In corporate America, cash is king: Moody’s estimated in January that non-financial companies in the U.S. hold more than $1.5 trillion in cash. Amazingly, a large percentage of these funds not only come from one industry (tech), but from six companies: Apple (AAPL), Microsoft (MSFT), Google (GOOG), Cisco (CSCO), Oracle (ORCL)Â and Qualcomm (QCOM) account for more than one-quarter of the total cash held by non-financial corporations in the U.S. To give you an idea of what this means for these individual companies, here’s a look at their cash piles (and other cash-like assets) in comparison to their market capitalizations:
| Company | Cash / Equivalents | Market Cap | % of Cap |
| Apple | ~$150 billion | ~$555 billion | 27% |
| Microsoft | ~$100 billion | ~$340 billion | 29% |
| ~$60 billion | ~$375 billion | 16% | |
| Cisco | ~$50 billion | ~$125 billion | 40% |
| Oracle | ~$40 billion | ~$190 billion | 21% |
| Qualcomm | ~$35 billion | ~$135 billion | 26% |
| Total / Average | ~$435 billion | ~$1.72 trillion | 25% |
As an investor, this brings up an interesting question: How should we account for this cash? In my experience, this is addressed in two different ways: (1) avoided entirely, or (2) backed out to calculate an earnings multiple excluding cash on hand. Clearly the first approach is too pessimistic, even if you assume that much of the value of this excess cash will be destroyed through expensive acquisitions – but I’m not sure that the second approach does us much better.
From my perspective, the second approach suffers from two serious issues:
(1) Taxes: A recent article in the Sydney Morning Herald estimated that roughly $1 trillion of U.S. companies cash on hand – more than 60% of the total – is trapped outside the U.S. If companies plan on repatriating this cash, Uncle Sam will want his fair share; many are holding out for a tax holiday, though the prospects appear grim. Even if you think a repatriation tax holiday is near, you still need to account for the reduced tax rate (in 2004, the rate was 5.25%).
(2) Time Value of Money: In the case of Microsoft (which I’ll use as an example because it’s the company from above that I know the most about), cash flow from operations has averaged nearly $30 billion per annum over the past three years. Over that same period, the company has spent about $14 billion a year on dividends and repurchases (cumulatively), about $4 billion a year on acquisitions, and another approximately $3 billion a year on capital expenditures; said in fewer words, their current level of cash generation significantly exceeds their ability to deploy it intelligently. Looking at the company’s business holistically, as well as compared to the level of cash required to run the business historically, one could easily argue that $50 billion or more of cash on the books is not needed, and will not be used, for the foreseeable future (that’s different than a company like Berkshire Hathaway (BRK.A, Financial)(BRK.B), where investors can be confident that funds will be intelligently put to use at some point down the road – even if they don’t know exactly when that will be). From my perspective, there’s a cost to holding those funds: While Microsoft’s management team might not find a way to intelligently deploy that roughly $6 per share in excess cash (and from my perspective has little to no possibility of doing so), I think I could. For these companies, that’s a real point of consideration for the equity investor.
So now that I’ve rejected the two conventions from above, where should we go? I believe a logical approach, based on those arguments, would call for somewhere in between: penalizing companies for the expected taxes to be paid and the time value of money (with the second component clearly being pretty subjective), while simultaneously recognizing that having billions in excess cash is likely to be a good thing — even if we can’t pinpoint when that will be.
The second component will require some study of the company’s previous capital allocation decisions — from repurchase timing (as I’ve discussed with Deere, here) to the use of M&A (how much did the $11 billion spent on Autonomy prove to be worth to Hewlett-Packard (HPQ) investors?), as well as the impact of outsiders or insiders (new management and board members); in the case of Microsoft, I think that final point is particularly important at this point in time (I’m confident that ValueAct will push for action).
You might be thinking that I’ve weaseled my way out of an answer: in the case of Microsoft, I’ve left you somewhere between $0 and $100 billion — quite a bit of space (to say the least!) and a conclusion you’ve likely come to on your own. For one key reason, I think that’s just fine: Putting a specific number on the value of the excess cash is pretty futile in almost all situations. Even in these cases, where the cash on hand is a large percentage of the market capitalization, how important is nailing the “real” value of Microsoft’s cash? If you settle on $60 billion and I settle on $50 billion, does that really address the big questions in valuing the stock? If you’re an investor that buys with a margin of safety of 30% or more, the difference is peanuts.
But that gets to a bigger point: You’re doing yourself a disservice if you pigeonhole the concept of “intrinsic value” into a P/E ratio (or EV/EBITDA, etc.); investors take false comfort in a world of comparable (to a company’s peers and the market as a whole) and historical multiples.
The reality is that valuing a company — or the cash on its books — requires thorough and thoughtful analysis; it can’t be simplified into a formula or rule of thumb. Depending on your perspective, and your willingness to devote time and energy to analysis, I think that can be a real blessing — a way to differentiate from the herd with justified confidence.
Something as simple as counting the cash on the balance sheet can be a step in that direction.
