Research

Where are the shareholders' mansions?

CEOs' home purchases, stock sales and subsequent company performance management

2007-05-09

To comment on this and other stories go to the Knowledge@W. P. Carey Blog. While U.S. corporate CEOs acquire enormous estates, shareholders are being left out in the cold. In a new study, W. P. Carey finance professor Crocker Liu, director of the school's Center for Real Estate Theory and Practice, and David Yermack, professor of finance at New York University, examined the size, cost, and financing of the primary residences of CEOs of the Standard & Poor's 500.

Liu and Yermack discovered important correlations with future company stock performance: The larger and more costly the home, the worse the stock performance. Also, when a CEO liquidates company shares or options to finance a home purchase, even if the sale represents a small share of the CEO's total holdings, it bodes poorly for future company performance.

Conspicuous consumption and conspicuous waste

In his "Theory of the Leisure Class," sociologist Thorstein Veblen coined the terms "conspicuous consumption" and "conspicuous waste." Impressive possessions such as homes, vehicles, clothes, jewelry, and art can be status symbols that signal personal wealth, power, and importance. Buyers use these purchases to impress and intimidate their peers and competitors. They hold them up as evidence of their position.

When advantage is not taken of another's ability to produce something cheaper, quicker, better, or with less resources, waste results and sometimes it can be conspicuous. On a corporate level, Liu explained, when CEOs buy new homes, they may be indicating commitment to their firms. Residential real estate is costly to acquire, impossible to move, and it generally can be liquidated only with time and expense.

Investing in company shares, agreeing to a deferred or performance-contingent compensation package, or taking some other costly and/or irreversible action are other commitment devices that bond companies and executives — something valued by boards, shareholders, and other constituencies such as bondholders and labor.

The executive's commitment provides a credible signal to these groups of the executive's high ability level or desire to work within the organization — perhaps because of confidence in the organization's future course. Alternatively, Liu said, CEO home purchases may indicate entrenchment. These CEOs feel secure in their positions and are not concerned with the possibility of removal by their boards, and are therefore unconcerned about the costs of liquidating the property.

Entrenched CEOs perceive themselves to be immune from discipline, and are uninterested in maintaining or improving their performance to attract outside offers. Acquiring an expensive house would provide a dual stream of benefits to an entrenched CEO: utility in consumption, and also a public signal about the executive's status and security. If CEO acquisitions of grand houses are consistent with entrenchment, they should augur badly for future shareholder returns.

CEO estates

Liu and Yermack searched online real estate property tax and deed transfer records, Lexis-Nexis, voter registration records, CEO employment contracts, Federal Election Commission campaign donation reports, and various Internet "people search engines" to identify the principal residences of CEOs in the S&P 500 at the end of 2004.

Given the strategies sometimes used by wealthy people to shift ownership of property for privacy, tax, estate planning, alimony, or other reasons, the researchers often had to work backward through a sequence of intra-family transactions involving the CEO, spouse, and various trusts. They were able to pinpoint residences for 488. The remaining 12 CEOs may be renters who own no property or may live outside the United States (one of the 12 works for a company with a Bermuda headquarters).

Liu explained that the team used www.zillow.com and www.reply.com to assess market values and property characteristics as of late 2006. Unfortunately, historical estimates of market values are not available, he said, but assuming that the market values prevailing in late 2006 are proportional to those at the end of 2004, the data is at least consistent.

The CEO residences are large: The median home has 11 rooms, four and a half bathrooms, a floor area of more than 5,600 square feet, and sits on a median area of one and quarter acres. Their settings are impressive as well: 12 percent of the CEO homes are situated on waterfronts; 8.5 percent are adjacent to or on golf courses. The median market value in 2006 was $2.7 million.

By comparison, the median sales price for all homes sold in the U.S. in 2004 was one-tenth as large, or $274,500, and the median home constructed in the U.S. in 2004 had 2,140 square feet of area with just two bathrooms. Liu observed that although a majority of CEOs still live in homes they owned before being promoted, the probability of a CEO trading up to a new home appears to increase with years in office.

For every doubling of a CEO's tenure in office, he or she moves up to a residence 15 percent more valuable. This may occur due to some combination of higher income, better job security, greater commitment to staying with the firm, or an increased sense of entrenchment — all of which should be correlated with tenure.

Bigger isn't always better

Due to mergers, IPOs, or bankruptcies, some firms were deleted from the S&P 500 index, so that an uninterrupted time series of 2005 stock returns is not available for all 488 observations. For those CEOs who left office or had their firms delisted during 2005, the study cumulates performance up to the times of their departures.

The 200 CEOs who live in homes with values above the sample median significantly underperformed their counterparts by an average of 3.4 percent in 2005. Those whose properties exceed the sample median and who acquired their grand residences after becoming CEOs underperformed their counterparts by a striking 8.7 percent.

Looking at just the largest estates — those with at least 10,000 square feet or ten acres of land, Liu and Yermack's study reveals even more excessive underperformance. On average, firms with CEOs living in the biggest homes lost 1.7 percent, while the remaining 85 percent of the firms gained 5.2 percent; that's a 6.9 percent difference in 2005 returns.

Power One, an energy infrastructure supplier, whose CEO Steven Goldman bought a 12,000 square foot beachfront property in Malibu, California in 2000, performed the worst, losing 30 percent of shareholder value in 2005. Berkshire Hathaway CEO Warren Buffett, a useful point of contrast though he's not an S&P 500 CEO, is noteworthy for having lived in the Omaha, Nebraska house he bought in 1958 for $31,500.

Investing with this knowledge

To measure the efficacy of an investment strategy that takes advantage of these findings, the team constructed a hypothetical trading rule in which an investor takes long or short positions in company stocks at the time of CEOs' home purchases depending on the size of the estate.

Since these home purchases occur at different times, the researchers subtracted the contemporaneous S&P 500 return from each observation's raw return to adjust for moves in the overall market. The cumulative difference between these two mean values equals 4.1 percent after three months, 15.0 percent after six months, 29.2 percent after one year, and 46.2 percent after three years! It's clear. "If [a CEO] buys a big mansion, sell the stock," Yermack says.

Financing

The other, possibly more illuminating finding in Liu and Yermack's study regards how CEOs finance their home purchases. CEOs on average finance about 27 percent of their home costs from selling shares and options, a percentage that is slightly higher for less expensive homes and slightly lower for new CEOs hired from outside the firm who likely don't own much internal equity at the time of home purchase.

Out of the sample, 160 CEOs were net equity sellers. In nearly every case, however, the amount raised from equity sales — a mean of about $450,000 and a median of zero — is rather modest relative to the CEO's overall stake in the firm. CEOs' median equity investment (shares plus Black-Scholes option value) exceeds $35 million, and even at the 10th percentile, CEOs own $6.7 million worth of company shares and options.

Because managers are more likely to sell shares when their firms' prospects are poor, these sales can be important signals about future company performance, despite the apparent personal liquidity rationale for the sales. Since management selling is sometimes discouraged or prohibited by boards and compliance divisions, as well as disfavored by outside investors, insiders can and apparently do rely on declared liquidity needs as a pretext for selling company shares.

Liu and Yermack found that when CEOs sell shares or exercise options to help pay the cost of a new house, the stock underperforms market benchmarks for at least the next several years. Conversely, when the CEO uses other methods of financing the home purchase and sells no equity, the company stock performs well thereafter.

Liu suspects, "these CEOs may see a home purchase as an opportunity to sell shares under a pretext that they are necessary to finance the property acquisition, while the true motive may be more closely connected to private information about the future path of his firm's stock."

Looking for other indicators

Conjecturing that CEOs who live far from the office may spend less time at work and have less direct contact with co-workers, Liu and Yermack examined CEOs commuting distances. Somewhat surprisingly, distance does not come close to being a statistically significant predictor, even in cases of CEOs who live thousands of miles from their offices. The study also looked at when CEOs acquire undeveloped land to build new houses and how being involved with new construction may affect job performance.

CEOs were not apparently preoccupied with blueprints and construction details; there was no significant underperformance. Using aerial photographs to identify CEO properties that are adjacent to golf courses and waterfronts, in the expectation that the leisure opportunities associated with these homes may lead to CEO shirking, the study again failed to find a statistically significant evidence of underperformance.

Conclusion

CEOs' decisions about the size, cost, and financing of their homes contain information useful for forecasting future performance of their companies. When CEOs purchase very expensive or particularly large estates, their companies' stock performance falters. Also, when CEOs sell stocks or options to finance their property purchases, their stocks deteriorate. These are signs of CEO entrenchment and waste.

"You know the old adage is true: Pigs get fat and hogs get slaughtered," said Crocker Liu. Entrenched managements are being beaten by leaner operations. "Boards aren't living up to the responsibility of setting reasonable limits on executive compensation. They've disconnected CEO and shareholder interests." Until shareholders get more active, these lackluster managers and their complacent boards will likely continue to underperform the market.

Bottom Line:

• Future company performance deteriorates when CEOs acquire extremely large or costly mansions and estates, regardless of the source of finance.
• Stock performance is inversely related to the CEO's liquidation of company shares and options for financing the transaction, even though these stock sales are often small relative to the CEO's total holdings in the firm.
• Large home acquisitions are evidence of Veblenian conspicuous consumption and waste, and signs of CEO entrenchment.
• Executive compensation packages need to be better aligned with shareholder interests.

Finance Real Estate Faculty