GAAP vs TI

Divend Exceeds EPS... How's that work? I noticed that NFI paid out $5.60 in Dividends last year and they are projecting $5.60 again this year. But their EPS for last year was only $4.40. How do they manage to do this?

There is a short answer and a long answer. The short answer is these guys run a mean MREIT machine, and know how to maximize shareholder value to the hilt, with all kinds of market and non market forces against them.

The long answer is what you refer to as "EPS" means different things to different people and the IRS as it relates to MREITS. When you say this company has paid a higher dividend than EPS, you're obviously referring to GAAP (Generally Accepted Accounting Principles) which means very little to most MREITS.

GAAP
----------
GAAP is the most conservative financial metric an MREIT can use because it reflects NON CASH EXPENSES that may never impact the bottom line as a cash outflow. A major example of this is LOAN LOSS RESERVES. Like many companies, NFI takes a NON CASH hit to the P&L each quarter to account for loan losses that MAY OCCUR based on a stratification of its loans.

In many cases, these loan losses may never occur if the borrowers never go delinquent. Under this scenario, NFI has overstated its losses and understated its income. Beyond that, NFI also books an "Impairment on mortgage securities available for sale" that hits the P&L with a non cash offset to the Balance Sheet. Since the actual cash impact doesn't occur until the securities are sold, NFI is being conservative in taking a P&L hit for something that may never occur if it turns out they don't book a loss on the sale.

GAAP income dropped to 69cents in 1Q2006 because they changed how they accounted for securitization pools on their balance sheet, and the TIMING of how these expenses roll off and hit the P&L. No cash leaves the company in such a scenario; It simply means they have higher costs sooner and lower costs later, so over time it has no impact on their income statement. The simplest way to think about this is let's say you own an asset that goes up and down in value over time, but doesn't result in a cash loss because there is no permanent "impairment". Same concept here.

For all the reasons mentioned above, GAAP IS NOT RELIABLE WHEN LOOKING AT NFI'S EARNINGS POWER OR DIVVYS BECAUSE IT IS TOO CONSERVATIVE.

TAXABLE INCOME
-----------------------------------------
On the other end of the spectrum is TI, which is the metric upon which the divvy is based. Among other things, TI is the difference between what they earn on the investment portflio (commonly referred to as WAC) less cost of funds (NFI uses short term Libor and 2 yr swap as their average COF) plus any costs to service the portfolio, relevant 3rd party costs and the like. Since TI does not reflect certain non cash expenses that eventually becomes cash costs in the future, there are some who say TI overstates the true cash profits of the company.

I would agree and tend to look at true cash profits as a hybrid of GAAP & TI. In the past, if GAAP was $1 and TI was $2, I've used an average cash profit of $1.50. This is a "back of the envelope" estimate since I don't really know how much of their non cash GAAP expenses (ie loan loss reserves, securitization pool, etc.) ever make it to the bottom line as a permanenet impairment. That said, I'm very comfortable with the true cash profits of this company being in the $1.30 - $1.50 range in slow quarters and perhaps as high as $1.80 in higher volume quarters if WAC breaks their way.