Originally I started looking into AFLAC (AFL) thinking that this stock would be very cheap. It’s got an amazing history of increasing dividends, growing revenue year after year, and maintaining impressive returns on equity (ROEs). Not only that, but the stock has traded down by 31% over the past year, compared to the S&P’s performance of up 7%. Huge underperformance. Is the risk/reward in your favor yet?
- Shares: 467mm
- Price: $36
- Market Cap: $16.8BB
- TTM Fully Diluted EPS: $3.80
- TTM P/E: 9.5x
- Dividend Yield: 3.3%
- Book/Share: $25.65
AFLAC sells supplemental health and life insurance, almost entirely in 2 markets: The United States, which is 25% of their business, and Japan which comprises the remaining 75%. They have no European business to speak of, and generally have a distribution network of agents and brokers that sell policies mostly through the workplace. In fact, the company is the world’s biggest underwriter of individually issued policies sold at worksites. Some examples of insurance they sell are disability, life insurance, and cancer insurance, which is quite popular lately in Japan. Another big product offering is supplemental health insurance, that is coverage of medical costs not reimbursed by Medicare or by Japan’s national health insurance system.
The Business Model:
AFLAC, like other insurers, makes most of its money on the so-called float of premiums. That is, a person buying say a life insurance policy pays premiums for years, until his demise. At that point, AFLAC pays money back out to the insured’s heirs, but for years AFLAC is able to keep the cash premiums, earning interest and capital gains on the investments made with that cash.
Looking at the premium revenue generated against the benefits and claims expense is quite telling. With premiums of roughly $20BB a year, AFLAC will likely pay $19.5BB or so in claims expense. That is, today they pay around 95% of their premiums back out in the form of claims. Over the past 6 years, they have run between 93% and 100% so-called combined ratios. That is, premium revenue divided by benefit costs. Anything under 100% is a decent insurance business model as long as you can successfully invest the float. Now, the net income margins have been running at high levels, fluctuating between 8% in 2008, to 13% in 2010. At the end of the day, the insurance business here is well run, never losing money, and showing consistent growth.
The problem is that most income from the firm is derived from the float. And in a low interest rate environment, it becomes extremely hard to earn much in the way of interest income. But before forecasting where earnings could be in 2012, I think it’s worth delving into the balance sheet.
If most of the profits come from earnings on the float, then really this is an investment company. They have hundreds of loans, bonds, and fixed income investments on the balance sheet. On average they own single A-rated, long dated (or long duration) bonds. In total, they have $93BB worth of cash and investments. Now, similar to a bank, their investments fall into 2 buckets: Securities Available for Sale, and Securities Held to Maturity.
Securities Available for Sale (AFS) means that in theory AFLAC intends to potentially trade out of these bonds before they mature. Therefore, they are actually marked-to-market. If bonds in this bucket trade down to 75 cents on the dollar, then they get immediately marked to 75 cents, and the company takes an unrealized capital loss that affects assets and shareholders’ equity. Interestingly, they do not have to recognize unrealized losses on the income statement, but it does hit their capital account, or shareholders’ equity. (To be precise, these losses hit “Comprehensive Income” which is reported on a separate “Statement of Comprehensive Income” page, but you don’t see it on the official “Earnings Statement”. I know, bizarre. Kudos to the insurance lobby.)
So, at least the book value per share is a decent measure of much of their portfolio, and what that portfolio is actually worth. At least for AFS securities.
Now, the second bucket, Securities Held to Maturity (HTM) is a different story altogether. In theory, these bonds will be owned, never sold, and hence never get marked-to-market. The trickery that some firms use is that they have AFS bonds that trade down. Voila, instead of marking them to market and reducing shareholder equity, they move them to the Held to Maturity bucket. Then they don’t have to take the loss or reduce their equity.
I did like the fact that in Q1 2010, AFLAC moved $1.6BB in securities from the HTM bucket, to the AFS bucket. That is they took the losses and didn’t hide them. AFLAC seems to be doing the opposite of what a shady firm would do. AFL owned Tunisian bonds, and bank bonds from Portugal and Ireland, two of the infamous PIIG countries. (page 18, 10Q). Losses on these totaled $665mm, which losses impacted BV in the quarter. Fortunately, one area in which AFLAC has gone awry, owning Perpetual Preferreds of a lot of banks, are in the AFS category and hence are marked down to fair value.
AFLAC’s Held to Maturity Portfolio
This is where you have to dig in. In total, AFLAC has $32BB out of their $93BB of investments in this non marked-to-market category. AFLAC’s June reports shows pages of bonds that they own, and there is decent disclosure on their holdings. Eyeballing their Financial Analyst Briefing (or FAB) docs, shows that they own significant amounts of financial institution debt. That is, bank bonds in addition to their piles of bank preferreds (or perpetuals). As just pointed out, the perpetual book is marked-to-market, and in fact all of the losses here have hit book value. For the record, they are marked at 92 cents.
However, the bank bonds they hold are over-marked. They hold an astonishingly large percentage of their portfolio in bank fixed income bonds, some $27BB out of their $93BB. That is almost 30% of their assets. As of June 30th, these were marked at 95.7 cents. However, fair value was $706mm lower, around 93.3%. That $706mm represents future losses that potentially will hit the income statement and balance sheet. Keep in mind that reported losses are as of June 30th.
The bad news is that bank bonds have sold off quite a bit since June 30th. The CDS market for Euro banks is quite telling. In fact, RBS CDS have reached spread levels higher than those of October 2008, to around 350 bps today! That is around 125 bps wider than June 30th, and considering AFLAC owns $431mm of their bonds, implies about a 6% hit to bond prices (20 bps is around 1% hit to 5 year bonds). 6% times 422mm bonds is a hit to AFLAC’s book of $25mm just on RBS. They have similar exposures to Lloyds (LYG) and Barclays (BCS) too, with AFL owning 400-500mm of each of these banks bonds. That is 75mm of pain right there.
Honestly I am not really interested in going through all their bank positions, but generally a 3% hit to their $27BB of bank bonds is $800mm, or a couple bucks a share. On the plus side, as of June they held other bonds with enough capital gains to offset the losses, so that fair value at June 30th was actually quite close to the marked Held to Maturity portfolio. It’s what has happened since quarter end that is worrisome.
What I just don’t get about this company is:
- why they own so much financial risk in general, and
- why they own so much European financial risk.
They operate out of Japan and the United States exclusively. Why own European bank debt? 51% of their financial institution debt is from Europe and the UK. At least their sovereign PIIG exposure has been reduced to around $1BB of exposure in total, 70% is Spain, and 30% Italy. That’s far lower than the $27BB of financial exposure they have. But a billion is still a lot of money in PIIG sovereign debt. It should be negative $1BB.
So, with European bank problems surfacing this year, AFLAC’s management decided to “derisk” the portfolio. PIIG exposure has fallen from 6% to 3% of their portfolio since 2008. They got out of Greece. But they are still exposed and in a European sovereign restructuring, I suspect more losses to come. Naturally, I can only assume that the huge selloff in Italian and bank bonds in July/August was a big contributor to the selloff in AFLAC’s stock. Good news is that recent ECB buying of Spanish and Italian bonds have actually kept pricing about flat QTD. Wish I could say the same about European banks.
Tallying Real Book Value per Share
Book per share today is $25.65. Compared to a $36 stock price, that is a 1.4x price to book ratio. Historically going back to 2001, the price to book has averaged around 2.6x. That does make AFL seem somewhat cheap. I think my problem as pointed out above is that bank losses have probably hurt the balance sheet since June 30th. AFLAC has begun to derisk, and the smart play for management is to sell down European bank bonds they own. Let’s just assume that their $706mm in losses is $1BB now, and that their capital gains have been cut in half (roughly $1BB of capital gains). That would imply a hit of around $2 per share to book to around $23.65. With the stock at $36, that is a price to book multiple of 1.5x.
While AFL has traded as high as 3.0x book back in the mid 2000s, in 2008 during the crisis the stock got as low as 1.0x book. In fact, the stock traded in the mid 60s in early 2007, when book value per share was approximately $18. By early 2009, book had fallen to $14 a share, and so had the stock price! Quite a disastrous result.
Earnings are not nearly as smooth as management would like you to believe, and ROEs haven’t been as high. I think you want to buy this after they have a truly awful year, as in 2008, when the stock reached $12-14, then rallied back to the high $50s.
In another crisis, which perhaps is inevitable out of Europe, book will fall $4-5 per share again. A $36 stock could fall to the low $20’s in a panic riddled tape. That is my downside, call it $22 a share, or down 38%. Could happen very quickly if Europe cannot resolve its debt issues.
However, if AFLAC can grow book value over the next 2 years by say 10% per year (close to where ROEs are today), then book could reach $30 a share. EPS could actually reach $6.00 - $6.50 a share. At 2.0x book and 10x earnings, then that is $60 a share, just a smidge higher than where the stock was trading back in the early months of this year. 66% upside, and with dividends could total 70% upside in 2 years.
I think at $30, it’s a buy. The risk reward would be solid. Down $7-8, up $30 is a good trade, there is a pretty decent chance that the world prevents a bank disaster in Europe. We have been down that road before. There won’t be another Lehman Brothers. Lately too central banks from around the world are intervening to provide unlimited dollars to stressed European banks. Probably euroland will survive without a major panic in the financial system. If fiscal unity does happen in Europe, via a new authority and we see Eurobonds being issued, than that should buy 2-3 more years of bull market bliss, and cause AFL to rally pretty hard.
Some Notes on ROEs
In Q2 2011, annualized ROEs were 9.8%. From 2006 to mid 2011, ROEs averaged 9.7%. I measured this by using CAGR on book/share, the best true measure of long term ROEs in my opinion. You really have to ignore management’s “Operating EPS” and even have to be careful with GAAP EPS. The reason for such low ROEs? Low interest rates worldwide and wider credit spreads that impacted results in 2008-2009. This business benefits from accounting that obscures its true returns.
What you are buying
Given that book value is $12BB, and that they have $93BB in cash and investments, you are really buying a zero cost interest bank that is around 6.75x leveraged. Leverage is actually reasonable, but unfortunately they have chosen to invest in the wrong spots worldwide. Paying a big premium to book seems like a bad idea. Paying a reasonable price to book could make this a homerun. I would personally really load the boat at $25. It’s a solid franchise, and really in a low default world, is a money making machine. The problem is, they own the very securities I deem likely to default or at least will continue to hurt them. Bailout after bailout in Europe only prolongs problems with the banking system there. Do you really want to get in front of this?
I am not an insurance guy. I struggled to make sense out of the statutory filings. Feel free to add color or insight if you have this kind of expertise. Also, if you are a US investor, then understand that 75% of their portfolio and their business is denominated in yen. If Japan’s currency ever breaks down, then the hit to AFL would be painful. Their biggest portfolio position is in JGBs, Japanese Government Bonds. Not a lot to like there, yields are tiny, Japan's Debt/GDP is the highest in the world, and their population is aging as well as actually declining. Structurally there isn’t much hope for Japan long term, yields there eventually will go up, perhaps by a lot.
I would conclude by saying I am a better buyer today, gun to my head. But you might get a chance to buy this for cheaper later, the trend is definitely down. Look for news of a fiscally unified Europe perhaps to save the stock. I wouldn't buy this for the dividend yield alone though, as there's way too much macro risk for 3.3%.
Disclosure: I have no positions in any stocks mentioned, but may initiate a long position in AFL over the next 72 hours.