By Tim McLaughlin
BOSTON (Reuters) - When investors unloaded their shares in Hewlett-Packard Co. <HPQ.N> last year, the computer giant's rank-and-file did the opposite. They bought nearly four times as much stock as they did in the previous year despite a raft of bad news.
It was a smart bet, and outside investors would have done even better than the insiders had they followed employees into the stock.
HP's stock is trading at a 27 percent premium over the average $17 a share price paid by workers during fiscal 2012. HP shares are up 53 percent since the company disclosed the results of its ESPP activity in late December.
Following a herd of thousands of HP workers does not seem like it would be a sound trading strategy. But when it comes to the buying patterns inside the discount employee stock purchase programs (ESPPs) offered by U.S. corporations to their employees, the wisdom of a crowd can be a barometer of future stock performance.
Recent academic research by Arizona State University's Ilona Babenko and Rik Sen of The Hong Kong University of Science and Technology found that companies in the top half of aggregate ESPP purchases outperform those in the bottom half by up to 8 percent in the year after purchases.
"Since ESPP purchases reflect the decisions of thousands of employees...they can provide a reliable signal of future performance," the researchers' working paper said.
SWEETENING THE DEAL
Those findings could become more pertinent going forward, as many firms now are sweetening the terms of their ESPPs, according to Kevin Barry, a stock plan services executive at Fidelity Investments.
ESPPs have been one of the most overlooked aspects of stock-based compensation for employees, Barry said, but more firms are offering workers larger discounts now to attract and retain employees. In some cases, the discount is being raised to 15 percent or more from the previous level of 5 percent. Employees fund their share purchases through after-tax payroll deductions.
Fidelity recently found that 51 percent of the companies it surveyed indicated they intend to modify their ESPP in the next two-to-three years.
Despite the obvious benefits of buying stock at a 15 percent discount, many eligible employees don't bother, said Kevin Brennan, president of U.S. share plans at Computershare, an administrator to 330 ESPPs. He said only 30 percent to 40 percent of the eligible employee population participates in ESPPs with a 15 percent discount. Participation rates decline if the discount rate is lower.
Meanwhile, Babenko and Sen's research on (ESPPs) found that increased buying by lower-level employees is associated with a higher likelihood of favorable events, such as becoming a takeover target.
Declines in ESPP participation also can signal a negative development like an earnings restatement.
Babenko, a finance professor at ASU, said lower-level employees find out plenty about their companies through their everyday interaction with customers, outside vendors, senior managers and their co-workers.
"I didn't believe initially that non executives had so much information," Babenko said. Most academic research has focused on the predictive power of trades made by senior corporate executives.
But worker bees could provide a better signal. Unlike senior executives, they do not have stock ownership requirements or complex estate planning strategies, which can trigger trades that have nothing to do with a company's future prospects.
For example, ESPP activity at optical telecom Infinera Corp <INFN.O> increased significantly in 2012 as employees bought 26 percent more shares than the previous year. If outside investors bought Infinera shares once the ESPP activity was disclosed on March 5, they would have netted a 35 percent gain.
'PROFITABLE TRADING STRATEGIES'
For outsiders, the details of corporate ESPPs are disclosed in filings with the U.S. Securities and Exchange Commission. A typical ESPP offers workers a chance to buy company stock at a discount of 5 percent to 15 percent. The offering period can be three-to-six months and the stock is purchased on set dates.
In the case of HP, 21,000 workers, out of 301,000 eligible, bought 6.21 million shares at a 5 percent discount for an average price of $17, according to disclosures around fiscal 2012 activity. The discount did not change from fiscal 2011, when 1.75 million shares were purchased at an average price of $25 by 18,000 ESPP participants.
In most plans, there is a holding period that prevents employees from making quick sales. In fact, Babenko said their research found that ESPP participation is a stronger predictor of future returns for companies with sales restrictions.
"These results provide further evidence that purchases through ESPPs reflect valuable information that has not been incorporated into the firm's stock price," Babenko and Sen wrote in their research paper. "...If the information on ESPP participation is not quickly and efficiently incorporated into prices, it might be possible to design profitable trading strategies."
Dan Walter, chief executive of Performensation, a compensation consulting firm, expressed doubt that an increase in ESPP activity could signal a stock's rise. He said companies can boost participation by simply providing better communication about the merits of their ESPPs, or sweetening the terms.
Last June, participation in Hologic Inc's ESPP, for example, tripled to more than 900 employees in the first enrollment period after the company announced a more generous discount. Hologic <HOLX.O> shares are up 16 percent since the first ESPP offering period began on July 1, but the Standard & Poor's 500 stock index has returned more than 21 percent over that same period.
(Reporting By Tim McLaughlin; Editing by Linda Stern and Leslie Gevirtz)