Posted by: Simon Lack | February 7, 2013

Unlocking Value In Microsoft The Einhorn Way

This morning David Einhorn on CNBC elegantly highlighted the inconsistency in pricing of different asset classes, specifically in the case of Apple (AAPL), which he owns. Einhorn believes AAPL should exploit the low yields/high prices of preferred equity to the advantage of stockholders.

You can read the transcript of Einhorn’s interview here so we won’t repeat what he said. We don’t own AAPL, but his suggestion isn’t limited to them. In fact, any company with a strong balance sheet, predictable earnings and a low P/E could avail itself of the same strategy.

So we’ve applied it to Microsoft (MSFT), another tech giant that we believe is attractively priced. The basic concept is for the company to distribute preferred equity shares to existing shareholders. No cash would change hands; each shareholder would just receive preferred shares proportionate to their shareholdings. But even though no cash has changed hands, we believe it illustrates how value would be unlocked.

Here’s how it works (my colleague Henry Hoffman crunched the numbers for this example): MSFT has 8.4 billion shares outstanding for a market cap of $228 BN. It has $43BN of cash net of debt (further assuming a 20% haircut to repatriate the cash from outside the U.S.). So the whole company is priced at $185BN ($228BN-$43BN). MSFT pays a quarterly dividend of $0.23, which costs $7.8BN annually. Suppose the company diverted this dividend fully to pay dividends on the preferred shares, which are all owned by existing stockholders. Assuming a 4% yield on these preferreds would value them at $195BN (7.8 divided by 0.04). So the shareholders would own new securities of this approximate value which they could keep or sell in the marketplace. But they’d still own the common equity. Consensus expectations are for MSFT to earn $24BN in the fiscal year 2013 (ending in June). MSFT’s P/E is 7.7X (excluding Cash).

Diverting $7.8BN to pay preferred dividends reduces the net income available to common equity to $16.2BN. Assuming the same P/E multiple on these reduced earnings would value the equity at $125BN. Combined with the preferred equity, the shareholders would now own securities worth $320BN. On this basis each common share would be worth $37.86 on the day MSFT announced its intention to distribute preferred equity securities to every holder of common equity.

It wouldn’t alter the leverage of the company because the preferred would count as equity and would sit just above common in the capital structure. It also wouldn’t alter the company’s dividend expense, since the dividends that were being paid on the common have simply been diverted to the preferreds. It would highlight the large disparity between the valuation of common equity, with an earnings yield of almost 13% (inverse of the 7.7 P/E ratio) and the 4% yield on a security that sits immediately above common equity in the capital structure.

The P/E on the common equity might fall following the distribution of the preferred equity, but even in the absurd case of the common equity being worthless (hard to imagine given $16.2BN of FY 2013 earnings) the preferred equity’s value is still higher than today’s market cap (ex-cash).

You can play around with the assumptions endlessly. One we like is to further assume that money spent on stock buybacks is additionally diverted to preferred dividends. Buybacks count as cash returned to shareholders in the same way as dividends. MSFT’s five year average buyback expenditure is $9.9BN. Adding this to dividends of $7.8BN creates a preferred dividend of $17.7BN and a value on this class of securities of $442BN. Net income left to common shareholders in this case would be$6.3BN in FY 2013.

David Einhorn made a clever suggestion. It’s a neat way to illustrate the discrepancy in asset markets between fixed income yields and earnings yields.

We hope Steve Ballmer was watching CNBC this morning.

Disclosure: We are long MSFT


  1. Dear Simon,

    Financial engineering to enhance shareholder value is always an interesting exercise. I caution, however, that there is many a slip twixt the cup and the lip. I have a number of thoughts… in no particular order: the lynchpin to value creation in this example is the yield on the preferred which I suggest would be much wider than 4% ; I thought I saw Einhorn quoted at 6.25% for his proposal. Given the large portion of the capital structure to be taken up by the preferred, the relatively weak 3X dividend coverage and no prohibition against creation of more senior claims, the MSFT preferred might trade like junk. And once the preferred is issued, the now more junior common will absolutely trade at an even high earnings yield that before. In any event the preferred stock idea should be measured against a more traditional use of debt leverage to buy back equity. Off the top of my head there’s no apparent difference between the two other than (1) the after tax annual financing of the preferred will be far more costly, though (2) the preferred entails no refinancing risk; if one is debt financing for 20 or 30 years, however, the latter consideration is neglible.

    I worked as a consultant on this problem with Salomon Brothers Corporate Strategies Group for many corporate issuers in the 1990’s. What we found (at that time at least) was that the cost of capital was minimized (and shareholder value maximized) by targeting capital structure and interest coverage parameters in the weak single A to mid triple B credit range (generally indicating 5 to 6 times pretax coverage of interest). Given these parameters and using the net cash (after repatriation haircut), a debt funded equity buy back approaching $200 billion seems possible.

    Clearly APPL, MSFT and many others have long ignored development of any financial policy and strategy, and any movement to light a fire under the managements is probably a good thing. The companies have truly massive untapped credit capacity for which their shareholders are paying a price. Financial engineering would add somesignificant value,


  2. Dear Simon,

    The fundamental theorem of corporate finance (capital structure irrelevance, not having been repealed, we are right to view attempts to create value from nothing with a skeptical eye. Dividing common investors’ claims into a new partition — part common, part preferred — certainly does nothing to change asset risk or tax shield. As such we’re faced with a choice: either

    a) M&M are wrong
    b) there is some mistake in the analysis

    Roger hits the nail on the head — the proposal ignores risk (specifically, the conservation of risk, even when it’s transferred from one bearer to another). A proper accounting will cause the specious arbitrage to evaporate.

  3. Roger,

    re: “What we found (at that time at least) was that the cost of capital was minimized (and shareholder value maximized) by targeting capital structure and interest coverage parameters in the weak single A to mid triple B credit range”.

    This too violates M&M, and we find again that the resolution of the paradox isn’t that M&M is wrong, but that the analysis was incorrect. Specifically, these attempts to minimize WACC almost universally rely on an assumption that debt beta is zero, which leads to an incorrect calculation of the cost of debt and the levered cost of equity. A correction is posted here: .

  4. At Salomon I was the credit market guy who explained to corporations what I thought they could get away with, but I had Paul Pilorz, Eric Lindenberg and Marty Leibowitz , all finance Phds, running interference on the theoretical considerations; Marty, of course, is probably the most published finance academic in the world. M&M makes enormous assumptions and, as a result it’s of limited value. Markets are made by (often irrational) people, not mathematical formulas and, as result there are at times significant inefficiencies to be found. This in one of those times. For whatever reasons (Fed action, investor fear, unrepatriated earnings looking for a home, etc) real interest rates are very low and equity risk premiums are very high. As I’ve noted to Simon, equities seem to be valued as if long Treasuries are trading at 6%, a level they may well attain over the next three to five years. As a result there appears to be an opportunity for conservatively financed companies to enhance shareholder value. Let’s use MSFT as a real life example using today’s market parameters with one important caveat – their business will experience at least modest long-term organic growth (otherwise I’d advise them to use their equity as currency to make acquisitions). I’ll use short hand – investment grade credit spreads average 140 basis points over Treasuries. We’ll finance at plus 200 and constrain the leverage at roughly 2X cash flow. This allows for $66 billion of financing in half a dozen tranches from 5’s to 20’s. Add $40 billion of cash (after repatriation haircut), and I’d advise management to tender for three billion shares at $35. If we get this far, we’re left with 5.4 billion shares outstanding earning around $4.25 each. Less than 4% of the Value Line 1700 trade at a PE of less than 8.0 and most of those companies have a lot of hair on them. I think this stock trades at at least $35 and maybe even at a higher multiple than before the financial engineering. In this case the financial engineering itself signals important information to the market that management has confidence in it’s business (an element that seems to be lacking in the current valuation).

  5. It’s funny Roger brings up the point about inefficient markets – as that is exactly what goes through my mind when I see (very valid) discussions like this. Financial engineering in reasonably efficient markets should not create investor value, yet here we sit with ostensibly 2 examples (AAPL and MSFT) where just that sort of financial engineering should result in increased equity value to shareholders.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: