We always scour the universe of prospective investments for companies with durable business models, strong operating track records, and high quality brands. These criteria led us to investigate the shares of Kellogg Company (K). Not that we need to remind anyone, but Kellogg produces some of the most ubiquitous cereal brands in the world: Special K, Frosted Flakes, Kellogg's Raisin Bran, Froot Loops, Corn Pops, etc. The company also makes Pop Tarts, Famous Amos cookies, and many other treats.
Kellogg's sales have grown steadily, and the company has been consistently profitable for a long time. I'm sure Kellogg bulls would argue that a 16x P/E multiple is perfectly reasonable for a company of this quality, especially because the shares currently pay a 3% dividend. You won't do much better than 3% investing in high-rated bonds right now, and you get the upside of a terrific franchise when you buy K. However, there is one problem lurking in the fine print of Kellogg's filings that will force us to stay away from the shares.
First, a word about Kellogg Company's internal operating targets. The company says that its aim is to grow "internal net sales" at 1-3% per year, "internal operating profit" at 4-6% per year, and currency neutral net earnings per share at 7-9% per year. Obviously, a company can grow its bottom line at a much faster rate than its top line only for so long. There's only so much efficiency that management can wring out of the operating model. Too often, the investment world focuses narrowly on growth for growth's sake, ignoring the price that the company must pay for its growth. In the past, Kellogg has been a huge repurchaser of its stock. Though buybacks will increase earnings per share, they only increase the value of a stockholders stake if the shares are cheap at the time of the buyback. It's wrong to think of Kellogg as a rapid grower without considering the price it's paying for its growth.
Now, on to an even bigger issue. Kellogg's pension accounting is troubling. There's no way around it. The company assumes a long-term rate of return on its pension assets of 8.9%. Since it fell way short of this target in 2008, it contributed $400M to its plan. Investors have to adjust earnings for these sorts of contributions, and they have to take into account the potential for further shortfalls in the future.
Let's consider how likely it is that Kellogg actually returns 8.9% per annum on its plan assets going forward. Thankfully, the company details the "logic" behind the 8.9% return assumption. It says that it assumes a long-term inflation rate of 2.5% and it gives itself the benefit of the doubt by forecasting a 1% active management premium. Though Kellogg claims that this premium is "validated by historical analysis," please pardon our extreme skepticism. Though 2.5% isn't an outlandish assumption for inflation, it is higher than most policymakers would desire. If we assume 3% real GDP growth, 2% inflation, and a 2.5% dividend yield, we get 7.5% return for stocks in the long run. We regard this as a reasonable (perhaps slightly aggressive) estimate, but even if it's a little conservative, Kellogg's pension return is likely to be far, far below its 8.9% estimate because its plan assets included fixed income holdings, the returns on which should be far below 7.5%.
Kellogg says that a 1% reduction in the assumed rate of return would increase annual benefits expense by about $40M. Since we think that the company's assumed rate of return is 2-3% too high, its reported net income is probably off by $80-120M (roughly 20-30 cents per share). Now, we don't claim that our estimates are precise, but pension accounting is inherently imprecise. That said, no one stands to gain anything if the accounting is too aggressive. It would be far better if Kellogg (and the hundreds of other companies who make unrealistic return assumptions for plan assets) would employ conservative accounting. That way, investors would be better able to ascertain worst-case outcomes.
Kellogg has a phenomenal stable of brands, but its pension accounting should be of concern to bulls. If you adjust its figures as we've done above, its P/E multiple is closer to 17-17.5x. Clearly, the shares aren't cheap right now. Currently, we're investigating General Mills (GIS) to see if the same forces are at work. We should have something to say about that soon.