REITs: The Uncertain Path To AFFO

Concept image of Business Acronym REIT as Real Estate Investment Trust. 3d rendering

Kwarkot/iStock via Getty Images

Those serious about understanding investments in REITs need to grapple with the problem of assessing their earnings. This leads to Adjusted Funds From Operations, or AFFO.

Unfortunately AFFO is quite treacherous. There is no standard for it. Like any other type of firm, REITs can produce adjusted numbers of any sort they wish and hope that investors will believe they represent actual earnings.

My view is that it does not make sense to leave the choice of adjustments entirely in the hands of the REIT. I often use the adjustments standardized by REIT/base, about which more below.

One important goal of this article is to provide a thorough explanation of why I do that. This will reduce the need to clog up other articles with such explanations. More generally, this article discusses the basic problem of assessing earnings for REITs.

It would seem to me that getting this right should matter, although for some investors this may be a lost cause. Quite a few of the comments on various articles seem more tribal than discursive. Woe to the author who suggests that a popular REIT is not perfect in all respects, as I did in my most recent article on W.P. Carey (WPC).

First let’s start at the beginning.

The Ideal Calculation

As Chris Volk, a former CEO of multiple REITs, discusses in his forthcoming book, the calculations one needs to understand a business and its profitability are not the calculations of the accountants. So we start with what would make business sense for a REIT.

Here is what one wants to know:

Simple AFFO defined

RP Drake

No sweat, eh? In pulling the numbers together one finds them in a few places. One can seek to find these components:

  • Cash revenue
  • Cash expenses of operating the properties
  • Cash costs of overhead (“G&A”)
  • Interest expenses
  • Other cash expenses, including recurring capital expenditures (“capex”) necessary to sustain the value and viability of the properties.
  • Dividends on preferred stock
  • Stocks and options issued to individuals as compensation.

We will come back to stock-based compensation later. In the meantime bear with me.

By subtracting all the other items from the first, one would arrive at earnings that could be distributed to investors in the common stock or reinvested in the business. This is the first thing one really wants to know. It is also one of many things sometimes meant by “free cash flow”.

Here is how you can come close to this ideal number:

  • Start with GAAP revenues.
  • Remove the adjustments for straight-line rent and market rents.
  • Subtract any other non-cash revenue.
  • Subtract property operating expenses.
  • Subtract GAAP overhead and interest costs.
  • Subtract recurring capex (one may have to dig for this, but it matters).
  • Subtract preferred stock dividends.

This gets you what I call Simple AFFO (“SAFFO”).

One source of “other non-cash revenue” may be from marking to market of certain investments. This is now required by GAAP to be reported as current revenue. Only a few REITs have much of this, but some have a lot.

The calculation just described probably gets one a pretty good value for the net cash that can either be paid out as dividends or reinvested to drive growth. There are ways to check, discussed below. If the dividends are less than 80% of this then they are well-covered (although no dividend is absolutely safe).

If not, the dividends are at least somewhat at risk against adverse developments. Also, the discussion here ignores issues related to liquidity that do matter when things go badly.

Stock-Based Compensation

The GAAP G&A costs include the cost of any stock-based compensation. REITs typically add these back if they provide an AFFO.

Notably some REITs do not report an AFFO or do not add back such compensation. I discuss two of these below.

The argument for adding it back in is that stock-based compensation does not incur current costs. This is true, but also in my view misses the point.

If a firm has some relatively steady level of stock-based compensation, then over time the resulting share issuance each year will be some typical fraction of existing shares. It does not matter whether what is issued is stock, restricted stock, or even options (though the impact of these may be lumpier).

This share issuance dilutes existing shareholders. The implication is that less dividends per share can now be afforded.

One can seek to do complex calculations about the impact of such compensation. But the easy thing to do, and the one that gives a conservative view of the finances of the REIT, is to leave those costs in.

I often rely on the REIT analysis firm REIT/base for estimates of AFFO. For those REITs they cover, I also benefit from their identification of all the adjustments illustrated below.

REIT/base does not add the stock-based compensation back in, which often leads their AFFO to be a few percent smaller than that claimed by the REIT. In this respect, my view is that the REIT/base number provides a more accurate indication of the dividend coverage.

As a result, in many cases I use the REIT/base number.

The Path to FFO

Unfortunately the earnings numbers one finds for REITs have only a limited relation to the ideal calculation described above. Sometimes it is closer than others.

GAAP accounting rules provide the basis for financial reporting. For REITs, one additional non-GAAP calculation is standardized. It is the NAREIT Funds From Operation, or FFO, and it is based entirely on GAAP numbers.

The main thing the FFO calculation does is to add back in Depreciation & Amortization charges. There are a small number of additional adjustments, one of which is to add back in any impairment charges.

The adjustments producing FFO make sense for both of these two goals:

  • Goal 1 — One may seek to understand the real estate operations of the REIT, as such.
  • Goal 2 — One may seek to understand its ability to pay dividends.

Unfortunately, the adjustments producing FFO are often insufficient to accomplish either goal.

A simple test of whether there are costs that may be impacting Goal 2 is to compare the NAREIT FFO to the Cash From Operations on the Statements of Cash Flows. If the latter is significantly larger, then there are real cash costs that have not been included in Cash from Operations. Instead they are included within financing or investing. This plays out in the next part of the path.

The Path From FFO to Core FFO

The path from FFO to AFFO, as it is practiced by many REITs, involves two parts. Here is a graphic illustrating these:

FFO to AFFO

RP Drake

The first step is to add back in various costs, usually on the grounds that they are non-recurring. They often arise from financing activity. This gets one to a number often reported as “Core FFO.”

The natural incentive is for a REIT to add back in many costs, in order to make themselves look as good as possible. In my view, such manipulations should be viewed with skepticism. Some of them are pretty egregious, as I discussed here.

Typical examples of such costs include “Debt Adjustments” and “Transaction Costs.” Here the two Goals from above diverge. On the one hand, one may indeed want to take these costs out to see the smooth, continuing ability of the real estate operation to generate earnings.

On the other hand, these are real cash costs. They reduce the funds available to pay dividends or reinvest. To serve Goal 2 above, one wants to track that. This is where my own interest as an investor lies.

Some REITs show such costs persistently and at levels that are not negligible. One can make an argument for smoothing them over time, but just dumping them back into earnings gives a misleading view of the coverage of dividends by the REIT.

Bear in mind that a typical REIT pays out 80% of AFFO as dividends. If, for example, the cash actually represented by that AFFO has been burned up in the costs added back to get to Core FFO, then the REIT has no spare funds for growth. So ratios of Core FFO to NAREIT FFO that are above 110% start to be a significant concern for the ability of a REIT to pay its dividend and grow.

Across the 95 REITs covered by REIT/base, the median ratio of CORE FFO to FFO is 1.02. So a typical REIT has 2% less cash available to pay dividends than is seemingly implied by the REIT/base AFFO. This is not enough to worry about.

But there are outliers. Here, for 2020 and 2021, are the top 25 of the REIT/base REITs ordered by the ratio of CORE to NAREIT FFO. Repeat offenders are highlighted in yellow.

Ratio of Core FFO to NAREIT FFO

Author plot based on REIT/base data. (RP Drake)

A number of the high ratios in 2020 were for REITs who had many tenants shut down by governments, and who were paying reduced or no dividends in response. Examples include EPR Properties (EPR), Brixmor Property (BRX), and KIMCO (KIM). Regency Centers (REG) shows up too, but they used their financial might to sustain their (uncovered) dividend. I specifically discussed EPR in this recent article.

Large ratios here also often do relate to one-time events. As an example, Independence Realty Trust (IRT) went through a major and important merger during 2021, which I discussed here.

While it does make some sense in such a case to not count the associated, one-time, transaction costs in evaluating AFFO, it is also true that such cash is not available to cover the dividend. But perhaps the synergies or improved earnings from the merger make up for this. The question merits examination in any such case.

I have not followed Apartment Income REIT (AIRC), for which the ratio of Core FFO to NAREIT FFO is high in both years. Notably, both years show large debt adjustments and large transaction costs.

REIT/base shows the AFFO payout ratio for AIRC at or above 100% in both these years. And the table above shows that AFFO is significantly larger than the cash actually available from earnings. I would want to dig deep before investing in that one.

I have also not followed Corporate Office Properties Trust (OFC). The story is similar. For both these years, the combination of the REIT/base payout ratio and the data in the table above implies that AFFO is significantly larger than the cash actually available to cover the dividend.

Now, it is true that a REIT can cover the difference between earnings from operations and the dividend with financing and investing activity. But doing so is not without adverse longer-term consequences.

My own analysis does not always keep the adjustments included in CORE FFO. When these are large, I may use a value of AFFO smaller than that of REIT/base, because my interest is mainly in Goal 2 above.

From Core FFO To AFFO

To see common adjustments between Core FFO and AFFO, we will take a look at Alexandria Real Estate Equities (ARE). Unusually, for ARE Core FFO is below FFO. The reasons are interesting but are not important here.

The following numbers are extracted from the REIT/base table related to annual FFO:

CORE FFO to AFFO

Extracted from REIT/base output. (RP Drake)

Here one sees the usual adjustments also found in company filings. The S/L Rent item corrects GAAP revenues to remove the non-cash adjustments to cash rent payments that distribute the revenue on a straight-line basis. This number is larger for any REIT with long lease terms and fixed escalators.

This adjustment also is typically negative because the REIT is typically growing its assets over time. Some authors miss this aspect.

The Market Rent adjustment similarly removes some non-cash revenues mandated by GAAP. And the Straight-Line Expense is nearly always small.

The Recurring Capex row subtracts from earnings the funds used to maintain the value of the properties. This is a real cash expense but unfortunately is not an explicit part of the GAAP Statements. For the lab properties owned by ARE, it is a substantial item.

As a ballpark, the REIT/base AFFO is around 70% of Core FFO for ARE. This is common for healthcare REITs, office REITs, and others with substantial recurring capex.

As you can see, REIT/base finds the payout ratio to range from 76% to 86% for these years. In most cases, including this one, the payout ratio quoted by the company will be smaller (i.e., better).

The reconciliation of these is provided via the hyperlink indicated by two blue lines. That reveals this:

REIT/base vs company AFFO

Extracted from REIT/base output. (RP Drake)

In this case the stock-based compensation is substantial. Overall, the Company AFFO is in the ballpark of 10% larger than the REIT/base AFFO. A 5% difference is more common.

As you saw above, what seems important to me is that the dilution associated with continuing stock-based compensation reduces the cash available to cover the dividend. For this reason, I prefer the REIT/base payout ratio as a measure of dividend coverage and retained earnings. But even that can be too optimistic, depending as seen above on the story with Core FFO.

Simple AFFO In Context

Returning to simpler calculations, here is an example from Federal Realty Investment Trust (FRT), in my view one of the class acts in the REIT space. They report only an FFO, and the numbers one needs to produce an AFFO for oneself.

Simple AFFO for FRT

RP Drake

Here the numbers in black are from the Income Statement (save for the dividend). The numbers in purple on rows shaded orange can be drawn from the SEC filings or taken from the compilation by REIT/base. Care is needed regarding the share count since any Limited Partner units must be identified separately.

The numbers in red are the result of calculations. One can see that the SAFFO is typically about $10M larger than the REIT/base AFFO. This mainly reflects an accumulation of small items such as expenditures attributed to non-controlling interests.

The payout ratio was above one for 2020 and 2021. Like REG, FRT chose to use their financial strength to keep the dividend going across the pandemic.

Another class act in my view is Camden Property Trust (CPT). They refuse to report a Core FFO (per earnings calls) and report an AFFO in their filings in which only Recurring Capex is subtracted from FFO. (As an apartment REIT with only short-term leases, they avoid some of the GAAP complications discussed above).

In my view the bulk of the REIT space needs to clean up its act. REITs should reduce the reporting of misleading numbers and increase the disclosure of what is misleading about the GAAP numbers.

Takeaways

We have seen that there is no value of REIT earnings that is right or wrong. There are only choices about how to view various costs and non-cash adjustments. Anyone who accuses some author or investor of being in error is suffering from very shallow understanding.

I have some hopes for what you might apply to your own investing from the above.

First, you have seen why I prefer the REIT/base AFFO for the REITs they cover, as a number to go with most of the time. The subscription is inexpensive and the presentation is clear. There is no reason for any REIT investor not to take advantage of this tool.

Second, you have seen that some attention to the relation of NAREIT FFO to Cash from Operations is an easy way to spot when deeper scrutiny is needed. Comparing this with a Simple AFFO evaluated as shown above can deepen the conclusion, again without a lot of added effort.

Finally, one can easily be fooled by NAREIT FFO. One can further be fooled by AFFO from the company (or worse, from some automated computer calculation). The ultimate point here is to not get fooled.

Be the first to comment

Leave a Reply

Your email address will not be published.


*