In my initial Seeking Alpha article on AGNC this past February, I described an interest margin forecasting model I created based on these three variables: These three factors are already making a difference - AGNC's CEO noted that since the end of Q2 its book value had risen by 4%. The Fed has made it clear through their actions and their words that their primary monetary policy tool is adjusting the federal funds rate, not balance sheet reduction." "We believe the risk of asset sales by the Fed is extremely low. Third, AGNC believes that the Federal Reserve will not be dumping a lot of MBS on the market, avoiding a further increase supply: Less supply means higher prices for MBS, and therefore a narrower spread. Today, however, with mortgage rates higher, house prices elevated and the economy slowing, the net supply of Agency MBS is now expected to be closer to $400 billion, with most of that supply having already occurred in the first half of the year." "At the beginning of the year, the net supply of Agency MBS to the private sector was expected to be in the $700 billion range, which would have made it the largest issuance year on record. Second, the supply of new MBS coming to the market is slowing. Other investors are undoubtedly seeing the same thing, as evidenced by a 30 bp spread narrowing over the last few weeks. "Agency MBS spreads have widened by more than 100 basis points over the last year to end the second quarter…This yield differential rarely gets that wide or stays that wide for any meaningful period of time…At current valuation levels, Agency MBS are attractive by almost any historical measure…We believe AGNC is entering a favorable environment that will be conducive to generating strong risk-adjusted returns for our shareholders." The first is that MBS yields are now quite attractive, as AGNC notes: So where is that spread going next? AGNC's book valuation is likely at an inflection pointĪGNC management sees brighter days ahead for the spread, and I agree. Now you can see why AGNC's book value got beat up this past year - because the mortgage-to-Treasury interest rate widened dramatically, to historic highs. Sources: Freddie Mac, Federal Reserve and company financial reports This chart shows the correlation over time, with the mortgage spread inverted:įreddie Mac, the Federal Reserve, AGNC financials A widening spread typically reduces the book value and a narrowing spread increases it. GAAP accounting requires AGNC to mark-to-market both the MBS and the swaps each quarter, so these wide swings in valuations are reflected in the company's book value.
#MARGINNOTE PRO 3 MAC#
Sources: Freddie Mac and the Federal Reserve But in the short run, valuations of AGNC's MBS assets and its Treasury rate-based interest rate swaps can vary sharply from each other, as this chart shows: The hedging substantially reduces AGNC's interest rate risk and stabilizes its interest income over the long run. For example, at the end of Q2, the duration of AGNC's assets was 5.4 years and of its debt, 5.0 years. AGNC funds the MBS with repos (short-term borrowings) and then buys interest rate swaps based on Treasury bond rates that create debt with durations close to those of the MBS. AGNC's assets are almost exclusively fixed-rate MBS with expected lives ("durations") of 3-6 years. AGNC's book value dynamicsĬritical to understanding how AGNC's book value moves is understanding how it invests and how those investments are accounted for. This article lays out my case, borrowing liberally from AGNC's comments on its second quarter earnings conference call. If so, not only does AGNC's 11% dividend yield appear safe, but I think there is a reasonable case for a dividend increase within a year or two. What should we think about this stock considering Q2's big mixed messages? The company's view, and mine as well, is that it has just managed through a major interest rate headwind that is in process of changing direction. That's a pretty bad thing.īuy the stock at this probable inflection point. Its tangible book value fell to $11.43 per share, a 30% decline from a year ago.