Almost two years ago, I bought shares in IBM (IBM), incidentally around the time Warren Buffett announced he had made a large investment in the shares himself. The company looked cheap and had an outstanding cash generation record. That is one of the main things I look for in an investment - free cash flow. If you take a look at the financial results of IBM over the last 10 years below, you can see the appeal.
Over the last 10 years, IBM's free cash flow has equaled 116% of reported net profits. It isn't uncommon for cash flow to exceed profits in the IT industry, as some revenue is deferred as it relates to services that will be supplied in a later period (or as a cynic would say - to defer paying tax on the earnings!).
IBM has a shareholder friendly capital allocation history as well, with 78% of free cash being used to fund share buybacks, and a further 20% on dividends. But it has also been trimming its balance sheet. A further 23% of cash flow has been spent on acquisitions and the astute will notice this adds to more than 100%. That's because it has been selling assets from the balance sheet. That isn't a bad thing, far from it, in fact it improves the return on investment, which has averaged 38% over the last 5 years.
But not long ago, I decided to sell my IBM stake, for the same price I paid for it two years ago. Why you might ask, well the answer is simple - Oracle (ORCL). Take a look at its 10 year financials below.
IBM - Oracle comparison
When I reviewed this, IBM's free cash flow generation of 116% of net profit suddenly didn't look so stellar compared to Oracles of 129%. And most notable of all is that IBM's revenues are declining, and earnings declines have been partially offset with cost reductions but there is only so far this can go. Oracle on the other hand is increasing revenues and earnings, and although they are forecasting to have a fall in revenues next quarter, this is a due to a very high comparison Q2 last year (which saw 18% revenue growth in one segment).
Now it should be noted that Oracle spends more on acquisitions than IBM, but if I account for that in a Discounted Cash Flow (DCF) valuation, Oracle comes out as 27% undervalued, whereas IBM could be overvalued unless it starts to grow revenues again.
Finally, compare the current prices of both companies, IBM trades at a P/E ratio of 12.9, compared to Oracle at 12.7 (cash adjusted). But Oracle is growing and generates more free cash flow, compared to IBM which is seeing declining revenues and earnings. My DCF model shows IBM is fairly valued even if we assume earnings will grow at 8% for the next few years, and is overvalued if it doesn't grow yet continues to spend money on acquisitions. Conversely, Oracle appears undervalued, even if we take account of the $8bn tax bill it would face if it repatriated its offshore cash.
After looking at this comparison I couldn't defend holding IBM any longer, so I sold my whole position at a roughly break-even price (very disappointing over 2 years) and bought Oracle shares instead. I like to have a small part of the portfolio in a big defensive company that is a long term buy and old, and Oracle is simply the better choice.