Jeff Davis, CFA, is a veteran bank analyst. The views and opinions expressed in this piece are those of the author and do not necessarily represent the views of S&P Global Market Intelligence or Mercer Capital, where he is the managing director of the financial institutions group.
David Einhorn of Greenlight Capital Inc. is no stranger to controversy. Over the years he has had a knack for shining a light on an asset and raising a valuation question. Allied Capital Corp. and Lehman Brothers Holdings Inc. are two notable examples among financials. He also will wade into a capital structure issue as was the case in 2013 when he sued Apple Inc. to issue a dividend paying preferred stock. The preferred was never issued, but the company has stepped up repurchases such that nearly $200 billion of cash has been returned to shareholders since 2013 through the first quarter of 2017 compared to $2.5 billion in 2012.
His current project is General Motors Corp. He put a flashlight on its common shares on March 28 arguing that they are unreasonably cheap. Immediately before the proposal was made GM's shares were trading just below $35 per share, which equates to 5.8x 2016 earnings of $6.00 per share and 5.6x the midpoint of management's 2017 guidance ($6.25 per share). The dividend yield is high at 4.4%, more than double the yield of the S&P 500. As Einhorn points out, the yield is not high because the payout ratio is high; the $1.52 per share dividend equates to just one quarter of (current) earnings.
Stated more directly, GM is a cheap stock. Investors know that. But valuation as a stand-alone proposition rarely is a catalyst and can often signal brewing issues when the shares are cheap.
His proposal is to split GM's shares into two: (a) a dividend stock (GMD) that is entitled to the current $1.52 per share dividend provided it is earned; and (b) a capital appreciation stock (GM) that is entitled to earnings in excess of $1.52 per share. Income investors would gravitate to GMD, while capital appreciation investors would own GM. In a sense, I guess GMD would represent a non-cumulative preferred, but in the form of common that is issued via a tax-free distribution. Presumably it would be pari passu to GM in liquidation, though not in voting rights as he has proposed it.
Einhorn's math implies a total value of $43.16 to $59.55 per share based upon a dividend yield of 7% to 9% applied to GMD and a P/E of 5.6x to 8.0x applied to GM. The plan is elegant in its simplicity. If the range is realized shareholders should be grateful. The post-bankruptcy GM priced its IPO at $33.00 per share in 2010. The economy has been growing moderately since, while car and light truck sales have trended much higher since then to a current annual rate in the vicinity of 17 million units. GM's EPS have increased, but the shares have remained range bound.
There are a few takeaways for banks and investors. An obvious one as always is that purchase price matters a lot when measuring returns, at least over short-to-intermediate holding periods. It is the one variable investors exercise complete control over. Investors overpaid in the IPO. Had the IPO priced at $23 per share it might have been deemed a failure for the government as a selling shareholder, yet the total return since then with dividends might not be so bad.
As it relates to Einhorn's proposal to unlock value the board would do so if it were easy and without much risk. Not surprisingly GM's post-bankruptcy board dismissed the proposal even though Einhorn does not call for a stepped-up pace of capital return to render the shares less cheap. Clever if not compelling is not something such a board would be expected to embrace.
Unlike GM's shares, I think few would call bank stocks cheap today even though the market likely is correct that profitability will improve for reasons other than declining credit costs following the national election last year. The better Einhorn analogy for banks is what to do about excess capital that is building, even within the context of enhanced capital requirements. Excess capital for a bank is worth at most dollar-for-dollar, and that arguably is from the perspective of a controlling shareholder that can extract it subject to regulatory approval.
If a bank is overcapitalized, better to distribute it than to loosen lending standards to drive loan volumes or overpay for acquisitions (at least with cash rather than with over-priced common shares). Buybacks seem to be the preferred method for returning excess capital in the industry, but are not buybacks at current high prices unattractive even if there is an incremental benefit to the EPS gods? What is the point of paying approximately 20x last-12-months earnings that have little in the way of current credit costs? As an alternative I see nothing wrong with periodic special dividends. It is a dollar-for-dollar proposition. Similar to Einhorn's proposition special dividends force investors to focus on cash distributions as a source of value.
The bigger issue I see is not forcing investors to fully value the dividend so much as what the shares say about the capital appreciation security that Einhorn has proposed. The P/E is low and the dividend yield is high because investors believe earnings have peaked or will do so soon. The industry's earnings are highly cyclical — perhaps more so than the banking industry's earnings cycle. Auto and light truck sales in the U.S. are running in excess of 17 million compared to around 10 million in 2009 when the industry was on the floor. GM's board knows this; hence a reluctance to take chances even though the economy appears set to continue on a moderate growth pace. Maybe that is why Einhorn is not pushing a greater return of capital as he did with Apple in 2013.
Is the U.S. economy likely to slide into a recession later this year? I doubt it. Bank stocks are not signaling that; nor are the high yield and levered loan markets in light of spread tightening that has occurred over the past year. But the credit cycle appears to be at work with GM. After loosening standards for years (think seven-year loans) and increasing subprime volume, lenders apparently have decided to stop digging a deeper loss hole. Capital availability is tightening for iffy auto loans; yet, much of our consumer-driven economy has been dependent upon finding and financing the incremental buyer to keep the cycle going. Last decade it was home mortgages; this decade autos.
I doubt an intense down cycle for auto loans will create big issues for the broader economy. Auto loans are not as infused in American society like mortgages, balances are relatively small, and the duration of the loans vis-à-vis a mortgage is much shorter. Still, the market is signaling coming pain for auto companies as fewer cars can be financed and lenders will see losses climb. Santander Consumer USA Holdings Inc., a subprime auto lender, trades at comparable P/E to GM based upon 2017 consensus earnings of 6.5x as of March 31. The cycle should play itself out like most. When the issue makes its way to the front of the USA Today it has peaked.
Coming auto losses notwithstanding, I doubt we will see traditional commercial bank lenders that are sizable players in the auto market struggle; rather, the units will incur higher loan loss provisioning for a couple of years with perhaps a couple of quarters that cause investors to grimace.
The cycle does point to a larger point about banks, however. Credit is cyclical. Borrowers and lenders always have and always will have a propensity to overindulge. For investors there is a lot to be said for looking at earning power—a through the cycle measure of average profitability that has little to do with current and next year’s expected earnings. Applying the concept to GM may mean its shares are trading for 9x-10x normalized earnings, which seems cheap to me but not as cheap as 5.6x 2017 earnings guidance. I suspect Einhorn has a view of that, though I do not think he has shared it.