Wednesday, January 11, 2012

Foolish Return Analysis

A while back, the Jaded Consumer ran an article on The Motley Fool's seemingly opposing investing advice and the dizzying array of stock picks offered in its numerous paid newsletters. In it, The Motley Fool's method of comparing users' stock picks to the S&P 500 index performance was criticized in passing as mishandling dividends. The point of this post is to illustrate that comparisons on The Motley Fool between users' stock picks and the S&P 500 are absolutely rubbish when dividends are involved.

The Motley Fool's method of handling dividends is not to add their return to the capital return. This method would err by ignoring the tax inefficiency of dividends, which are taxed in the year paid while capital gains go untaxed until realized, and err further by utterly ignoring the time value of money – that is, the difference in value of having a dollar of liquid return today versus having it paid to you in three years. Instead, what The Motley Fool claims it does is much worse: it subtracts dividends from users' basis. Confusing things further, The Motley Fool avoids negative basis by applying an undisclosed method to avoid ever hitting zero. According to the site, "The adjustment will be a fractional deduction, and not a flat dollar reduction. This prevents the cost-basis from ever reaching zero." The method also prevents anything like accuracy, or even duplication by third parties.

For purposes of example, and to show the absurdity of The Motley Fool's method by looking at a dividend stock, The Jaded Consumer will look at American Capital Agency Corp. (AGNC), which has been written about here on numerous occasions (mostly because it is managed by ACAS, about which much is also written here).

Let's have a look at a user who picked AGNC as a "buy" at The Motley Fool. The "buy" pick for this user was made on Thursday, September 4, 2008 – a day in which AGNC traded between $18.25 and $18.69. To give The Motley Fool the best possible case of accuracy, we will assume that the user made the pick at the exact time of day needed to nail the day low of $18.25. Given today's closing price of $$28.11, the user would (if the user had made a real buy, and held) enjoy an unrealized (and untaxed!) gain of $9.86 – for a not-too-shabby 54%. (By comparison, the SPDR S&P 500 ETF Trust – designed to mimic the performance of the S&P 500 – which traded at a low of $123.96 on 9/4/08, closed at $129.15 today, thus yielding a rather lower return even after considering several years of its single-digit dividend yield.) Holding for all this time would have entitled a user to some dividends, the history of which is readily available from the NASDAQ web site. Holding from 9/4/08 through the 1/27/2012 payment date would have entitled an investor to a $1.00 dividend, a $1.20 dividend, a dividend of $0.85, a dividend of $1.50, and for the last ten quarters a dividend of $1.40 apiece. This totals $18.55, which (ignoring time-value-of-money considerations) would amount to 101.6% of a the hypothetical $18.25 purchase price. Adding the capital gains and dividend returns would yield at best (ignoring time value of money considerations) a gain of $28.41, or 155.6%.

Yet, the CAPS participant we're looking at, having given the thumbs-up and a 3-year investment horizon to ACAS on 9/4/08, is credited with having already bagged a whopping return of 212.93%, even though the results at The Motley Fool cannot yet reflect the 1/27/2012 dividend (which hasn't yet been paid and won't be owed a shareholder until the ex-dividend date passes):

The real mystery is the purported Start Price. AGNC never traded for $9, ever. And this user's happy return is not really greater than a total return of 300% (yet) as suggested by the 212+% gain claimed at The Motley Fool.

The Motley Fool CAPS system that produces these bogus returns is not the only place one can get confusing results. The Motley Fool has a portfolio-tracker that will help calculate an annualized return that can be compared to that of the S&P, but that tool ignores dividends altogether. An investment in AGNC over the last year doesn't look like it's enjoyed a double-digit return, it looks flat.

For a site dedicated to helping users understand investment returns, The Motley Fool has a few things to straighten out – particularly with respect to making sense of the return obtained from dividend investments.

MTGE Likely Better Than Its Dividend

The Jaded Consumer has a new article at Seeking Alpha.


Thursday, January 5, 2012

MSFT Imperils Ford's Quality Reputation

Ford's decision to use a Microsoft platform to power its user-facing electronic controls includes a price paid in lost quality reputation. Of two quality issues that drove Ford from the #10 spot to the #20 spot in automotive brand survey rankings, one is the unstable and user-baffling MyFord Sync. Where "Quality is Job One", Ford's Sync system was outsourced to Microsoft.

Ford can do better. Maybe not Microsoft, but definitely Ford.

ACAS Repurchases Raise NAV

Share buybacks at ACAS were new in 2011, but there's some background the investor should be aware of, in order to understand the limited time-frame in which share buybacks are likely and the specific effect of share buybacks on an investment company trading below its net asset value.

The Old Days
In the halycon days in which American Capital Ltd. (formerly "American Capital Strategies") traded in the $30s and $40s, stock repurchase made no sense for the company. ACAS' BDC status required it to pay as dividends at least 90% of its taxable earnings, leaving ACAS little opportunity to plow cash into share buybacks without reducing its deployed capital. And here's the kicker: because ACAS' debt covenants included net worth terms requiring ACAS to maintain a certain level of net assets, ACAS couldn't risk reducing its deployed capital. Share buybacks would require shrinking the company, and – even ignoring the above considerations – management made it clear on several conference calls that it wanted to grow the company and grow the breadth of share ownership. Apparently, management wanted to grow American Capital as a brand for individual participation in the private equity market. Share buybacks were anathema to this mission.

Worse: on occasions when ACAS traded above its net asset value, buybacks actually diluted value per share. That is, if ACAS were to have paid $40 on the open market for shares backed with $38 per share in net assets, ACAS would in effect have lost $2 per repurchased share – the benefit of the $2 would flow to selling shareholders rather than ACAS or shareholders who elected to hold.

How The World Turns
Share buyback has been discussed by investors since the crash because ACAS' share value has been notably in the toilet. Last year the Jaded Consumer summed the situation thusly:
ACAS' management made it very clear that it didn't think its existing cost of borrowing was attractive: ACAS prefers to borrow at AAA rates, and will repay debt to get there. Share buyback, by contrast, constitutes a leverage-increasing transaction that impairs liquidity without improving the balance sheet: don't expect share buyback soon.
On ACAS' 4Q2010 and full-year 2010
By 2010, American Capital's debt restructuring eliminated its net asset covenants. By mid-2011, a deliberately failed RIC test eliminated ACAS' dividend-paying requirement (not to avoid dividends, but for a tax purpose: this allows ACAS to roll forward operating losses from the crash era so that taxes can be avoided on more future earnings). ACAS is therefore not obligated to shed any of its earnings, and has no special incentive to maintain a specified amount of deployed capital. Under the new debt agreements, ACAS' lenders are secured by ACAS' investment holdings (look at the 10-K; the superscript numbers offer a guide to each portfolio company's situation), but have no claim on cash ACAS generates form its portfolio companies.

And instead of being required to distribute cash, ACAS is able (due to its loss carryforwards, which for the time being prevent it from having a tax payment obligation) to retain everything it makes and invest it as it pleases, without concern for enterprise value. With shares trading at something like 0.6x the last-published NAV, ACAS has an interesting investment opportunity that previously was unavailable. Sure, ACAS purchased publicly-traded ACAS debt for below face value when sentiment against ACAS was at its lowest and the debt was especially cheap (buying a $1 obligation for $0.60 causes an immediate risk-free gain of $0.40, for example). But as described above, ACAS had been loathe to buy shares. And with a debt:equity ratio currently reported to be 0.41:1, ACAS has little fear that buying shares will threaten its leverage discipline (or its 1:1 BDC limit).

The New Deal
When ACAS first announced its plan to repurchase shares on the open market, it announced a $75m purchase of 9.1 million shares. As described earlier, the impact of the transaction was to raise ACAS' NAV by 18¢ without any tax consequence. In other words, a company paying 35% taxes would have had to earn 28.25¢ to get the same after-tax impact –a feat ACAS accomplished without consuming any of the valuable loss carryforward that drove it to abandon its BDC tax status. Due to the market's sharply downward action in 3Q2011, this 18¢ was swamped by the downward pressure on all the stocks whose values are used as comparables in the calculation of the fair values of the holdings that comprise ACAS' investment portfolio.

We don't yet know what the NAV was at the end of the quarter ended December 31, 2011, but based on the 3Q2011 NAV of $11.92 per share, ACAS' repurchases during the quarter at an average price of $6.97 per share had an impact on ACAS' net asset value of $4.95 per retired share. Since ACAS retired 8.4 million shares over 4Q2011 (at a cost of $58.7 million), the value ACAS added through underpriced share retirement is $41,580,000. Working from the 344,800,000 shares outstanding at the end of 3Q2011, ACAS at the end of 4Q2011 would have had no more than 336,400,000 shares (less, if ACAS repurchased shares during 4Q2011 prior to November). This implies an immediate tax-free impact on NAV of about 12.4¢ per share.

There is another way to measure the impact of share buybacks than to think of the repurchase's impact in terms of the pennies per share it is worth in the quarter in which it occurs – which is tricky anyway, given the movement of NAV over a quarter due to the impact of price changes among the portfolio's known-priced comparables. One can instead look at the percent increase of the company's earnings and NAV gains the buybacks will cause as earnings and net assets are distributed across fewer outstanding shares. The 336.4 million shares ACAS seems to have had at the end of 2011 is 18.3 million fewer shares than ACAS held at the end of 2Q2011. Share buybacks haven't just improved NAV performance per share, but have had an anti-dilutive effect of more than 5%. That's 5% better performance per share, without having to reinvest dividends on which investors would have paid taxes. Share buyback is, in effect, a tax-free DRIP.

Given that Net Operating Income at ACAS has been in the neighborhood of 20¢ per quarter (in 2011's Q1, Q2, and Q3), a 5% concentration would seem to add 1¢ of NOI per share per quarter. This is a performance improvement superior to ACAS' fee income from capital managed under a 28,000,000-share issuance at AGNC, which yielded approximately 0.7¢/sh/q. Moreover, the concentration of ownership means that as markets rebound, yields will also be magnified by the 5% reduction in share count. Factoring in ACAS' leverage, the concentration should be good for those who continue to hold shares.

The Status Quo
ACAS currently trades at $7.25. Since ACAS' 4Q2011 report (and per-share NAV at the end of the period) has yet to be published, it's not clear just what discount $7.25 per share is to ACAS' current NAV. However, since ACAS' diverse collection of portfolio companies move in value with the broader market's collection of comparables (from which ACAS' portfolio companies' value is estimated from quarter to quarter), and since ACAS is leveraged, it follows that the broad gains in the market lately have buoyed ACAS above the $11.92 per share NAV that existed at the close of 3Q2011 and that current prices result in a NAV discount greater than 39% or $4.67 per share. Buying a profitable investment at a discount greater than 40% is exciting stuff for a value investor.

ACAS suffered a massive flight when its investors – attracted by BDC dividends and ACAS' track record of increasing dividends – bolted following ACAS' dividend cessation. When ACAS finishes consuming its loss carryforwards, and resumes meeting RIC tests, ACAS will once more be obliged to pay dividends. The NAV discount should begin to evaporate as that occurs; though the continuance of the discount would be favorable to DRIP investors eager to add shares of a company managing an internally-diversified and leveraged portfolio whose contents are artificially depressed in price because FAS 157 requires illiquidity discounts and other factors to be applied to NAV – yet which produce a return based on their actual intrinsic value.

Another note about the status quo: a dividend that changes over time raises worries from observers who could conclude that a lowered dividend is a sign of financial weakness. ACAS, by implementing a share buyback program rather than a dividend, avoids this sort of criticism while offering shareholders a tax-efficient way to enjoy management's willingness to return money to remaining shareholders. The fact that ACAS has been able to pay more than $50m per quarter the last few quarters in share buybacks is interesting, but should not be mistaken for a guarantee of its initial dividend. ACAS' share buybacks have generated NAV improvement of nearly $5 per retired share, which makes them an investment. Dividends are not an investment, do not raise NAV, and are not tax-efficient except where the BDC rules make them a condition of corporate tax avoidance. Since many investors' tax rates are less than the corporate rate of 35%, dividends under that circumstance are definitely efficient with comparison to the alternative (corporate rates, and if combined with a dividend, double-taxation). The current situation will likely endure as long as ACAS is required to fail RIC tests to avoid losing loss carryforwards.

The Future
ACAS has already moved from making only limited follow-on investments in existing portfolio companies to participating in genuinely new investments again. ACAS is a $10m participant in RBC Capital Market Corp's financing of Permira Advisors' purchase of Rennaisance Learning Inc., and a $15m participant in BMO Capital Markets' financing of Teachers' Private Capital's purchase of most of Flexera Software Inc. ACAS' follow-on investments have grown in scope: in the second half of 2011, it invested $50m in the mezzanine financing of the merger of Survey Sampling International and its existing client Opinionology Inc. ACAS is now no longer treading water, but apparently back in business. Its aggressive shedding of debt has put it in a position to retire shares without impairing its ability to enter new investments, and offers ACAS flexibility in determining the terms with which it may more favorably restructure its debt.

ACAS looks like it's heading in a productive direction. I'm still wondering what the "special situations group" has been doing while the markets have been in turmoil: haven't they spotted a "special situation" in which to invest? Ahh, well. Things are looking good for the present and foreseeable future. Wondering about the fish that got away never profited anyone.