The Motley Fool's Rich Smith recently wrote in praise of share buybacks at Apple (AAPL) while disparaging buybacks at American Capital (ACAS). This article seeks to describe why Mr. Smith has it exactly backward. The share buybacks announced at Apple won't amount to a hill of beans (compare the amount involved over the life of the program to the size of the market cap; it will do little beyond countering dilution), but at American Capital they will materially move the needle on long-term shareholder returns.
The Thesis Against American Capital's Buybacks
Mr. Smith cites Jim Cramer's criticism of pre-crash share buybacks at big banks as such compelling evidence that share buybacks are a bad deal that he is willing to dismiss another authority cited in the same article - a Boston University professor of finance - for the proposition that share buybacks raise stock prices. Nothing in Mr. Smith's article seems to delve into the value proposition of any of the shares discussed in the article. The surface treatment given the topic includes:
- comparing the average 1Q2012 purchase price ($8.79) to American Capital's market price
- concluding that share price moves prove some thesis about management's purchase strategy
- suggesting that management's purpose is to prop share price rather than to increase relevant per-share metrics
Comparing the price of share buybacks to the market price might make sense to a trader (who is necessarily obsessed with current market prices), but is nonsense to investors. Mispricing in the market isn't proof of anything to an investor, but instead represents a buying opportunity. Mr. Smith needs to reread his Warren Buffett. Buffett himself announced share buybacks at Berkshire Hathaway (BRK.B), which like American Capital hopes to capitalize on the market's view of share value to lower the share count at prices below enterprise value. Neither Berkshire nor American Capital have announced any intention of trying to put a "floor" under shares - indeed, a price collapse would offer management at either company an even greater opportunity to increase value for continuing shareholders by retiring a greater number of shares with funds allocated to the buyback.
The Math At American Capital
With shares trading at a substantial discount to their net asset value per share ("NAV"), the opportunity existed in 1Q2012 to retire shares with a last-published NAV exceeding $13 at prices less than $9. The fact that the average purchase price in 1Q2012 was 4% below the market price the day Mr. Smith wrote his article is about as important as the fact that the shares could be sold today for prices 12% above the 1Q2011 average purchase price. So, what? It's not as if American Capital will realize a gain or loss issuing or retiring its own shares. Who would think this way but a trader?
The objective of share retirement at a value-focused enterprise is the concentration of assets into fewer shares. As explained by Warren Buffett in the Berkshire Hathaway 2011 Letter to Shareholders, this only works when the purchase price is lower than the intrinsic value of the company per share. At American Capital, this value can be approximated by the shares' NAV.
To understand NAV, it's necessary first to understand how American Capital is required to value its assets (and the accuracy of this valuation), then to consider the impact of American Capital's share count on the value obtained by adding its portfolio companies' value and subtracting the debt by which its portfolio is leveraged. First, a chart:
Since the book value is total assets minus liabilities, the graph from mid-2009 through the end of 2011 depicts a company steeply raising book value while sharply cutting leverage. Note that the company isn't shrinking: this graph does not depict per-share metrics, but an increasing book value for the firm as an entity. According to new numbers just released this afternoon, American Capital at the end of 1Q2012 had total assets of $6.4 billion, total liabilities of $1.3 billion, and a net asset value for the firm of $5.1 billion. Overall, management is making the company more and more valuable.
Yet, company-wide metrics aren't the metrics that move an individual shareholder's investment. Per share metrics are the order of the day. And with share buybacks decreasing the number of shares into which American Capital's results must be divided, the per-share metrics look even better. Over the period charted above, American Capital's per-share metrics looked like this:
The 1Q2012 performance derided by Mr. Smith - and not depicted in the above chart due to the newness of the 1Q2012 numbers - was rather enhanced by the share buybacks criticized as a waste of shareholders' money. In particular, the repurchase of 5.5 million shares for $48 million resulted in an increase in NAV per share of 12¢ entirely aside from the effects of management's performance at the level of the firm. This gain in NAV - equal to about 1.4% of the per-share price on the dates management bought shares - reflects a concentration of corporate assets into fewer shares, enabling continuing shareholders even better results in the future.
And what results they have been. Rebounding from the crash, management has grown NAV to $15.71 per share at the end of 1Q2012 - about 80% over the $8.76 shareholders had invested on their behalf at the end of the second quarter in 2009. The increases in American Capital's investment values have been concentrated into fewer shares just as they have been offset by less and less debt per share.
Why American Capital Is A Buy For American Capital
With the market price per share south of $10 and the NAV per share solidly north of $15, it's evident management can raise NAV per share by retiring shares at a huge discount - more than a third of NAV. American Capital continues to make new investments ($83m in 1Q2012, up from $31m in the year-ago quarter), retire debt ($119m repaid in 1Q2012), and hold cash ($518m in cash and cash equivalents at the end of 1Q2012, including $313m in unrestricted cash and cash equivalents). Operating income remains positive, allowing American Capital to benefit from a tax asset recorded last year. (Sale of depreciated investments led to a realized net loss for the quarter, though earnings were $1.71 per share over the quarter.) Despite retiring shares, the overall assets of the company increased over the quarter. Things look good at American Capital.
The suggestion that American Capital "cannot afford" a dividend lacks any cited authority, and seems to be without any basis except in the inclination of critics desiring to insult management, or possibly the company's investors. In truth, American Capital has a dividend policy, which the Board extended on April 26 of this year to be effective through the end of 2013. The dividend policy calls for the Board to set an amount management may utilize for dividends and share buybacks, which management is to use for share buybacks while the NAV discount continues to make buybacks the better bargain for shareholders.
American Capital continues to offer investors entry into a private equity market segment in which retail investors have little opportunity to participate - and at a significant discount. The shares are so hated by bewildered onlookers that the market's disparagement of their value is easy to understand. However, the steady increases in assets per share tell the tale of the tape and indicate the long-term direction of American Capital shares. As Warren Buffett appears fond of quoting from Benjamin Graham about the market, "In the short run it's a voting machine, but in the long run it's a weighing machine."