Contributor Since 2012
For those who have not read my comments on the matter, I view Realty Income Trust (O) as a bond substitute. In other words, why buy a bond that yields 5.5% today, if I can buy a stream of future cash flows (dividends) that grows over time through the power of reinvestment and dividend growth.
Because I view the investment as a bond substitute, I use a discounted cash flow model to value the future dividend stream.
If I set my discount rate at 9% (my expected annual rate of return on investment), discounted cash flow analysis tells me that based on a current yield of ~5.5% and a current price around $40/share (today's price is $38.78 which is within 5%), the future dividends would need to grow at about 6% per year in order to justify a price of $40 (+/-5%) today that returns ~9% annually to me over a 20+ year period.
With a 6% dividend growth rate target in mind, I can start to create a model for O's operations to see if this is achievable.
Starting from the revenue side, O earns rents from triple net tenants and those leases provide for about 1.3% annual rent bumps. When those leases mature, a good number of the tenants will re-lease and a conservative assumption is O would get about a 10% bump in the rent in that lease year. For those tenants that don't release, O has the opportunity to lease the space to a new tenant at the market lease rate (or sell the property and reinvest the proceeds in order to generate a market lease yield on the investment). If interest rates have gone up considerably, the market lease rate should also go up considerably. For modeling purposes, I'm assuming 50% re-leases/50% new leases. If we take a look at page 55 of O's 2013 Q3 10Q report, we can see that O provides a lease expirations schedule. I've circled the "Total Portfolio" to call your attention to this data.
Accordingly, I can use the data in their Q3 lease expirations schedule to model O's expected future cash flows from rent. First annualize the Q3 rent; second, adjust each expiring rental stream by 1.3% annual bumps (both the pre- and post- expiration rent streams); and third, in the expiration year apply a +10% rent bump on 1/2 of the rent for releases, and a +X% bump on the other 1/2 for new leases (the X% will vary with our interest rate assumptions).
On the cost side, O has two main sources of cash cost: 1) Corporate/Property expense; and 2) Interest. Capex is relatively immaterial at this time due to the leases being triple net retail (not industrial/data center).
For Corporate/Property expense, O essentially incurs about $93M annually for these. For modeling purposes, these costs can be considered as fixed.
For Interest expense, O has two main pools of debt: Unsecured notes and Mortgages. For Unsecured debt - From page 15 of the 2013 Q3 10Q, O reports the following (weighted average interest rate is 4.9%):
Note that O's interest expense is essentially fixed, so if interest rates rise tomorrow, it does not change the amount of interest expense O needs to pay tomorrow on its existing debt. Higher rates only have an impact on re-financing rates and new debt.
When I take this data and schedule out their annual interest expense by maturity tranche, I can add various assumptions about future interest rates to see the affect on total interest expense. For example, I may expect interest rates to remain low for a very long time, I may expect interest rates to rise ~50% from their current level, or, I may expect interest rates to spike upward, say double their current level or more. I'll come back to this in a moment.
If I then take the scheduled revenues based on lease expirations and match them up with scheduled interest expense based on debt maturities, and subtract fixed overhead charges, the balance is essentially FFO available to pay both preferred and common. Backing off preferred distributions (estimated around $50M annually - see my first instablog post where they have ~124M sitting in their credit facility after Q3 which I believe will ultimately be financed with preferred), the balance is the FFO available for distribution to common shares. It is this amount that needs to grow by ~6%/year per share in order to achieve my 6% dividend growth expectation.
What I've found is that in the low interest rate scenario (i.e. the current ~5% portfolio interest), the organic growth of O's lease revenue (i.e. 1.3% automatic bumps, re-leasing bumps and new leasing bumps) can safely cover growth in their dividend by around 2.5% per year from the current level. In the moderate interest growth rate scenario (i.e. ~7.5% portfolio interest), the organic growth only supports a dividend growth rate of around 1%. However, in the high growth scenario (where the average portfolio rate is 10%+), the dividend would likely need to be cut as the organic growth in the rent pool would not cover the higher interest rates upon a refinancing.
However, because each of the interest scenarios does not achieve my 6% dividend growth threshold, there needs to be another source of FFO growth. O essentially fills this gap by borrowing or issuing equity at a cost of capital that is ~1.5% below the return from the asset they are purchasing. In other words, if interest rates spike up to 10%, O could issue debt with a 10% coupon and use the capital to acquire a portfolio of properties that had a lease rate return of at least 11.5%.
Interest rates are not a one-way street only raising the cost of borrowing, while having no impact on the assets. O, or any other company for that matter, would not borrow at 10% to buy something that returns 7%. On a macro-level, asset prices would have to come down, or the lease rates paid by the tenants would have to come up, for the economics to work.
So if O earns a 1.5% spread on new capital, how much would they need to grow their balance sheet by each year in order to achieve a 6% dividend growth rate? Based on my estimates, in the low interest rate scenario, I believe they'd need to issue new debt/equity capital and acquire about $1.0Bn per year in new assets. In the moderate interest rate scenario (+50% from today's rates), they'd need to issue/acquire about $1.5Bn. In the high interest rate scenario, they'd need to issue/acquire over $2.1Bn (increasing as rates exceed 10%+).
Realistically, the low (5%) and moderate (7.5%) interest rate scenarios are the most likely outcomes that my crystal ball seems to tell me. It is possible that we see corporate BBB+ borrowing rates at 10%+, but that would seem to suggest hyperinflation - certainly possible, but I don't believe is probable given the coordination between Western central banks + Japan. That said, because these risks do exist, I don't believe its wise to be overweight REITs in my portfolio at this time.
As to whether or not there are a sufficient amount of triple-net assets available to acquire, I look to Sumit Roy's (EVP, Chief Investment Officer) comments from O's Q3 earnings call transcript seekingalpha.com/article/1796002-realty-incomes-ceo-discusses-q3-2013-results-earnings-call-transcript. Essentially, they looked at $17Bn in deals in the third quarter alone and only closed on $503M in transactions because the other opportunities did not meet their portfolio objectives.
In other words, the acquisition opportunities are clearly there for O to meet my 6% dividend growth target. Accordingly, I believe O continues to be fairly valued today around $40 (i.e. +/-5%).
As always, I take no responsibility for your investment decisions and you must perform your own due diligence. This data and any analysis are for informational purposes only and are intended to promote financial literacy.
Best of luck to everyone on their investments.
Disclosure: I am long O.
Additional disclosure: I have no other affiliation with the company and I wrote this post myself.