Spread between Loan Constant and Cap Rate

IRRelevant:

I'm sorry but your comment is wrong. What the OP is asking is if people look to see if there is positive leverage or negative leverage in a deal. The answer is absolutely.

In the OP's original example the loan was amortizing and in your example is the loan is IO. The "amortization is negligible" comment is incorrect, it makes a huge difference on ROE. The OP had negative arb on the debt YOC vs. cost of debt (4.75% vs 5.77%). In your example there is positive arb (YOC vs. cost of debt) because the loan is IO. Positive arb = higher ROE, negative arb = lower ROE.

If you are in a deal and you have negative arb you are basically betting that the property appreciation (through either income growth or cap rate compression) will make up for you having negative leverage throughout the hold period. Developers for example make this bet all the time (they have construction debt and zero income from the property during construction).

In the OP's original example unless there is a value-add component to increase income, the property is not stabilized etc or the OP is betting on cap rate compression (scary) then the debt is too expensive and it's probably not a good deal.

Positive arb?

Yeah, the only example I could think of for OPs question was on an unstabilized property (whether it be development or otherwise). If people are buying stabilized properties that can't cover debt service at today's rates betting on rent increases and cap rate decreases to save them, please let us know so we can run for the hills.

IRRelevant:

I'm sorry but your comment is wrong. What the OP is asking is if people look to see if there is positive leverage or negative leverage in a deal. The answer is absolutely.

In the OP's original example the loan was amortizing and in your example is the loan is IO. The "amortization is negligible" comment is incorrect, it makes a huge difference on ROE. The OP had negative arb on the debt YOC vs. cost of debt (4.75% vs 5.77%). In your example there is positive arb (YOC vs. cost of debt) because the loan is IO. Positive arb = higher ROE, negative arb = lower ROE.

If you are in a deal and you have negative arb you are basically betting that the property appreciation (through either income growth or cap rate compression) will make up for you having negative leverage throughout the hold period. Developers for example make this bet all the time (they have construction debt and zero income from the property during construction).

In the OP's original example unless there is a value-add component to increase income, the property is not stabilized etc or the OP is betting on cap rate compression (scary) then the debt is too expensive and it's probably not a good deal.

You are thinking about the cost of debt incorrectly. Given this scenario (and assuming amortization), we are in agreement that the debt constant is 5.77% THROUGH THE BASIS OF THE LOAN. (i.e $403,724 / $7,000,000 = 5.77%). What this fails to account for is the overall capital stack. If you take the debt constant as a % of purchase price (i.e total cost aka our capital stack), the debt constant falls to 4.03724% ($403,724 / $10,000,000).

Now think about it this way. You are buying the property (i.e. the total unlevered cost) for $10M at a 4.75% cap rate (or rate of return). You then turn around and "sell" the 0-70% portion of your capital stack for a return of 4.03724% (LESS than your overall rate of return of 4.75%). What you are effectively doing is splitting the unlevered return into 2 buckets - the 0-70% LTV portion for 4.037%, and then the equity (70-100% LTV) portion. iIf you do this out in excel, the return on equity (IRR) is 6.3%, assuming 360 amort and a 5 year hold.

Now, to the point about why amortization is negligible. The reason the above IRR for the equity is 6.3% and not 6.5% is due to amortization. You are going to have to pay back the loan eventually and in the same $ amount, whether you do it at the maturity of the loan, or through the loan term via amortization. Because this money is coming out of your pocket either way, it is now "negligible". It does, however, have an effect on overall interest payments (as your balance is slowing decreasing in the case of amortization.

For the simple math of this exercise, amortization is unimportant because that money will be repaid in the exact same $ amount, no matter how you skin the cat.

I will concede the time value of money aspect of an IO vs. amortizing loan is Accretive to returns, but by a small amount, unless you are looking at a long, stable hold period. The bigger reason investors like IO loans are that the annual payments are much lower. Why is this good? A lower annual debt obligation means the property can support a higher LTV. A higher LTV means the equity basis is going to be smaller than would be had the owner taken out an amortizing loan. A lower equity basis ultimately drives higher returns to the equity (as long as the cost of debt is not greater than unlevered returns on the deal - see above comment)

Now, let's analyze what goes into a loan payment and how everything is accounted for. Let's start with the definition of interest. Interest is the rate of return on a certain principal amount. This is something everyone intuitively understands (hopefully).

The bigger question is what is amortization? To be brief, amortization is the periodic repayment of PRINCIPAL. Amortization has nothing to do with interest (i.e. the rate of return on a loan). Amortization on a loan is negligible (again, for this simple analysis, not a full blown DCF on a core asset) because that principal is going to be repaid regardless. The only effect amortization has is higher payments earlier in the cash flow stream. Time value of money dictates that these principal repayments are "more expensive" than if you were to pay off the loan in its entirety at the end of the term (as in a 100% IO Loan). This is why an amortizing loan will return a smaller IRR than an IO loan (but not by much). Again, that principal is being repaid regardless and in the exact same amount, no matter how you look at it (IO, partial IO, full Am, etc).

Here's a test. I dare you to run a 10 year DCF on a constant (read unchanging) cash flow stream in excel. For fun, let's assume the exact scenario our OP has outline. Now, let's assume we run 2 separate capital structures. One with a full IO loan, and one with a fully amortizing loan over 30 years with a balloon payment at maturity. I'm willing to bet the IRRs you compute will be very close, but the IO version's will be slightly higher (and by slightly, i mean ~20 bps).

So, that is why in a back of envelope calculation (as outlined by the OP), amortization is negligible. Amortization simply dictates how and when principal is repaid. But, the cost the lender charges is the key, because that is the return the lender is receiving on his principal (in this case, he is receiving 4% for the 0%-70% portion of the capital stack). Because the unlevered deal returns are higher (4.75%), the debt is Accretive.

bing bang boom