"No Free Lunch" - A Discussion of Leverage and REITs Real estate has long been an industry in which very high debt levels were common. Since there are no compelling reasons why the real estate industry should carry high debt levels, the current trend toward higher leverage ratios is of some concern. by Mike Kirby and Jon Fosheim

At first glance, higher debt levels seem to make considerable sense for REITs. As the top two graphs on page 47 show, by increasing a firm's debt, two "good" things happen. First, AFFO (adjusted funds from operations) per share increases. For the hypothetical REIT (assets generate a 9.5% yield and debt costs 7%) shown in the graphs, AFFO per share with no leverage is $1.00. At 40% leverage that increases to $1.18. If leverage is increased to a still prudent 60% level, AFFO per share increases to $1.39—39% higher than the unleveraged figure. For a management team who would have to develop a lot of successful properties and sign a lot of leases to generate a 39% increase in AFFO per share, this is enticing.

The second benefit of higher leverage is that it enhances AFFO growth. Simply put, whatever growth does occur is spread across a smaller share base. The turbocharging effects on growth are powerful as evidenced by the fact that the same hypothetical REIT, with a core, unleveraged, growth rate of 5%, can enhance that rate to 8.3% by merely taking on 40% leverage. At 60%, the growth rate is boosted to 12.5%.

The CFO's job seems easy, then. The 40% leveraged structure generates $1.26 per share of AFFO and offers 8.3% growth, while the 60% leveraged structure generates $1.39 per share of AFFO and offers 12.5% AFFO growth. Which financing structure would you choose? If you chose the latter of the two, you gave the same answer that virtually every real estate developer in the history of the world would have given. Unfortunately, it's also the wrong answer.

The correct answer is that neither structure is better than the other. As put forth by Nobel laureates Franco Modigliani and Merton Miller (see sidebar) in their classic 1958 and 1963 papers on "firm value," financing structure probably has little, if any, bearing on share value. The value of a REIT is determined by the assets comprising the left-hand side of the balance sheet coupled with whatever value is ascribed to management. How one chooses to divide the pie between debtholders and equityholders has no impact on overall value. Yes, if everything goes according to plan, shareholders of the more leveraged REIT will fare better.

However, one of the most basic tenets of economics/ finance is that investors are risk averse and need to be enticed by higher yields to accept more volatile cash flow streams. Because leverage can serve to magnify negative surprises just as it can boost positive projections, investors will neither reward nor punish a REIT for adding leverage. Because the higher AFFO per share and AFFO per share growth are directly associated with an increase in volatility, the REIT's multiple will adjust (see bottom graph on page 47), and its share price will be unaffected.

Debunking Some Myths
The academic view of leverage is therefore quite clear. Furthermore, the ramifications that this line of thinking has on REITs are not difficult to understand. In a nutshell, the leverage ratio for a REIT should not matter too much, as long as a REIT is operating within the band of prudence—say, 0% to 65% leverage. When a REIT operates within this band, shareholders should largely be indifferent, and the leverage ratio will have no impact on share pricing. Seems straightforward? Unfortunately, that doesn't turn out to be the case. A number of market participants, both at public companies and among the cadre of securities analysts following REITs, speak and act as if either they have never heard of this or don't believe a word of it. If something became true by repeating it often enough, the following list of leverage-related myths would be gospel.

Equity holders are concerned with return on equity, so the goal should be to maximize the return on equity. This line of thinking ignores the higher volatility of cash flows that is associated with leverage and violates the basic principal of finance that investors need to be compensated for increased volatility. That compensation comes via a lower multiple.

Investors are not really sensitive to leverage, as long as it doesn't get out of hand. After all, who really notices if leverage is 50% instead of 25%? This comment assumes that the only negative ramification of higher leverage is bankruptcy risk. It again ignores that every dollar of added leverage adds incrementally to the volatility of the company's cash flow. Investors will demand to be compensated for this volatility with a higher expected yield.

Leverage serves as a way to enhance returns. Ask the person making that statement if he/she recalls how many solvent real estate developers existed in 1992. Leverage does enhance returns, if things go as expected. Things don't always go as expected, however.

If a REIT utilizes secured rather than unsecured debt, its shareholders are exposed to less risk. In one relatively unimportant sense this is true. The risk of bankruptcy is lower for a REIT that utilizes secured debt vs. a REIT that utilizes a similar amount of unsecured debt. Put another way, the REIT that relies on secured debt may be able to employ a higher leverage ratio before its shareholders wince. However, bankruptcy risk is likely viewed as immaterial at the leverage ratios at which most equity REITs operate today. The more important ramification of higher leverage is the added volatility it brings to a company's cash flow stream. From this standpoint, there is no difference between secured debt and unsecured debt, and the multiple adjustment on the shares will be identical.

Preferred stock should be viewed as equity because that's how rating agencies treat it. Of course, rating agencies treat preferred stock as equity; their constituency is debtholders. Preferreds have a subordinate claim relative to bondholders, so preferreds do nothing to impair their claim on a company's assets. However, preferreds have a superior claim to common shareholders on after debt service cash flows. From the common shareholders' perspective, preferred stock is viewed no differently than debt. In other words, preferreds render the cash flow stream available to common shareholders as more volatile, and the market will ascribe a lower multiple to this riskier stream.

If one is attempting to assess bankruptcy risk, preferreds are more equity-like than debt-like, but at the debt levels at which most REITs operate today, bankruptcy risk is not the risk that is uppermost in investors' minds. Since we believe the primary goal of measuring leverage is to measure its potential impact on earnings volatility, we treat preferreds as debt.

Since debt costs less than equity, a firm can decrease its weighted average cost of capital by borrowing more. A company's weighted average cost of capital is a function of its business risk. It is not affected by how a firm chooses to finance itself. If more debt is added, the cost of equity will increase as equityholders demand to be compensated for the increased volatility of their cash flow stream. If the interest tax shield (see sidebar) has value, some reduction in the weighted average cost of capital may be obtained by leveraging, but this is likely not much of a factor for tax-exempt entities such as REITs.

Measuring Leverage
The majority of Wall Street REIT analysts utilize some combination of FFO/AFFO yield and FFO/AFFO growth prospects to reach conclusions about a REIT's appropriate valuation. If one uses that valuation approach, it is essential that the analyst explicitly factor the impact leverage has on a company's AFFO yield or AFFO growth into his/her earnings estimates (see table below). Not doing so ignores, as noted, that though leverage does not impact share price, it has a profound impact on multiples.

In addition to not placing enough weight on how leverage should impact multiples, most analysts also do a poor job of measuring leverage. There are three primary methods in use to measure leverage. Among the most often-cited measures is debt as a percentage of the company's total market capitalization (that is, equity market capitalization plus total debt). Though easily calculated, this measure is largely a function of a REIT's share price.

Another frequently cited measure of leverage is a company's coverage ratio (EBITDA divided by interest expense). In our view, that measure also is flawed because it fails to differentiate between the debt capacity of a portfolio of junky real estate vs. that of a high-quality portfolio.

The third and, in our view, the best measure of leverage is debt as a percentage of the company's current value capitalization. We calculate this ratio by marking-to-market the value of all assets, as well as adjusting the value of any debt that is well below or above market (e.g., we reduce the liabilities of apartment REITs that utilize the subsidies available for tax-exempt debt). This measure also avoids the criticism attached to the use of coverage ratios by ascribing a lower valuation multiple (higher cap rate) to low-quality real estate.

The value of this measure is further enhanced by including, as is our routine, a pro rata share of all joint venture (JV) assets/debt in our current value balance sheets. GAAP accounting rules generally allow JV debt to be removed from the balance sheet (just the net equity in the JV shows up as an asset), and many analysts make the mistake of not looking beyond the figure that appears on the balance sheet. This results in an improvement in a company's measured leverage ratio when a REIT enters into a highly leveraged JV! In addition, we treat most preferred stock as debt because it has a superior claim over common equity and serves to lever the cash flow stream available to shareholders. In many instances, these two adjustments result in a significantly higher (and much more accurate) leverage ratio.

The lure of leverage is twofold. The first two graphs below show how added leverage can boost both the AFFO per share and the AFFO per share growth rate for a typical* REIT. But there is no free lunch. The increased AFFO per share will be entirely offset by a decrease in the multiple. The share price will be unchanged. * This hypothetical REIT is presumed to generate a return on assets of 9.5% and incur borrowing costs of 7%. Source: Green Street Advisors

Quantifying the Impact of Leverage on AFFO Yield and Growth
As noted, any comparative analysis of REIT AFFO yields or AFFO growth has to include the effect that leverage can have on those figures. Because REITs can so easily manipulate AFFO, AFFO yields, and AFFO growth by adding leverage, investors need to be cognizant of leverage ratios.

The table below attempts to back out the relative impact of above-average or below-average leverage ratios. It is important to note that several simplifying assumptions are made that make the analysis hypothetical in nature. Therefore, the results shown in the table are more indicative of how a typical REIT with a given capital structure may be impacted by leverage, as opposed to a detailed review of each REIT's debt structure. The two most important of these simplifying assumptions are that all REITs earn the same return on the marked-to-market value of their assets and all REITs (with the exception of some apartment REITs that utilize extensive tax-exempt financing) have the same average interest costs. While it is true that these assumptions result in slightly misleading results for a small number of companies, this should not detract from the general conclusions that companies with high leverage are reporting "inflated" AFFO and AFFO growth, while the two figures are understated for companies with low leverage.

In computing the inflating effect of leverage on AFFO yield/AFFO growth, the assumption is made that a company with above-average leverage issues shares today at the prevailing share price and uses the proceeds to retire debt. By contrast, a company with below-average leverage is assumed to borrow funds, using the proceeds to buy back shares. Transaction costs are ignored, and our leverage measure is total liabilities as a percent of our estimate of asset value. The results shown in the table on page 46 answer two hypothetical questions. First, how would a given REIT's current year AFFO and AFFO yield be restated if it had an average amount of leverage? Second, how would next year's projected AFFO growth for that same REIT be impacted if the REIT maintained an average amount of leverage?

It is probably most helpful to look at the effect that leverage has on the combination of yield and growth, as opposed to focusing on one or the other. Some leveraged REITs with richly priced shares would experience very little, if any, impact on their first-year yield if they delevered, but every leveraged REIT would experience a slower growth rate if it did so. The results of this exercise should be useful to any investor who focuses primarily on AFFO yields and AFFO growth in valuing the shares of a REIT.

Download a copy of Impact of Leverage on AFFO Yields and AFFO Growth chart

The Propositions That Changed Modern Finance M.I.T.'s Franco Modigliani and the University of Chicago's Merton Miller, both Nobel laureates, are credited with coming up with the propositions that changed finance theory. In two papers, published in 1958 and 1963, Modigliani and Miller put forth the proposition that in their "perfect world" where there were no taxes or transaction fees, the level of debt at a company should have no impact on the value of the company (i.e., the price of its stock). Put differently, the value of a firm is tied to the value of its assets (both tangible and intangible), and how one chooses to divide this value between debtholders and equityholders has no bearing on overall value. As debt increases, risk-averse investors demand a higher return on the equity, and any improvement in net income per share generated by taking on more debt is fully offset by a decline in the multiple the market will award the shares. The "perfect world" assumptions of what are referred to as the M&M propositions are artificial. In other words, Modigliani and Miller created an economic model which could then be manipulated at will by economists studying corporate finance theory. More important, the ways the real world deviates from these assumptions have been dealt with in subsequent work by M&M and others. Here are the most meaningful issues that need to be addressed before applying M&M's basic theory to the real world: 1. The tax law in this country allows companies to deduct interest expense but not dividend payments. This creates an "interest tax shield" that has very real value to any tax-paying company that has debt. Taken on its own, this issue would suggest that all tax-paying companies should be very highly leveraged to minimize the entities' tax bills. 2. Another argument sometimes cited in favor of higher corporate leverage is that large corporations can generally borrow in a more efficient manner than individuals. A basic precept underlying M&M's perfect world is that all entities—individuals and corporations—share the same borrowing costs, and individuals can achieve their own desired degree of leverage/risk by borrowing in lieu of having the company do it for them. In the real world, transaction costs can be lower for corporate borrowers, and they can often achieve a better interest rate than the companies' shareholders. The difference here is often not as sharp as it might first appear, however. Individuals oftentimes can achieve very competitive interest rates through the use of home mortgage or margin debt (both of which offer tax-deductible interest). 3. The argument against high debt levels has to do with the risk of bankruptcy and the high costs and divergence of interests that begin to appear as the risk of bankruptcy becomes meaningful. The potential transaction costs of bankruptcy are sizable and obvious. In addition to the bills from lawyers and other professionals, damage can be done to a firm's basic business and reputation. Less obvious, but equally important, are the agency problems, otherwise known as potential conflicts of interest that begin to surface at high debt levels. Also at high debt levels, the risk of default/bankruptcy increases and the potential for abuse of bondholders by shareholders/managers increases along with it. Bondholders are not naïve and will demand a risk premium higher than what would be warranted in the M&M perfect world. An extension of this rationale is that some firms that own tangible, low-volatility assets (e.g., REITs) can afford to be more leveraged than firms with substantial intangible, high-volatility assets. Modigliani's and Miller's views have been extensively scrutinized. Numerous studies have been done to see if the theory can be confirmed. While the results thus far have been less than crystal clear, M&M's preachings appear to be standing up well under the scrutiny of others. In an exhaustive study published in the June 1998 issue of The Journal of Finance, the University of Chicago's Gene Fama and Yale University's Ken French concluded that higher leverage does not result in higher firm value (read: share price). In fact, if anything, higher leverage was associated with lower value.