be unwilling to buy shares if they fear the ensuing price increase would lead the government to abandon its intervention plans.
The past year has seen many examples of such turmoil. In one week last July, the price of shares in Fannie Mae nearly doubled to $14.13 from $7.07 on hopes of a government rescue for the mortgage finance company. Similarly, banking giant Citigroup has seen its shares bounce wildly as investors tried to get a handle on whether the government would take some type of corrective action to stabilize the banking behemoth.
The hall-of-mirrors problem occurs when government action is expected to increase the value of a security, as was the case last fall when Congress was debating the $700 billion Troubled Asset Relief Program. It is not an issue when government action, such as nationalization, threatens to wipe out equity holders. In that case, there is no feedback loop because the stock becomes virtually worthless.
Cutting the Loop
The guessing game could be ended by improved communication between the market and the government, with the responsibility lying largely with government, Goldstein and Bond suggest. “You have to cut one side of the loop,” they write. One way to accomplish this would be for government to state clearly what it will do under particular circumstances, instead of acting in an apparently ad hoc manner as it has done during the financial crisis. Troubles at Bear Stearns, Lehman Brothers and American International Group each elicited different responses from regulators. That added to the tremendous uncertainty and turmoil in the market.
In the case of banks especially, government regulators have access to direct information about the condition of firms, so why shouldn’t they act exclusively on that? They should — at times, according to Goldstein and Bond, who add that it is crucial for regulators to have their own sources of information so that they do not rely too heavily on the markets for data. Some argue that market prices should play a bigger role in the regulatory regime, but the professors’ research suggests that such a move would be fraught with problems because of the feedback loop.
Nevertheless, Goldstein and Bond do see an important role for market-based information. “We believe there is additional information in the market that the government does not have direct access to,” Goldstein said. That information comes from thousands of market participants bringing their own views to bear on thousands of different companies, creating a broader and — in some ways — more detailed view than government can achieve through its direct information from the banks’ books.
As it is, government can ascertain the market’s views of a company only through securities prices, which, unfortunately, can be tainted by the market’s expectation that government is likely to intervene. Ideally, government could distinguish between the portion of a security price attributable to fundamentals and the portion attributable to expectations of intervention.
For example, suppose investors think a bank’s assets are worth $5 a share without a government bailout and $10 a share with a government bailout. If investors think a bailout is likely, the market price would be $10 a share. Unfortunately, there is no way for government to know that the expectation of a bailout is built into the price. That leaves regulators unable to tell what the market thinks the bank is worth based on economic fundamentals.
Government can make its process of gleaning information from market prices more efficient if it observes the market’s treatment of different securities issued by a single company at the same time. For example, it might observe not just stock prices, but also debt prices to arrive at a more precise understanding of conditions at a bank.
With a similar goal in mind, Goldstein and Bond propose the establishment of a prediction market on government interventions, similar to those that allow traders to bet on the outcome of presidential elections. Such a market would exist alongside the securities market, and it would allow investors to buy and sell options that are essentially bets on whether or not the government will rescue a company.
For example, said Bond, if Investor A believes that the government will bail out a specific bank by a certain date, he can sell an option to Investor B, who believes the government will not step in by the set date. If the government steps in to rescue the firm, Investor B must pay Investor A the value of the option. If the government does step in, Investor A must pay Investor B. Higher demand for options on one side of the bet would drive up prices for those options, providing a very clear indication of market sentiment. “If you had such a market, the government, by looking at prices in the prediction market and in the security market, could tell what the market really thinks,” Goldstein said.
He pointed out that the hall-of-mirrors effect bedevils not only government, but also corporate managers trying to decide whether or not to make an acquisition, and boards of directors trying to decide whether to replace a chief executive. “If the board knows that the CEO is of low quality, it will replace him. This corrective action will benefit the shareholders of the firm and thus increase the price of its shares,” Bond and Goldstein write. “So, inferring information from the price about the quality of the CEO is a challenge: A moderate price may indicate either that the CEO is bad and the board is expected to intervene and replace him, or that the CEO is not bad enough to justify intervention.”
That leaves Goldstein wondering how many bad CEOs are left in the job because “the expectation that the CEO will be replaced is already priced in, meaning that the board doesn’t get the right information.”