The Treadmill’s Real-World Effects  73 The expectations treadmill explains the mismatch between TSR and the underlying value created by the two companies. Using the ratio of enterprise value (EV) to net operating profit after taxes (NOPAT) as a proxy for market expectations, J&J’s EV/NOPAT started the period at 23 times, while Tyson started at 13 times. This means that J&J’s treadmill was already running fast, with high expectations already built into the share price. Tyson’s EV/NOPAT was below average, reflecting modest performance expectations. The EV/ NOPAT for both companies increased during the period—J&J from 23 times to 29 times, and Tyson from 13 times to 17 times. Another source of the difference in TSR was changes in ROIC, driven pri- marily by changes in margins. Tyson’s adjusted EBITA/revenues increased from 4 percent to 9 percent, while J&J’s remained flat at about 12 percent. Similarly, Tyson’s ROIC (excluding goodwill) increased from 12 percent to 22 percent, while J&J’s declined from 25 percent to 21 percent.3 Which company did a better job? You can make arguments for either one: Tyson succeeded in outperforming its expectations, and J&J Snack Foods suc- ceeded in delivering against high expectations. TSR might have been a fair mea- sure of the performance of Tyson’s managers, but it would not have reflected what a great job the J&J team did. For TSR to provide deeper insight into a company’s true performance, we need a finer-grained look inside this measure. EXHIBIT 5.1  Tyson Foods vs. J&J Snack Foods: Growth, Return on Invested Capital (ROIC), and Total Shareholder Returns (TSR) Dec 2014–Dec 2017, % 3 6 19 24 27 14 Tyson J&J Snack Foods Revenue Growth Average ROIC Annualized TSR 3 J&J Snack Foods’ ROIC declined while its EBITA margin went up because it used more capital (work- ing capital and net property, plant, and equipment) to generate each dollar of revenues in 2017 versus 2013. 74  The Alchemy of Stock Market Performance Decomposing TSR We recommend analyzing TSR by decomposing it and quantifying its compo- nents in the manner outlined in this section. The effort serves two purposes. First, when managers, boards of directors, and investors understand the sources of TSR, they are better able to evaluate management. For example, it’s important to know that J&J’s TSR, though lower than Tyson’s, reflects strong underlying performance against high expectations. Second, decomposing TSR can help with setting future targets. For example, it may be challenging for Ty- son’s managers to repeat their high TSR, because that would probably require raising profit margins and earnings multiples much higher. The traditional approach to analyzing total shareholder returns is math- ematically correct, but it does not link TSR to the true underlying sources of value creation. The decomposition we recommend gives managers a clearer understanding of the elements of TSR they can change, those that are beyond their control, and the speed at which their particular expectations treadmill is running. This information helps managers focus on creating lasting value and communicate to investors and other stakeholders how their plans are likely to affect TSR in the short and long terms. Decomposition of TSR begins with its definition as the percent change in a company’s market value plus its dividend yield (for simplicity, we assume the company has no debt and distributes all its excess cash flow as dividends each year): TSR = Percent Change in Market Value + Dividend Yield The change in market value is the change in net income plus the change in a company’s price-to-earnings ratio (P/E).4 Adding the dividend yield gives the following equation for TSR: TSR = Percent Change in Net Income + Percent Change in P/E + Dividend Yield This equation expresses what we refer to as the “traditional” approach to analyzing TSR. While technically correct, however, this expression of TSR misses some important factors. For example, a manager might assume that all forms of net-income growth create an equal amount of value. Yet we know from Chapter 3 that different sources of earnings growth may create differ- ent amounts of value, because they are associated with different returns on capital and therefore generate different cash flows. For example, growth from acquisitions may reduce future dividend growth because of the large invest- ments required. 4 Technically, there is an additional cross-term, which reflects the interaction of the share price change and the P/E change, but it is generally small, so we ignore it here.