IS BIGGER REALLY BETTER?
by Donald V. Potter
The biggest wave of mergers and acquisitions in history is now underway. Each year seems to bring more deals involving more money. Recently, the financial services world saw two truly impressive mergers, one between Citibank and Travelers and the other between NationsBank and Bank of America. Each merger creates a behemoth in the financial services industry. In the automotive world, Daimler and Chrysler are joining forces. The recent past has seen similar waves of acquisition driven consolidation in a range of highly diverse industries: pharmaceuticals, oil services, disk drives, cosmetics, telecommunications, defense and health care, to name a few.
The assumption that profitability depends on market share leadership and increasing size relative to competition propels these transactions. The same assumption also drives many day-to-day management decisions about a business. Managements assume that as a company gets larger than its peers, it gains control over both its customers and its cost structure — that size creates economies of scale.
But is bigger really better? Do the market share leaders in most industries make better returns than do their smaller compatriots? In most cases, the answer is no. The relationship between market share and relative profitability is weak, not strong. Today, size may create headlines, but it is not producing commensurate returns.
Thoughts of industry leadership bring to mind companies such as Microsoft, Wal-Mart, Intel, Home Depot, Steelcase, and Deere. Some of these companies do lead their industries in both share and returns. Some do not. Steelcase is, by far, the largest office furniture manufacturer. But Hon Industries, at about a third the size of Steelcase, earns over twice the returns of its larger competitor. As big as it is, Wal-Mart rarely beats the returns of the smaller Family Dollar and Dollar General companies in the same industry. In pharmaceuticals, Merck leads in share but Pfizer earns better returns. Gannett is the largest newspaper publisher, but smaller publishers return more on their assets.
Windermere reached these conclusions after an extensive study of large, publicly traded companies in a variety of industries. Specifically, we used 1996 data for about 240 industries that have five or more competitors reporting line of business sales of at least $50 million. These industries ranged in size from $500 million (e.g., residential care, outdoor advertising services) to over $600 billion (e.g., motor vehicles, petroleum refining); the median industry had sales of about $7 billion. In each industry, we studied the top four competitors in sales in detail, measuring their market shares and returns.
We calculated the percentage of the time a company ranked first in market share and performed at the top of the four industry competitors in pre-tax return on assets. If market share leaders were to lead in proportion to their representation in the sample, the industry share leaders would lead their industry in return on assets 25 percent of the time. In all 240 industries, the share leader led the industry in pre-tax returns on assets 29 percent of the time, only 4 percent more than the random chance of 25 percent. Seventy percent of the time, then, the industry leader failed to lead his industry.
Leaders fared slightly better in industries that are larger or more concentrated. For example, we sorted these industries by size of total industry sales and examined only the top quartile of the 240 industries. The industry share leader then had a 38 percent chance of being the return leader, a result more in keeping with traditional expectations. We did a similar sorting analysis on industry concentration, ranking industries from high to low on the percentage of total industry sales controlled by the top four competitors. When we examined the top quartile of this industry sort, we found industry share leaders producing the best in industry returns 38 percent of the time. Industry leaders also led industry returns in mid-30s proportions when we looked at the top quartile of the 240 industries ranked by relative size of the sales of the top competitor compared to the second ranked company. Yet, even in these analyses, none found industry share leaders commanding the heights of industry returns as much as a simple majority of the time.
In other industry conditions, leaders fared even worse. For example, in the lowest return industries and in the most asset intensive markets, share leaders achieve the highest returns only 20 percent of the time. Growth is not particularly kind to share leaders either. The fastest growing quarter of these 240 industries had three-year compound growth rates over 20 percent, but only 27 percent of these industries’ share leaders reaped the highest returns on industry assets.
While industry leaders in market share led industry returns only a minority of times, the industry followers were stronger than we expected. In our total industry sample, the competitor ranking number four in market share led its industry in returns 23 percent of the time, only slightly less than the 25 percent random chance. These competitors who ranked fourth in market share performed the best in large industries and the worst in highly concentrated industries.
Why aren’t industry leaders more often in the lead? How is it that industry share leaders so infrequently achieve the highest returns, while distant followers seem to have the ability to lead in returns far more frequently than would normally be expected? Industry leaders often fail to lead because they become too enamored of market share for its own sake. They serve some customers who yield low returns, even in good times. In the press to be the largest, they accept customers who pay low prices and insist on high-cost services. We call these customers’ business “customer sludge” because it holds back the returns of the share leaders. Many of these leading companies also rely on size alone to create their cost advantages in the marketplace. In reality, size, and especially size relative to the other competitors in the marketplace, creates potential economies of scale — but that potential is not always realized. Economies of scale will not come into existence on their own. Companies have to force the development of scale economies by managing their cost structures (particularly in sales, general, and administrative cost areas) to ensure that unit costs fall as sales volume grows.
It appears paradoxical that the recent attention to costs, especially in re-engineering, has not ensured that share leadership yields return leadership. Re-engineering has surely helped many companies reduce their cost structures. But re-engineering has not separated competitors since virtually all competitors in most industries have undergone their own course of re-engineering. High returns within an industry are the result of a relative cost advantage over competitors, not the result of an absolute decline in costs. A share leader who would lead its industry in returns must exploit its relative size to create cost advantage. Competitors can copy most other cost advantages — but they cannot copy advantages derived directly from the leader’s superior size. The successful smaller competitors in a marketplace also share characteristics. For the most part, these smaller firms avoid direct confrontations with the much larger industry leaders. They target customers that the industry leaders under-serve, then shape their product service offerings and pricing policies to make a unique and compelling value proposition to those customers. And despite their small size, these companies also have an intense focus on cost control and reduction; their cost control efforts are every bit as disciplined and stringent as are those of the best industry market share leaders.
Despite this study’s findings, there is no inherent conflict between industry share and return leadership. They belong together, not apart. But they will not move in tandem until managements insist that unit costs decline as volume grows.
(Note: This Perspective was written in the context of the economy in 1998. While some of the companies may have changed their policies or indeed no longer exist, the patterns they exhibit still hold today.)
|For a greater overall perspective on this subject, we recommend the following related items:
Symptoms and Implications: Symptoms developing in the market that would suggest the need for this analysis.