Bob -
Cap rate is a mathematical tool for modeling the return on a cash flowing asset. As far as I can tell, the math doesn't discriminate based on what that underlying asset is. Why should it matter if the underlying asset is a multi-family property, a single family property or a steaming pile of poop (if you can find one of those that cash flows)?
While I love Ben to death, I completely disagree with his assertion that cap rate is only valuable when analyzing multi-family deals (and I'm not sure that comment of his was cut-and-paste in an appropriate context). As an example, I've used cap rate to help analyze billion dollar business acquisitions; yesterday, I used cap rate to analyze a potential self storage deal; etc. I'm quite certain it's not just used for multi-family analysis.
Ben also asserted that in the single family world, comp analysis is used to value property. That's certainly true a majority of the time, but absolutely is not always the case. When I purchase single family rental properties, I don't give a crap about the market value (as determine by a CMA). If I can generate a 30% IRR on a single family rental, I'm happy to pay above market value...hell, I'll pay WAY above market value if the returns justify it. In many cases, I won't even do a CMA, other than to determine what my owner occupant competitors may be able to pay for the property. Market value is meaningless when the bulk of the return will come from cash flow and when the resale of the asset will be to another investor who will be holding for cash flow.
I have a feeling Ben will agree with both of these statements above (correct me if I'm wrong, Ben)...
So, Bob, if you're going to argue that cap rate isn't a good measure of return for one particular type of cash flowing asset (in this case, single family rentals), you're going to have to justify it better than just cutting and pasting someone else's ideas.
Please tell us why you (not others) think cap rate is suitable for analyzing/comparing the return of all other cash flowing assets, but automatically breaks down for single family cash flowing assets (or if you think it breaks down for other types of cash flowing assets, please tell what other ones it's not appropriate for).
To say that a mathematical model breaks down based on the physical configuration of the underlying asset (one unit versus many units) doesn't ring true for me. Maybe you're right, but I certainly haven't seen a logical argument here to support that view.
Please -- in your own words -- explain it to me and convince me.