Friday, June 19, 2009

Leverage and Return on Investment

Last week Barry Ritholtz at The Big Picture had a good post on  interest only CRE mortgages. In the comments I talked about why borrowers wanted IO (to goose the initial cash on cash return numbers), and in response one of the commenters made the point:

Maximizing leverage implies boosting ROI.

No, no, no. If anyone should have learned anything in the last two years, it’s that maximizing leverage does not always boost ROI. For leverage to boost ROI, income has to go up. When income goes down, leverage destroys you.

There are a couple more subtle cases where leverage doesn’t help you. Even if income goes up, leverage doesn’t help you much until you exit the investment, because debt service sucks up a lot of cash flow. And, there’s no guaranty cap rates won’t go up and/or financing will be unavailable when your leverage is due (again, a lesson that should be painfully obvious today).

Finally, leverage hurts you when equity is cheaper than debt (i.e., cap rates are lower than interest rates). This doesn’t happen often, because since equity takes the first loss it typically has a higher return than debt. But, it does happen. The usual case is in an inflationary environment when interest rates are up and equity is relatively cheap because investors believe rent boosts will provide additional return.

Below are some examples of how the numbers work. The first set of examples are high, normal, and no leverage scenarios when cap rates are lower than interest rates. The cash on cash returns are towards the bottom, and show what happens when net operating income (NOI) goes up a little, a lot, and down. Note NOI has to increase 38% to get the same return as a non-leveraged deal:


A small decline in NOI eliminates cash flow on a highly leveraged deal, but has a minimal impact on an unleveraged project.

The next example shows what happens when cap rates are higher than interest rates:


The next time someone extolling the virtues of leverage, you can send them to this post.