One thing I wanted to do this afternoon is talk about the rare class of company—ExxonMobil, Wal-Mart, and IBM come to mind—that has no choice but to repurchase shares of the company’s stock. If you read much investment commentary on the company, you will walk away with the impression that IBM has no plans for growth or moat-building and is surviving slowly on the use of existing cash flow to retire stock to create an illusion of progress.
To understand why IBM makes the capital allocation decisions that it does, I find it wise to keep IBM’s size in mind. IBM is one of those two or three dozen firms in the world that when you buy an ownership stake, you are really buying into something that is the size of a small country. For instance, the company is going to generate about $97 billion in revenue or so this year, conditional upon the next quarter’s report. As a point of reference, the entire country of Libya produces $95 billion worth of goods and services in a year. The entire country of Morocco produces $95 billion worth of goods and services each year. The entire country of Puerto Rico produces $105 billion worth of goods and services per year. So when we talk about IBM, we’re talking about something that has an output roughly the size of Libya, Morocco, or Puerto Rico.
Pretend you’re in charge of running IBM. You want to continue the company’s legacy of doing something, and on the investment side, you eyeball that the historical return of S&P 500 stocks is around 10% per year. You acknowledge that IBM is a behemoth in size, but you set the target of delivering 11% annual returns to shareholders—you want people who part with their hard-earned money to show you the ultimate act of trust by tying their economic fate to your stewardship to be rewarded at a rate above what you’d get with a plain S&P 500 Index Fund.
You note that, although the current dividend yield is 2.7%, this is unusually high, and most starting investors usually receive an annual payout of 1.5% when the stock is fairly valued. This means that, assuming the stock is fairly valued, you must grow profits per share by 9.5% over the long haul to get those 11% returns.
Let’s say that one option is to try and reinvest all of the cash flow that does not go towards the dividend, trying to achieve 9.5% annual revenue growth that translates into 9.5% annual profit growth (this would be the product of big spending and sales intoxication in which you take “must get bigger” to be your primary investment imperative). To grow revenues 9.5% in one year, you’d have to produce $9.21 billion in fresh goods and services over the next twelve months while also retaining the $97 billion that you’re already generating. In other words, you’d have to add one-ninth of the Moroccan economy to your production in a given year. The Bahamas has the GDP of $8.3 billion. To achieve 9.5% revenue growth, you would have to add more than the annual production of every man and woman in the Bahamas economy to achieve that growth. Achieving that kind of revenue growth in one year would be extremely fortuitous, doing it every year: impossible.
A stock repurchase program, on the other hand, guarantees certain results. Sometimes, this can be a problem. If you have money to allocate, and you think there’s a 50% chance you can get 8% growth, why go through the hassle and uncertainty of building new factories, hiring new workers, and rolling out new products if you have a guaranteed path to 6% earnings per share growth via stock buybacks?
As a broad economic principle, stock repurchases can discourage management teams from intelligent risk-taking that can grow an economy through new products and services, as well as give people new jobs. Why take that gamble when you can just buy back your own stock? As a matter of social and economic impact, stock repurchases disproportionately benefit the investor class, and only benefit the broader economy to the extent you take the dividends or capital gains caused by the increase in profits per share and actually consume things.
But, when a stock is quite undervalued, few things can deliver returns quite like a share buyback. IBM earns around $16 per share in profits, and trades at $160. That P/E ratio is 10. This is ripe for wealth creation. IBM buys back and retires about $3 billion worth of stock per quarter, though it does vary. Imagine if IBM continues its current practice and the price of the stock stays low at $160. IBM is a $160 billion company. Even if profits stayed the same, the repurchases alone would increase your share of the profits by 7.5% over the coming year. Without factoring in growth, the current 2.7% dividend yield + 7.5% buyback yield (which would go up as the price of IBM stock goes down, and would go down as the price of IBM stock goes up). That’s a 10.2% annual return, assuming the continuance of present conditions without factoring in actual business growth. That’s why IBM is stuffed in the portfolios of famous value investors across the country. You don’t need a whole lot to go right for IBM to deliver satisfactory returns over the coming years five to ten years.
Now, you might be thinking—hey, it doesn’t have to be either extreme between capital investment and buybacks. You could invest more towards organic growth, and make it a balanced attack with buybacks. And plus, it would be evidence of an inefficient operation for a company to need 9.5% revenue growth to achieve 9.5% profit per share growth. Consumer staples grow revenues by 4-5% annually all the time, and easily turn it into 7-9% profit gains.
There are two responses to that:
First, I don’t think people truly realize how cheap IBM has gotten. During the worst of 2009, IBM traded at 10.9x earnings. Now, it’s at 10x earnings. This is the time to strike via buybacks, because the valuation of the stock allows the management team to greatly increase earnings per share when the stock exhibits price weakness. Just look at what happened during 2009 for evidence: IBM grew its earnings per share from $8.93 in 2008 to $10.01 in 2009, and that is because the company was buying back stock at 10-11x earnings. It may not seem like it now, but this price decline in IBM is where the earnings per share growth gets turbo-charged, and the results will be obvious in the coming years (it just doesn’t seem so obvious to many people right now).
And second, I don’t think people realize how much IBM is dedicating to future investment. The buyback narrative has taken such a hold that investors don’t seem to examine IBM’s research-and-development budget anymore. They spent $1.35 billion on research and development in the past ninety days. They spent $1.45 billion on R&D during the ninety days before that. And the company spent $1.50 billion during the first quarter of this year. And last year, they spent over $6 billion in R&D. Their budget is exceptionally strong, and they continue to pour money towards organic growth. The problem has really been the switch from 30% operating margin businesses to 20% operating margin businesses as a result of large businesses switching from hardware to the cloud (though the cloud presents security risks that will reveal themselves in the coming years).
IBM is currently repurchasing stock at 10x earnings. As the price of the stock has declined, the margin of safety for prospective owners has increased. And when IBM repurchases stock at lower prices, your margin of safety continues to increase. I would be disappointed if IBM announced any kind of slowdown in their buyback program right now, because repurchases at 10x profits are what make investors rich. Given Buffett’s detailed explanation of his preference for lower IBM prices to buy back the stock in his initial letter after the IBM purchase, I imagine that’s what he has counseled CEO Rometty after the recent earnings report.