Testing the Value Based on Multiples of Peers  407 The overall average NOPAT multiple across the entire peer group is 18.0 times, which would suggest a significantly higher value than the DCF esti- mate (which has an implied NOPAT multiple of 16.0). But the peers in this group appear to be clustered in two groups with very different underlying re- turns and growth rates, making the overall average less meaningful. There is a group of leading players with outstanding returns and growth rates that are valued in the stock market at an average of 21.0 times NOPAT. Based on the multiple for this top peer group, ConsumerCo’s branded-products business would be valued at $6,883 million, which would be a clear overestimation, given its actual performance and growth (see Exhibit 19.10). At best, it could represent what ConsumerCo’s business would be worth if it were able to at- tain the economics of these leading players in the sector. In contrast, the play- ers in the peer group with returns and growth rates closer to ConsumerCo’s business have an average multiple of 15.6 times NOPAT, leading to a value estimate of $5,060 million, which is much closer to the DCF results. Adopting the same approach of using close-peer multiples to value all of ConsumerCo’s other segments, including ConsumerCo finance and the cosmetics joint venture, the estimated equity value is $8,774 million (Exhibit 19.10). Note that by using top-peer multiples for the valuation, Consumer- Co’s value would be estimated some 30 percent higher than its DCF value, at $11,956 million. Showing the range of value estimates for close-peer and top-peer multiples helps to triangulate the DCF valuation results. In our experience, close-peer multiples typically lead to valuation results within EXHIBIT 19.10  ConsumerCo: Valuation with Multiples, January 2020 EV/NOPAT Multiples-based value Business NOPAT, $ million Close peers Top peers Close peers, $ million Delta to DCF, % Top peers, $ million Delta to DCF, % DCF value, $ million Branded products 325 15.6 21.0 5,060 -2 6,833 32 5,188 Private label 93 11.7 16.0 1,084 -4 1,482 31 1,128 Devices 102 14.0 19.5 1,422 -4 1,980 34 1,474 Organic products 134 24.5 26.5 3,285 -5 3,553 3 3,440 Corporate center (54) (1,123) (1,123) (1,123) Eliminations (2) – – – – – Total operations 597 9,727 -4 12,726 26 10,107 Customer finance 121 12.01 12.01 149 0 149 0 1502 Cosmetics joint venture 81 17.0 22.0 589 -3 772 27 6093 Excess cash 250 250 250 Gross enterprise value 10,716 -4 13,897 25 11,117 Debt (1,941) (1,941) (1,941) Equity value 8,774 -4 11,956 30 9,175 1 For customer finance, P/E and net income are shown. 2 At equity value, net of debt in customer finance. 3 At equity value of minority stake in cosmetics joint venture. 408  Valuation by Parts 10 to 15 percent of the DCF outcomes—in other words, within the normal margin of error for any valuation. However, many analysts and other practitioners often base their valua- tions on top-peer multiples. The valuation by parts then easily leads to a con- clusion that a company suffers from a so-called conglomerate discount and a recommendation that it should be broken up into parts to unlock the valua- tion gap versus its peers. The conclusion is as wrong as the recommendation. The discount simply reflects the fact that compared with its top peers, the company is at a lower valuation level because of lower performance. Splitting up the company does not automatically fix that performance gap (and might not even be needed). Over the years, practitioners and academics have debated whether a con- glomerate or diversification discount exists. In other words, does the market value conglomerates at less than the sum of their parts? Unfortunately, the results are incomplete. There is no consensus about whether diversified firms are valued at a discount relative to a portfolio of pure plays in similar busi- nesses.10 Some argue that they may even trade at a premium. Among studies that claim a discount, there is no consensus about whether the discount results from the weaker performance of diversified firms relative to more focused firms, or whether the market values diversified firms lower than focused firms.11 In our experience, however, whenever we have examined a company valued at less than pure-play peers, the company’s business units had lower growth and/or returns on capital relative to those peers. In other words, there was a performance discount, not a diversification or conglomerate discount. Summary Many large companies have multiple business units, each competing in seg- ments with different economic characteristics. Valuing such companies by their individual parts is standard practice in industry-leading companies and among sophisticated investors. Not only does it generate better valuation results, but it also produces deeper insights into where and how the company is generating value. To value a company by its parts, you need statements of NOPAT, invested capital, and free cash flow that approximate what the business units would look like if they were stand-alone companies. In preparing such statements, 10 P. Berger and E. Ofek, “Diversification’s Effect on Firm Value,” Journal of Financial Economics 37 (1995): 39–65; and B. Villalonga, “Diversification Discount or Premium? New Evidence from Business Infor- mation Tracking Series,” Journal of Finance 59, no. 2 (April 2004): 479–506. 11 A. Schoar, “Effects of Corporate Diversification on Productivity,” Journal of Finance 57, no. 6 (2002): 2379–2403; and J. Chevalier, “What Do We Know about Cross-Subsidization? Evidence from the Investment Policies of Merging Firms” (working paper, University of Chicago, July 1999).