Understanding American Capital's 2011 Results

| About: American Capital (ACAS)

Following publication of "Loving American Capital Before Valentine's Day", commenters here and at the blog asked about the company's dividend and its freedom to flexibly deploy capital without a revolving line of credit. Some readers interested in American Capital (NASDAQ:ACAS) apparently weren't in a position to listen to the earnings conference call - or, having heard it, were left disoriented by the enormous fourth-quarter effect of a tax asset, the impact of which dominated fourth-quarter operating earnings.

The Basics
Earlier this week I identified net asset value per share and net operating income - NAV and NOI - as places to measure management's performance at American Capital Ltd. The reason these metrics are important at American Capital is that other metrics quickly confuse observers with double-counting and accounting fictions that conceal actual performance. NOI offers investors a quick snapshot of the company's performance outside its portfolio gains/losses, and NAV offers investors a view of management's overall performance in managing the value of the portfolio (ignoring whether the constituent investments are exited, held, entered, merged, etc.).

For example, American Capital is required by FAS 157 (an accounting standard) to report unrealized gains and losses as part of its SEC-reportable "earnings." To judge performance by both "earnings" and NAV gains would therefore double-count management's investment-based performance by showing unrealized appreciation twice, while leaving one unsure what gains were really made outside of changes in the value of portfolio holdings. Investors presumably want to know answers to both (1) How is American Capital's steady-state business going, and how is management performing in its role as a manager of businesses it owns? and (2) How is American Capital's portfolio doing, and how is management performing in its role as an investment manager? Management's success as a business manager helps investors understand American Capital's capacity to patiently wait for good deals (it it's cash-strapped or losing money quarterly, one fears pressure to raise liquidity with ill-advised exits or dilutive issuance). Management's performance as an investor is the basic reason to buy a Business Development Company. Both questions are key, and NOI and NAV are the easiest windows into management's performance.

On Net Operating Income
The first question - about the NOI - is complicated because NOI in 4Q2011 includes a tax asset which resulted from crash-era losses and would have already have been on the books if (1) American Capital had been taxable as a C-corp then, and (2) American Capital's earnings made it evident earlier that it would be able to use the tax asset. In short, the appearance of a $428 million deferred tax benefit during the quarter is not a result of some strange tax event during the quarter but results from the year's profit finally entitling management to begin stripping a valuation allowance from the balance sheet that completely eclipsed the value of a $1.2 billion tax asset. $1.2 billion, you say? Yes. American Capital, by making more profit, could realize the benefit of even more deferred tax benefits than it has been permitted to book. This isn't earnings the way you and I think of earnings, but by being able to earn money without paying taxes (due to the loss carry-forward behind the huge tax benefit) it magnifies the portion of earnings that consists of taxable income. Since corporate tax rates are 35%, going tax-free for a while amounts to the same thing as growing 65¢ in earnings to $1, for a gain exceeding 50%. Thus, American Capital gets nearly 54% more bang out of any taxable income realized while the deferred tax benefit holds out.

This still hasn't helped us understand American Capital's NOI - it's just helped us understand why the NOI number published on Valentine's Day for the 2011 calendar year is not very useful in its raw form, and why current buyers of American Capital shares at these discounted prices stand to get turbocharged benefit from earnings for a while. The answer lies on page 4 of American Capital's 4Q2011 shareholder presentation, which is avialable on American Capital's web site. The 2011 NOI excluding the tax benefit was $0.85 per share, or about 8% of the NAV with which American Capital entered the 2011 calendar year. To appreciate whether American Capital's finances support its intent to behave as a patient investor, it's also reassuring to see that during the calendar year American Capital collected $1.24 billion in liquidity from payment of debts owed it, investments exited, and so forth. The third quarter of 2011 was the first quarter of negative earnings since the United States' GDP turned positive in 2009, and in it American Capital had the liquidity not only to make new investments but to launch a new publicly-traded managed fund and to initiate a share repurchase program. American Capital is liquid enough to carry on its intended operations just fine.

On Net Asset Value
The second question should be easiest to answer. During 2011, management raised NAV from its year-end 2010 value of $10.71 to $13.87, a gain of $3.16 per share, or 29.5%. Cool, right? American Capital has more money with which to make shareholders money, right? But think to the NOI problem. American Capital just credited itself a deferred tax asset that now sits on its books as a tax asset. Management can't invest a deferred tax asset, and a deferred tax asset pays no interest. A deferred tax asset simply shields the company against tax liability until it runs out. Now that American Capital carries on its balance sheets $428 million of the $1.2 billion in deferred tax benefit it acquired around the time of the crash, it will instead of paying taxes on taxable income chew into its deferred tax asset: neither earnings nor NAV will distinguish from here on the difference between American Capital being taxable or non-taxable for the next $428 million in taxable income.

So we can either take the accrual approach and view American Capital's temporary tax holiday as a capital asset that will be consumed as American Capital earns taxable income, or we can try to look at what changes in assets actually occurred during the year that don't include the tax asset, then remind ourselves that future earnings will be turbocharged by a temporary tax holiday that will boost taxable income's value to shareholders by 54%. Removing from NAV the $1.30 associated with the deferred tax benefit (see slides 3 and 4 in the 4Q2011 shareholder presentation), we would see the end-of-2011 NAV as $12.63 - an increase of $1.86 over the prior year, for real NAV growth of over 17%. In most businesses, a growth of shareholder value by 17.4%, particularly combined with the fact that any taxes built into the gain will be deferred until disposition, would be considered an outstanding home-run success. To understand what kind of result this is, however, I think a little perspective is necessary.

First, think about what sorts of businesses American Capital has been operating over the reported year. According to the earnings call and Slides 6 and 26 of the 4Q2011 shareholder presentation, American Capital has employed some 40,000 people to collect some $3.1 billion in revenue, capturing some $711 million in EBITDA across approximately 200 portfolio companies. The portfolio's diversification is depicted in the shareholder presentation:

This size and diversification of operation makes American Capital a rough proxy for the market as a whole. Standard and Poor's offers this breakdown of the S&P 500 (NYSEARCA:SPY):

Sector Breakdown
Some interesting differences between holding a fund that tracks the S&P 500 and holding American Capital are, of course, that American Capital has a taxable-income get-out-of-jail-free card that hasn't expired, and American Capital operates with 0.3:1 leverage. Yet, these alone can't explain the substantial differences in performance between the performance of the two baskets of investments. Over the calendar year in which ACAS grew NAV 17.4% (excluding the tax asset recognition), the S&P 500 closed with the smallest percent change since 1947 - a decline of 0.0028% - which including dividends would have provided a pre-tax yield of 2.11%. With a NAV increase of 17.4% (excluding its tax asset), ACAS trounced the S&P 500 - and did so without exposing investors to any current-year taxes.

American Capital is therefore not only liquid enough to pursue its intended investment strategies, it is fully capable of executing on them for the benefit of shareholders. American Capital's effectiveness as an asset manager, as measured by its own management of its owned assets, is outstanding. (For a view of American Capital's effectiveness as an asset manager of others' assets, consider its recent performance at the publicly-traded funds American Capital Agency Corp (NASDAQ:AGNC) [recent article here] and the recently-launched American Capital Mortgage Investment Corp. (NASDAQ:MTGE)[recent article here].

The Basic Case for American Capital
The 2011 NAV performance at American Capital is the benchmark for assessing management's performance as an asset manager, and as one can tell from the performance of the broader market during the same period, American Capital's performance as an asset manager of its own assets was nothing short of phenomenal. Only part of this can be attributed to the 0.3x leverage: it takes much more than a little leverage to move a performance like that of the S&P until it looks like that of American Capital. The difference is the performance of the portfolio companies themselves.

American Capital is a private equity company. Unlike a mutual fund full of publicly -traded securities [who wants to pay some CFA to hold shares of Apple (NASDAQ:AAPL) for you?], American Capital invests in assets not available to the small investor. When American Capital shops for portfolio companies, it is not looking at firms whose shares have a liquid market on a public exchange. Illiquid portfolio companies' valuations are generally subject to an illiquidity discount, which compensates buyers for the opportunity cost of investment in an asset that can't be quickly traded out when another interesting opportunity knocks. Just as private takeovers of publicly-traded companies involve well-publicized premiums associated with the control block sought by the buyer (ever wonder why takeover bids aren't at market, but at a premium often near a third of the shares' value pre-offer value?), illiquidity discounts affect the prices paid for companies that aren't being exchanged in tiny minority increments on a liquid public exchange.

By buying illiquid portfolio companies for which no exchange exists, American Capital involves itself in a market in which the availability of exceptional deals is higher for several reasons. First, the lack of a public market for the shares means that owners can't simply liquidate to a near-infinite number of unknown distant market participants by placing an order with a broker. This means that the market has relatively few buyers, and that prices may vary wildly based on the selling pressures (did an owner die, leaving competent managers worried about squabbling heirs, motivating them to lead an employee takeover before the company is destroyed by the new owners?) and the willingness of buyers to spend tens or hundreds of millions on a mid-market company that may need substantial operational improvements before it will be attractive to a future buyer. Prices can be affected by market cycles, personality, prevailing P/E ratios in the public markets, the appeal (or lack thereof) of the business' industry - all of which can provide an outstanding entry price for a new investment. Second, buying with an illiquidity discount allows firms to buy a particular amount of EBITDA for less than one might buy a comparable company that is a publicly-traded constituent of the S&P 500. The same earnings for less money is a recipe for long-term returns. Third, American Capital has long claimed to be a patient long-term investor. American Capital need not buy a particular company, however attractive it may be, unless it's priced properly. American Capital can keep saying "no" over and over until it has a steal on its hands. By buying only carefully-selected underpriced winners, and doing so in a buyer's market under the influence of an illiquidity discount, the basic premise is that American Capital should be able to build and develop a diversified portfolio, the return of which substantially exceeds those of the broader market.

Particularly in management's view is correct that the recovery is underway, American Capital's prospects for future appreciation appear extremely favorable.

New Investments
During 2011, American Capital deployed $317 in new investments, closing the year with 203 portfolio companies (down from a 2008 high of 296). Among the new investments was $40 million committed to a new managed fund that is already paying a dividend equal to 16% of American Capital's purchase price (equal to about 2¢ per American Capital share of taxable income per year) while additionally paying American Capital a 1.5% management fee on all funds within the new investment (currently, about 1¢ per American Capital share annually). The combined ~3¢ per American Capital share of taxable income amounts to an ongoing return on investment well north of 24% per year, excluding capital appreciation, just from one new fund. This return will likely rise with secondary issuance at its new externally-managed fund. As described in the pre-earnings article, American Capital by the close of 2011 had begun earning over 25 cents per year per share of American Capital from its two publicly traded external funds.
This wasn't the end of American Capital's investment, though. American Capital trades at a discount to NAV exceeding 30%. American Capital has begun purchasing shares on the open market at less than NAV, providing an immediate and non-taxable net-asset-per-share benefit to its shareholders of approximately 50% of the funds dedicated to the share buybacks. An overnight 50% return isn't a bad deal. American Capital has also paid down debt over the calendar year in an amount exceeding one billion (with a "b") dollars. This immediately creates a risk-free return equal to the interest rate avoided by repayment, but it has also de-levered the company to a debt ratio of 0.3:1.

American Capital is Hated, And You Can Profit
American Capital broke the hearts of a number of dividend investors in 2008 when it suspended its dividend. As a Business Development Company taxed as a Regulated Investment Company, American Capital was required - much like a REIT - to distribute at least 90% of its taxable income. American Capital's situation during the liquidity crisis was impacted by two mutually-magnifying problems. First, FAS 157 required assets to be valued at their immediate liquidation value. during a crash brought on by a liquidity crisis that actually caused negative interest rates to be paid for a while by the United States treasury, illiquid little portfolio companies of uncertain credit were virtually worthless and their debt more so, particularly if they failed to make timely payment due to temporary business trouble brought on by the broader economy. This caused another catastrophe for American Capital: its unsecured lenders had American Capital's promise that it would maintain a net asset level of a certain number of billion dollars, and when this asset covenant was breached American Capital's debt was immediately in default, notwithstanding that American Capital made every payment timely (even at default-rate interest). Renegotiating American Capital's debt to eliminate the technical default took over a year, during which American Capital not only paid default-rate interest on its substantial pre-crash debt, but also paid enormous fees to the legal teams working on the debt renegotiation. (When I write "teams" I refer to the overwhelming certainty that the consequences of being in default with each of American Capital's lenders included payment of any legal fees they incurred collecting the "defaulted" debt. On the renegotiation of the debt, American Capital overnight dropped not only the default-rate interest obligation, but also the obligation to float the huge invoices generated by each lender's legal team and its own outside counsel retained to develop the strategy that kept American Capital operating despite broad fears it would be liquidated in a bankruptcy auction. At one point, shares were less than 60 cents apiece.)

Since the dividend suspension, American Capital has consistently traded at a discount to its net asset value. Until it renegotiated its debt, it was not free to retire shares, but it did manage to repurchase some of its own debt well below face value for an overnight gain. Still, the company is viewed with suspicion. Despite the mathematical beauty of American Capital's shareholder-friendly below-NAV share repurchase program, widely-read publications seem to find fault with American Capital that they would never find fault in another asset manager conducting even a less-favorable buyback.

For example, The Motley Fool decries the share buybacks as inappropriate because American Capital, while still bearing the burden of its enormous defaulted pre-crash debt, issued shares for less than American Capital is now paying. At the same time, The Motley Fool uses the share buybacks at Berkshire Hathaway (NYSE:BRK.B) as a shining example of how a company can buck the trend by making good share repurchases. In so many articles one finds The Motley Fool consistently praising Warren Buffett and his management of Berkshire Hathaway - with justification - but apparently their attention to detail is slight. How can Berkshire's repurchases be friendly and American Capital's hostile when (1) Berkshire's buyback program authorises repurchase above book, and American Capital is buying well below net assets per share, and (2) Berkshire issued Class A shares (NYSE:BRK.A) at $111,452.57 when it bought Burlington Northern in 2010 (and Class B shares at $74.30), making its current repurchases above its last-issued price, which is the very sin laid at the door of American Capital. Non-insiders can't claim knowledge of the rationale behind selling Paulson a stake in American Capital on sale nearly two years ago, but the fact that American Capital was in default of billions of debt and could stand to benefit from a relationship with a deal-partner might be real reasons to suspect that the primary substance of the transaction was not to issue shares. Assuming that issuing shares to Paulson was worse than issuing Berkshire shares to Burlington Northern owners, this still offers no help at all to the problem of ascertaining whether repurchase of shares now is good for either company at present (higher) prices. If buying back near $80 shares it issued near $74 is good for Berkshire when the shares are slightly above book value (which includes intangibles like goodwill), how can buying at a third or more less than NAV be anything but good for the returns of current shareholders of American Capital? Taken on its face, the logic of The Motley Fool would forbid either company to pay more than its last-issued price ever again, no matter how undervalued the shares or how cheap the repurchasing cash.

The conclusion to be drawn from the completely different standard applied to American Capital under Malon Wilkus than to Berkshire Hathaway under Warren Buffett is that our affections rather than mathematics and economics should dictate the predicted performance of proposed investments. This is most certainly the exact opposite of the guidance endorsed by Warren Buffett. If Berkshire can be lauded for repurchases near but above book value per share, American Capital's repurchases substantially below net asset value that immediately accrue to the benefit of shareholders' NAV per share is certainly a benefit any time the opportunity cost of the spent cash doesn't eclipse the 50% overnight gain to be had buying shares one-third below NAV. The real question must be, what opportunity must American Capital be missing to make this a bad deal?

The answer is that American Capital has bid on numerous deals throughout 2011, but has been outbid. Shareholders should laud American Capital's discipline in not chasing investments at higher and higher prices until the deal is a dog. Shareholders should hope for patient and diligent work to acquire high-quality businesses at a good price, which in the market for illiquid privately held companies should provide some of the cheapest EBITDA to be found.

So long as American Capital remains despised, suspected, and hated, we should continue to be able to acquire shares at a substantial discount to NAV.

Historically, American Capital was not only a Business Development Company but also a Registered Investment Company. As a RIC, American Capital was allowed to avoid income tax by paying at least 90% of taxable income as dividends to shareholders in the same manner as a REIT. The benefit of this is that shareholders' after-tax returns reflect only one tax event - at the time of their receipt of the dividends. Berkshire Hathaway pays no dividend precisely because of the tax inefficiency of the double-taxation ordinarily occurring when a company pays taxes on its taxable income, then again when investors receive taxable dividends (which is why "qualified dividends" for the time being are afforded a lower tax rate: they've been taxed once already). In order to preserve the $1.2 billion tax loss carry-forward it acquired in the crash, American Capital became taxable as a C-corp during 2011. Its tax status now carries no dividend-paying obligation.

In the interest of tax-efficiency, American Capital would do well to avoid paying dividends until it is again able to do so without double-taxation. However, in 3Q2011, American Capital announced a dividend policy that made plain that it would in fact return capital to shareholders, though it would do so through share buybacks while share price trailed NAV. Repurchasing shares below NAV is an outstanding way to provide value to shareholders: it doesn't result in a risk of current-year taxation to anyone but the selling shareholders, it allows in-effect tax-deferred reinvestment of the funds that would have been paid as a dividend, and best of all it allows the continuing shareholders to benefit in the form of increased NAV per share from the discount from NAV by which American Capital is able to repurchase shares in the market. Management made plain that it continues to see outstanding value in the shares and will continue to repurchase them until the market stops pricing American Capital shares below the value of the assets held by American Capital for each share.

The other thing to like in the share repurchase is that American Capital's fee-based management services do not depend on American Capital's investment of capital. For example, American Capital no longer owns the share block of American Capital Agency that it bought in the private placement simultaneously with the fund's IPO in 2008. (Which was before the crash, through which American Capital led American Capital Agency with outstanding performance - thanks in large part to careful and diligent hedging.) Without any capital tied up in the shares, American Capital shareholders earn nearly 6 cents per quarter on each American Capital share from American Capital Agency fees. The more shares American Capital retires, the more concentrated this fee income. Especially while NAV remains north of market prices, American Capital's dividend policy of making share repurchases in lieu of dividends is certainly the correct way to provide a return to shareholders.

Management expressly stated during the Q&A that it did not view dividends as linked to its tax status. In other words, dividends are not contingent on the consumption of the loss carry-forward or on a corporate conversion to a tax structure designed to avoid double-taxation. Investors wishing for a dividend can take heart. On the other hand, investors wishing for an optimized after-tax return would cheer the idea that American Capital pay no dividend until it must, as dividends will be taxable to shareholders when paid and if American Capital is able to reinvest without paying taxes, that is clearly the tax-efficient course of action for shareholders.

Also, the more capital that remains deployed by American Capital, the more rapidly American Capital should expect to consume the loss carry-forward. Since the loss carry-forward pays no return on its own, it's not the sort of asset shareholders can really get excited about. It's extremely valuable at preventing tax liability, but it's worthless as a productive asset, and taxable income is absolutely essential to deriving value from it. If you are reading this, Malon Wilkus, this is a plea for more untaxed reinvestment and less threat of taxable shareholder distributions.

However, taxable shareholder distributions are not really a risk in the near term. Until American Capital trades at a premium to NAV, there's little risk of taxable distributions. Until American Capital resumes its dividend, there's little risk that American Capital - which among BDCs is reviled for its lack of dividend - will trade at a premium to NAV. The track record for NAV premiums among funds managed by American Capital has been that at both American Capital Agency and American Capital Mortgage Investment , shares traded at a discount to NAV until it became evident that management was paying a high dividend to shareholders. The high dividend was affordable largely because it was paid from pretax dollars: if American Capital Agency's recently-announced $1.25 per quarter dividend were subjected to corporate taxes before being paid, shareholders would get only 81.25 cents per quarter, on which they would then pay taxes of their own. (Assuming a 15% qualified dividend rate, this would leave a hair over 69 cents to reinvest after paying Uncle Sam, a mere 55% of the sum currently paid by American Capital Agency, which some shareholders reinvest before taxation due to their use of traditional, Roth, or SEP IRA accounts.) Because double-taxation has such an awful effect on after-tax returns, it is especially awful for small investors with small incomes whose tax rates are nowhere near the 35% corporate tax rate. How many small investors use American Capital to participate in deals in which they otherwise could only dream? How many IRAs are reinvesting double-digit returns from American Capital's managed REITs and would rather grow American Capital shares without any taxation than to reinvest American Capital dividends paid from after-tax profits?

Investors who need dividends rather than outstanding after-tax returns generally can obtain them from funds managed by American Capital. None need buy shares of American Capital itself. Income investing in American Capital Agency and American Capital Mortgage Investment is alive and well without competition from an American Capital which pays as dividends it could as readily reinvest without tax consequences. Especially since booking the rest of that $1.2 billion tax asset will require earning a lot more taxable income, it seems reinvestment is strongly favored over distribution for the long-term shareholder.

That said, American Capital enjoyed over a billion dollars in liquidity last year, and has made plain that dividends aren't waiting on tax status or the consumption of the loss carry-forward. While we can hope that American Capital finds outstanding deals in which to invest capital, we can be assured that if American Capital has funds it wants to return to shareholders it can afford to do so either as share buybacks (as Berkshire Hathaway is doing when its premium over book is slight) or as dividends (as Berkshire has assiduously avoided since its current management took over in the '60s).

Over 2011, American Capital changed its tax status to preserve an opportunity to make use of a $1.2 billion tax asset. To derive value from the asset, American Capital must earn much more taxable income. Currently, American Capital is only able to book just over a third of this tax asset as a net asset, because of the doubtfulness of making full use of it. Thus, American Capital now has a strong incentive to grow taxable income (so as to access more of the tax asset), whereas previously its incentive was to produce after-tax returns. Quarterly operating income for 4Q2011 was 24¢, a 20% upside surprise from estimates, excluding the impact of the tax asset. Some of that "surprise" resulted in one-time gains caused by non-accrual loans' movement back into the timely-paying category - a factor that should improve operating income going forward as portfolio companies' debts become easier to bear with the recovery of their business with that of the broader economy.

Annual operating income of 85 cents provided a 7.9% return on the company's asset value per share at the beginning of the year. This return was augmented by both unrealized appreciation and realizations on exit from investments. Because valuations depend on both portfolio-company EBITDA and broad-market factors such as the valuation multiples assigned by investors to publicly traded comparables, improvements in the economy and the broad markets should have a multiplied effect on American Capital's results even before the company's leverage and tax assets are considered.

In view of American Capital's apparent performance as a leveraged proxy for the broader market and the tax asset that will prevent it from paying income taxes, the company should be attractive to investors with an overall bullish view of the broad economy. The substantial discount of current share prices to net asset value offers a margin of safety, as the company's portfolio performance will be based on the invested assets represented by NAV rather than on the share price assigned from time to time by the market. Given the substantial and consistent liquidity enjoyed by American Capital - over a billion dollars in the last year alone - it is evident that American Capital has the ability to weather near-term concerns and will have no difficulty continuing to meet obligations to creditors.

Disclosure: I am long ACAS, AGNC, MTGE, BRK.B.