Saturday, November 1, 2008

AGNC: High-Yield Income For Market Bears

After releasing its 3Q2008 quarterly results, and fielding questions in a conference call with analysts, American Capital Agency (AGNC) is now trading above net asset value. As described before, AGNC is a REIT that owns no real estate. AGNC, rather, makes money investing in government-backed home mortgage instruments. The relatively modest rates paid on home morgages doesn't support outsized returns without leverage; AGNC's portfolio being comprised solely of high-quality agency-backed securities, it is capable of supporting enough leverage that even modest spreads support high net returns. However, management made clear in the call that the industry is in "the midst of what we would call a 'deleveraging vortex'" and that a REIT levered 15x will look in this environment as overleveraged as a REIT levered 25x in prior days.

In order to get ahead of this delevering curve, AGNC has delevered.

During the third quarter, AGNC reduced its debt:equity ratio from 8.3:1 to 5.4:1, selectively reducing borrowings under agreements whose terms management thought least favorable. Since revenues from holding debt are a function of leverage and rate spreads, one reasonably asks whether AGNC's income will be torpedoed by its relatively conservative leverage. Volatility may make a high debt:equity ratio dangerous, but it creates other opportunities.

During the last quarter, a dramatic increase in volatility placed AGNC in the happy position to make money selling options. Because AGNC invests only in agency instruments, and not in complex hard-to-value CDOs or the like, the quality and transparency of its holdings enable it to negotiate options agreements with potential buyers. By writing (in effect) covered call options against a few hundred million dollars' worth of the portfolio, AGNC collected substantial premiums (because volatility increases options premiums, and AGNC was selling) with which to protect shareholders against portfolio valuation risk. In the event the options are exercised, AGNC will sell securities for cash at a price above the market value at the time the options were issued. Considering AGNC's view that the market for agency securities remains liquid, and that there are numerous opportunities to be had in a market that has become inefficient (e.g., fifteen and thirty year notes trading at the same price for the same yield), the possibility of becoming a shopper with cash isn't such a bad risk.

There is an annual limit to the fraction of non-real-estate income that a REIT can earn while maintaining its tax status, but management indicated on the call that it was managing options income with that limit in mind. Moreover, AGNC did not anticipate long-term opportunities to use options rather than debt to achieve its desired leverage; management expected to maintain conservative leverage, but to enjoy increased spreads, resulting in adequate returns without the need of leverage levels that seem risky in the current market. AGNC's management observed that federal government intervention has thus far only had the result of widening rate spreads. Consequently, AGNC takes the (pessimistic) view that the mortgage market will continue to suffer. AGNC's strategy to profit from the mortgage market seems to benefit from mispricing, high mortgage interest rates, and other features of an ill home mortgage industry.

AGNC invests in what management described during the call as the highest-quality fixed-income securities in the universe. Following the Freddie/Fannie bailout, all AGNC's investments are in effect agency securities, backed by the United States government for the timely payment of principal and interest.

As foreshadowed in an earlier post here, AGNC intends retaining some earnings (within the REIT distribution limits) in order to moderate volatility in the dividend. AGNC's net asset value has actually increased since June 30, from $17.45 to $17.85. Management stressed that as AGNC grows, its results should improve because many of its material costs are fixed -- accounting, compliance, etc. -- and that its financial tools are intended to handle a much larger portfolio. AGNC's seemingly comprehensible strategy, its government-backed portfolio, and its apparently rosy future have bouyed AGNC's shares above NAV, something that ACAS' other managed but publicly-traded fund (ECAS) hasn't achieved, and something ACAS itself lost within the last few quarters. I imagine AGNC issuing shares to capitalize on dislocations involving investements with a long-term horizon, improving its margins, and enhancing existing shareholders' total returns.

The fact that AGNC is trading above NAV and is cranking out large cash dividends is probably a very good sign for AGNC's manager, ACAS, which holds something near 50% of AGNC's common stock. ACAS' holdings -- 7.5 million shares -- provide both dividend income (on the order of $7.5 million per quarter) and management fees (1.25% of the net asset value, payable in monthly installments, on approximately $265 million would be nearly another million per quarter), and both stand to increase as AGNC grows in value and profitability. The dividends and management fees are slightly above the 4¢ per ACAS share estimated here previously. AGNC's results -- and its premium share price -- imply an understanding of the current market that has not been attributed thus far by the market, in the form of per-share pricing, to ACAS. Part fo the pricing difference is a result of ACAS' increased opacity (investors would have a hard time understanding the hundreds of businesses which drive ACAS' results, and since those companies don't have individually-disclosed financials, investors are left to worry that performance might be silently slipping as the economy is jolted by economic shocks), and part of it is a result of AGNC's simpler-looking investment portfolio (valuing agency-backed mortgages is less uncertain and less speculative than valuing illiquid and more individually distinctive private companies). The issue here is comprehensibility: especially in a market panic, people fear what they do not understand. People have run for the doors in ACAS, and it won't be clear until November 10 whether fear led to safety or to abandonment of a good investment. The repeatedly-recurring allegation that ACAS' numbers result from fictitious accounting should tend to lose credence as ACAS and its publicly-traded managed funds continue to produce taxable income and continue to pay outstanding dividends.

As in ACAS conference calls, passing emergency vehicle sirens could be heard intermittently in the background. I wonder if management is saving rent money by buying in a bad neighborhood, or if the ACAS offices are simply near a hospital. Malon Wilkus appeared briefly on the call with a prepared statement, but the water was clearly being carried by specialists in mortgage instruments and not Mr. Wilkus. Mr. Wilkus' statement was fairly generic:
Despite the turmoil experienced in the financial markets during the third quarter of 2008, AGNC was able to navigate the market and deliver strong results to our shareholders. Our relative value approach to investing in agency securities and prudent management of our balance sheet has allowed us to produce excellent results for our shareholders while reducing our overall leverage.
--Malon Wilkus, Oct. 29, 2008
By contrast, some detail was offered by AGNC's Chief Investment Officer Russ Jeffrey:
Given the turmoil within the financial system in the third quarter of 2008, AGNC felt it was prudent to proactively manage our balance sheet through opportunistic sales of agency securities, which lowered our leverage ratio to 5.4 times. Market volatility created the opportunity for AGNC to develop and execute an option strategy, which was accretive to both ROE and earnings per share while allowing us to further reduce our risk profile. Overall, we remain optimistic about our ability to navigate these markets while delivering an appropriate risk- adjusted return for our shareholders.
-- Russ Jeffrey, Oct. 29, 2008
The specific solution utilized by AGNC to create value during the recent volatility -- options -- would not have been available to AGNC were its portfolio comprised of less liquid investments, and it is almost certainly not the solution ACAS could itself have used. ACAS' individual vestments -- debt and equity interests in privately held companies -- are not liquid. However, ACAS has a portfolio of nearly three hundred companies, and at the time of the last conference call was in the process of selling twenty-two of them. The question for ACAS is essentially whether its ability to achieve exits in a challenging environment is as solid as management seemed confident in August; other sources of liquidity seem unlikely to bear fruit for ACAS in this environment. On November 10, we'll know.

As for ACAS' managed company AGNC, the result is in: deft leverage management and successful underwriting has allowed ACAS to achieve a solid result in crazy times. The market's answer is clear: AGNC is worth more than the sum of its assets. AGNC's opportunities seem especially rosy as credit dislocations and market mispricing continue to create opportunities to identify solid values. The solidity of these values isn't based solely on federal guarantees, but careful underwriting and financial modeling that so far have played out quite well.

1 comment:

Anonymous said...

First, I appreciate your commentary on ACAS, AGNC and GOOG. Three of my largest holdings. However, again, my largest position is GFG - a $16 billion texas based thrift w/ nearly $1.1 billion in capital and strong franchise value in my home state.

if you remove home-state bias, of which i have plenty, the facts remain that GFG is grossly undervalued. Here's why:

1) Lack of Investor Coverage - most likely due to the recent spinoff. However, TRT Holdings (Robert Rowling), Carl Icahn and David Einhorn (Greenlight Capital) all participated handidly on the Private Placement - at +/- $5.17 per share.

2) Q2 Loss of 85MM - included in this was a $46MM charge to tax expense to establish a valuation on a deferred tax asset acquired in the spinoff. Net Loss w/o the charge and keeping the $100MM provision was closer to $37MM

3) Q1 Loss of $10MM - not that scary to a company with a fully diluted equity raise TBV of $8.74 per share and a BV of $10.31 and a risk based capital ratio of 14.6%

4) MBS Portfolio: Amort Cost basis of $5.11 billion vs Carrying Value of 3.7 billion. Non-Agency paper are ARMS w FICO > 700 and 80% LTV. Majority are HTM.

5)Diversified Loan Portfolio: GFG was NOT a subprime lender. They are hurting on the 12% of their loan portfolio being related to homebuilder exposure. These are regional and public companies, not entreprenuers and cowboys. - ie work throughs not work outs. Mkt is missing value in rest of Commercial LOan portfolio; " franchise value'

Earnings Prediction; or lack thereof - Loss of 50MM on provision growth.

Curveball - TARP Capital Purchase Prgm. Pluses - if they take, then TBV of 50% is in the cards in next 12 months due to the removal of any 'ongoing concern issues'
pluses - if they DONT take, investors find premiums in banks that aren't regulated by TARP rules, etc.

These are my highlevels on GFG. Earnings announced at close of business tomorrow. i am vested in this deal. not many other people are. time will tell...