FOR IMMEDIATE RELEASE
For More Information, Contact: Barbara Buell at [email protected] or 650 723-3157
STANFORD--When an initial public offering of stock hits the market, it is typically underpriced. Frequently the underwriter will buy shares to prop up the price and attract investors during the first few days of trading. Are these supported IPOs merely small offerings from fringe underwriters who are buttressing stocks at artificially high prices?
New evidence, analyzed by Stanford Business School Assistant Professor of Finance Manju Puri and N. R. Prabhala of the Yale School of Management, shows that supported IPOs are usually larger, have lower underwriter commissions, offer higher prices than unsupported stocks, and are not likely to be underwritten by lesser-known investment banks. "It's not the lemons that are supported," says Puri. "It's the big issues. Price support seems to have a role in helping after-market liquidity."
For about 60 years, Securities and Exchange Commission rules have exempted price supports from anti-manipulative measures on the grounds that they improve liquidity--the reason bankers often cite for the practice. Yet until now there has been little proof that liquidity justifies price support. "Our paper fills this gap by providing theory and empirical evidence for this often-cited but never-explained rationale," says Puri.
The researchers also debunk some common criticisms, based mostly on anecdotal evidence. One is that price support is a manipulative practice aimed at creating artificially high IPO prices. "We find no evidence that this is a systemic motivation for price support," says Puri. Another suggests that banks sometimes support stocks to help favored customers out of bad issues. If a stock price flags, the bank buys back the issue. Puri and Prabhala reject that idea because, they reason, if the stocks were dogs, prices would drop. The data showed that once price supports were withdrawn, the unsupported stocks actually drifted down in price more than the ones that had been supported, demonstrating that bankers had not propped up lemons. Finally, the researchers refute accusations that banks participate in underwriter reimbursement. That is, they supposedly require a larger gross spread -- the difference between the price at which the underwriter buys stock from the corporate issuer and the price at which it sells the stock, effectively the underwriter's commission -- in exchange for promising to buy back some of the stock after the offering. There's no evidence of higher commissions. "We find lower gross spreads rather than higher ones among supported stocks," says Puri.
Puri and Prabhala used rare data from SEC filings to analyze the effects of price supports. At one time, underwriters had to fill out a form detailing the amount and duration of their support. The requirement was discontinued in the early 1980s, and old SEC records were destroyed. However, the researchers were able to find a small but surviving group of these forms to analyze the returns and characteristics of 46 IPOs in 1981 and 1982. They also looked at a large group of 2,723 IPOs between 1985 and 1994 to see how they compared to the smaller group. What they found in both groups was that underwriters tend to support IPOs with a low initial price risk--issues whose prices they can pinpoint accurately. Puri and Prabhala's model verifies the hunch that price support provides the incentive for underwriters to improve the after-market liquidity of IPOs.
In a third test, the researchers looked at data from recent IPOs that were supported for three to five business days following the first offering. They compared the performance of those IPOs to public offerings made prior to passage of the 1934 Securities Act when there were no restrictions on price support and underwriters typically supported prices for about six weeks. They found that IPOs that were supported longer were less risky and were not as underpriced when they first came to market. They also found that supports explain no more than 4 percent of average initial IPO returns. That is, IPOs are underpriced, even after accounting for price supports. This is useful information for investors or issuing firms that want to know if shares are underpriced and, on average, by how much. "They are underpriced but not by as much as we thought," says Puri.
3/00
RSS