Director Compensation and CEO Bargaining Power in REITs

For example, Hermalin and Weibach (1998) argue that loyal outsiders can always be found through interlocking.

Conner and Halle (2006) echo the same concern, “Transparency and governance have improved, but they are not perfect.” (page 17).

Ryan and Wiggins (2004) note that, “Over time, the CEO nominates new directors who are indebted for their appointments, which erodes board independence.” (pp 500) Obviously, the CEO has greater opportunity to select his preferred directors if he is on the nomination committee.

However, the requirement that REITs must payout 90% of taxable earnings to qualify for tax-exemption can mitigate agency costs. Besides return of cash flows to investors, high mandatory payout forces raising funds from investors, exposing REITs to the scrutiny and disciplining forces of the capital market.

Ryan and Wiggins (2004) use compensation data from Standard and Poor’s ExecuComp database which includes firms from S&P 500, the Midcap 400 and the Smallcap 600 Indices. There are systematic and significant differences between Ryan and Wiggins (2004) and our sample. For example, the average size (total assets) of firms in Ryan and Wiggins (2004) is $4b while that in our sample is $1.8b. The average number of directors in their sample firms is 9.2 with 48% outsiders; the corresponding numbers in our sample is 8 and 52%.

It may be argued that the requirement that REITs must pay out 90% of net “taxable” earnings mitigates agency costs by limiting management’s access to free cash flow. Further, generous payout forces REITs to raise funds from the market, which, as Easterbrook (1984) observes, subjects REITs to the disciplining forces of the market. Ghosh and Sirmans (2006) report that dividends and board monitoring act as substitutes. However, several authors note that REIT managers have access to significant free cash flow due to their large depreciation expenses (Bradley et al. 1998; Kallberg et al. 2003; Ghosh and Sirmans 2006).

Anecdotal evidence and analyses of board composition lead us to believe that board structure changes slowly. Indeed, it is unusual for a director to be appointed to a board and not be retained for a number of years. We feel that 1-year is too short a period for new directors and their compensation mix to have discernible impact on performance. On the other hand, 5-year is possibly too long. We chose 3-year as the period under the premise that any new director can be expected to serve for at least 3 years, on average. In our data, we do not examine when an individual director is appointed, rather we assume that an existing board will serve for a period of 3 years, on average. Recently, Hardin et al. (2005) study 3-year stock returns following stock splits. They argue that it is tenuous to attribute 3-year return to a particular event. We take the opposite view because changes in board structure and compensation mix of directors are infrequent so long as the firm is performing uniformly.

See Ryan and Wiggins (2004) for detailed discussion. The average total compensation in 1997 for 1,018 industrial firms is $110, 995, with 50% in stocks and options.

We collect data on institutional ownership of individual REITs from the SNL and CDA Spectrum databases. Because of unavailability of detailed data on individual REITs, we do not distinguish among the type of institution although several authors have observed that monitoring effectiveness varies depending on the type of institution. Our analysis uses the aggregate measure of institutional ownership. Evidence on institutional ownership of REITs is limited. The impact of the type of institution on performance of REITs is an area of potential future research.

An alternative measure of performance in the REIT context that has been used by several authors is Net Asset Value (NAV). The general consensus is that NAV measures developed by GreenStreet Advisors are the most useful and reliable. The NAV measures reported by SNL Financial are questionable because it uses the same capitalization rate for all properties. Unfortunately, GreenStreet publishes NAV measures for a select group of REITs. For our sample, we could obtain the NAV for only a few REITs in our sample. Results based on this sample are not dependable because of serious sample size issues. Therefore, we do not report these results.

Feng et al. (2007) find that, in 1999, 31 out of a sample of 118 REITs had nomination committees and 16 of these did not involve their CEOs. In 2000, 42 out of the same 118 REITs had nomination committees, 24 of which had no CEO involvement.