Over the last two years, the general economy has righted itself and profits have rebounded strongly. Corporate cash balances are at an all-time high in many sectors, and shareholders have been rewarded for staying the course through the turbulence of the recent past. At long last, it seems, growth is back on the corporate agenda.
It is tempting, because it is easy, to succumb to the notion that as companies grow they will see a decline in either or both of their profitability and shareholder returns. But what if it were possible to avoid those trade-offs? What if we could find and learn from those companies that have delivered not only superior growth but also league-leading profitability and shareholder returns at the same time? In other words, what does it take to win growth’s “triple crown”?
Who would be king?
Baseball and horse racing provide two popular analogues. Baseball batters with the greatest number of home runs, runs batted in (RBIs) and highest batting average in a year are triple crown winners. In U.S. horse racing, the triple crown goes to the 3-year-old thoroughbred that wins the Kentucky Derby, the Preakness Stakes and the Belmont Stakes in the same year.
Triple crown winners are rare and, in fact, increasingly so. The last hitter to win a triple crown was Carl Yastrzemski with the 1967 Boston Red Sox, 45 years ago. (In the 45 years prior to The Yaz’s achievement there had been six triple crown winners.) Affirmed took the last horse racing triple crown in 1978. (In the previous 33 years there had been five.)
Part of what makes triple crown winners so special is that the events one must win to capture them demand not merely different, but often contradictory skills. Hitting home runs demands swinging for the fence, which often leads to strikeouts and a lower average; RBIs require putting the ball in play on demand—when runners are on base—in ways that get runners across the plate even if it gets you out, which reduces your home run total and your average; and a strong batting average might mean hard line drives (fewer home runs!) that result in the runner on first getting forced out at second (lower RBI total!). As for horse racing, for 3-year-old thoroughbreds the 1¼, 13/16 and 1½ miles of the Derby, the Preakness and the Belmont, respectively, are different enough that victory in any one of them often requires specialization at that distance.
Whether or not you’ve won the triple crown in baseball or horse racing is entirely unambiguous. But when it comes to corporate performance, as regular readers of this space will appreciate, identifying exceptional outcomes is not a straightforward undertaking. Unlike triple crowns in sports, we are not interested in performance over a single year. And when picking any particular period of time—five, ten, or some other number of years—we cannot help but make arbitrary choices that threaten to materially bias our results. If we want to find companies that have been truly remarkable on any measure of performance over time, we need to draw from the largest sample possible, which means making comparisons across different industries, different eras and different periods of time.
In the service of that objective, our sample consists of over 22,000 companies that traded on a U.S. exchange at any time between 1966 and 2008. However, when looking at so many companies over different lengths of time over more than four decades it turns out that systematic variance in company performance is very high and, at the same time, quite “sticky” at the high and low ends of the spectrum. This means that companies can deliver eye-popping results for a seemingly significant period of time even though it remains practically impossible to separate out the contribution of luck from that of company-level attributes.
To compare companies from such a diverse population we built a regression model for each performance measure. This regression allows us to predict what each company’s performance “should have been.” In each year, a company’s actual performance will typically deviate from this predicted value, falling either above it (a positive residual) or below it (a negative residual). The sum of these residuals is a company’s raw R-score (for “residual”).
Most researchers understand the importance of “controls” when comparing the performance of different companies. How can you compare an auto manufacturer in the 1980s with a software firm in the 2000s? Very often these controls take the form of looking only at similar companies over the same period of time – for example, an examination of retailers from 1995-2005. Whichever company comes out on top is the “great” company. This approach has its merits, but it is highly constraining: it relegates our assessment of great performance to identifying “big fish in small ponds.”
A more robust approach would allow us to compare any firm to any other over any period of time. Our approach allows us to do this. We have reliable and essentially complete data on every company that has ever been traded on a U.S. stock exchange between 1966 and 2008. Using these data, we build a regression model for each of ROA, revenue growth and TSR in which each performance measure is estimated as a function of a company’s values for each of the following control variables:
Year: Recessionary or expansionary economies make it harder or easier, respectively, to perform better. We correct for the “rising tide lifts all boats” phenomenon.
Revenue: Some performance measures, for example ROA, are systematically associated with company size. We want to recognize great management, not the mere ability to sit on a lead.
Industry: Industry structure has a profound impact on firm performance, and unless it is corrected for we can end up admiring firms that just happened to be in the “right” place.
Total observations: Our research indicates that companies that are around for a long time are less likely to enjoy strong performance due to luck alone. By correcting for longevity we can tell the difference between a flash in the pan and true “greatness.”
We cannot, however, simply compare raw R-scores, since companies with longer life spans will have systematically higher R-scores simply by virtue of having more of them. To correct for this, we ran a series of simulations to determine what the expected R-score is for companies with different numbers of observations. This allowed us to compute a “corrected R-score” for every company: a single measure for each of revenue growth, profitability (as measured by return on assets, or ROA), and total shareholders returns (TSR) that allows us to compare performance among all companies.
Given our particular interest in how to break the performance trade-offs that often accompany the pursuit of growth, we have focused first on those companies that fall into the top decile for growth, and then identified companies that also fall into the top decile along one or both of the other two performance measures. For convenience, we refer to the four categories as “G” (top decile for growth), “GR” (top decile for growth and ROA), “GT” (growth and total shareholder returns), and “GRT” (the triple crown winners).
In order to facilitate a more tractable examination of the profile and behaviors of firms with different performance profiles, consider for now only those that have at least ten years of data and that were going concerns in 2008 with at least $1 billion in revenue. This leaves us with 1,116 companies, of which 118 are at least great growers. (Statistical analysis suggests that this sample is representative of the full population.)
As shown in Figure 1, 55 percent of the great growers in that sample (66 of 118) delivered on more than just growth, consistent with the notion that, over the long run, growth cannot be an end in itself, but should drive, or at least be associated with, some form of value creation. At the same time, those firms that delivered on growth only are the single largest group, implying that delivering on more than one dimension is a genuine challenge.
Source: Compustat, Deloitte Consulting LLP analysis
Figure 1. Company count by performance category
Some descriptive statistics shed light on the profile of the companies that fall within each of our four categories. Intriguingly, there is no meaningful difference in the age or size (in 2008) of the companies in each of our four categories. The bubble chart below provides a sense of the observed performance by category. Note that ROA, TSR and growth rates for the GRT firms are higher than for any of the other categories, suggesting that not only is it possible to do well on multiple dimensions of performance simultaneously, it is possible to do very well. For although all the companies in this sample are in the top decile for their relevant performance measures, once a company clears that top decile cutoff, there is literally no theoretical limit to how high they can go.
Source: Compustat, Deloitte Consulting LLP analysis
Figure 2: Metrics by performance category
Statistical analysis reveals no significant differences among any of the categories on growth, which is to be expected since our sample is made up of companies that reached at least the ninth decile in growth for the total population. Intriguingly, neither is there any significant difference between GR and G firms on ROA or TSR. The GRT and GT firms, however, outperform the G firms on both ROA and TSR as indicated in the table below.
Other features of our sample are somewhat counterintuitive. For example, the average ROA of our GR firms is lower than the ROA for our GT firms. How can companies that have higher ROA on average (the GT companies) fail to meet the cutoff for top decile ROA when they perform better than the GR firms? We see the same effect on individual firms within categories: some G and GR firms have TSR values that are higher than some GT companies, yet they do not make the cutoff for top decile TSR performance.
Unadjusted performance differences when compared to pure growth companies
The answer lies in the importance of our controls, especially industry effects, on company ROA and TSR. Some industries have systematically higher TSR and ROA, and so companies in those industries must deliver better absolute performance in order to achieve a similar relative ranking.
With this in mind, perhaps the two most fascinating observations about performance differences by category are these: First, GRT firms achieve the highest raw and corrected levels of performance across all three measures. In other words, our triple crown winners do not “just” clear the top decile hurdles, ceding highest honors to “specialists” in each category. Rather, they finish at the top.
Second, the firms that fail to achieve ninth decile performance on multiple measures do not “just miss” on the other measures, they miss by a mile. The G firms, for instance, do not achieve ninth decile on growth and then eighth decile on other measures; they are in the sixth and third deciles for TSR and ROA.
It is important to note that we cannot claim a causal relationship between being in the triple crown category (our GRT firms) and the magnitude of their performance along each dimension. To do that, we would have to show that superior performance along all three dimensions was a consequence of clearing the ninth decile hurdle in each. However, at the risk of mixing metaphors, these findings suggest that GRT firms are rather like an Olympic triathlon champion with times for each stage (swim, bike, run) that would have earned them the gold in those individual events.
With individual event champions (G, GR and GT firms) doing so poorly on other dimensions, and GRT firms doing so well on all three, it’s just possible that instead of trade-offs among multiple dimensions of performance there is instead a positive feedback loop. Could it be that, beyond market share and influence with customers and suppliers, growth can create self-perpetuating momentum, the kind that draws to a company the best people, the most promising opportunities, and enthusiastic customers? That enthusiasm in turn might drive profitability, making possible the sort of reinvestment in R&D and people that over time establishes a daunting lead over one’s competition. The resulting growth might drive strong shareholder returns, creating a currency for growth, starting the cycle anew. Call this the “performance synergy hypothesis.”
Source: Compustat, Deloitte Consulting LLP analysis
Figure 3: Corrected vs observed revenue CAGR, TSR, weighted ROA
Elements of trade-off-free growth
In an attempt to identify some of the possible differences in behavior among our performance categories that might be associated with their performance over time, we looked at the total growth achieved by every firm in our sample that was generated through acquisition and through merger activity.
Determining the contributions to growth of M&A and geographic expansion
To calculate the contribution to growth from geographic expansion, we divided the change in non-domestic sales by overall sales growth over our period of observation. For example, a company entering the database in 1994 with $150 million in sales, all domestic, that generated its first foreign sales of $10 million in 1998 and by 2008 had $1.3 billion in sales with $300 million in foreign markets, has a geographic growth contribution of 300 / (1300-150) = 26 percent.
For M&A, we added the summed revenue of acquired firms in the year of acquisition and divided by total growth. A company that acquired three companies with revenues of $50 million, $75 million and $100 million while the firm’s total revenues grew from $200 million to $1.75 billion has a contribution to growth from M&A of: (50 + 75 + 100) / (1750 – 200) = 14.5 percent
Note, however, that these two numbers can add to more than 100% since our calculation of each is independent of the other; that is, we cannot identify geographic expansion driven by an acquisition.
The box plots below illustrate the distribution of total growth driven by each of these behaviors. A regression analysis reveals what is not immediately evident in a visual representation of the data: that GRT and GT firms generate on average about 40 percent and 25 percent less, respectively, of their growth from M&A than the pure growth companies, despite, recall, having entirely comparable rates of lifetime growth. GR and G firms are not materially different from each other.
These findings are in many ways consistent with both the conventional wisdom and the academic research on M&A. It is not uncommon to hear the refrain that acquisitions—especially larger ones—are systematically associated with lower profitability and lower shareholder returns for the acquiring firm. What we observe is that triple crown winners generate a mere 7 percent, on average, of their lifetime growth from M&A, while pure growth firms—which fail to deliver exceptional TSR or ROA performance—derive on average 46 percent of their growth from dealmaking. GT firms—companies with strong shareholder returns but less impressive profitability—split the difference, with an average of 22 percent of their growth coming from such sources.
Source: Compustat, Deloitte Consulting LLP analysis
Figure 4: Growth by M&A
Source: Compustat, Deloitte Consulting LLP analysis
Figure 5: Growth by geographic expansion
There are any number of interpretations for these data, but in our view they are consistent with the view that merger activity does not necessarily improve profitability, and often erodes it. And although few executives are likely to go into a deal intending to sacrifice their firms’ profitability or their shareholders’ wealth, it is noteworthy that so many G firms manage to do so consistently.
(The relatively high percentage of growth from M&A by the GR firms warrants explanation: there are only four companies in this category, and the range of growth from M&A is from 0 percent to 148 percent, with the high end of that range driven by a very large and dominant company competing in a highly regulated industry.)
Geographic expansion shows the opposite trend, with GRT firms deriving on average 40 percent of their growth in this way, about 16 percent more than pure growth companies; the other three categories are no different from each other. This suggests to us that successful geographic expansion delivers growth that is surprising to shareholders (contributing to the superior TSR – the “T” in GRT) and can be done in ways that are consistently profitable (the “R”). GR, GT and G companies tend to exploit geographic growth far less, at 19 percent, 26 percent, 24 percent of their total growth, and apparently less effectively, given their lower levels of overall profitability.
Beyond their standalone impact, are there any synergies between these two growth levers? That is, is there any relationship between the degree of geographic expansion and M&A activity within firms by performance category? Statistically, the answer is “no”: firms within each category appear to pursue their geographic expansion and M&A-driven strategies largely independently of each other. Consequently, while definitive conclusions are elusive, it would appear that among companies achieving significant growth, there is little evidence that their international expansion is fueled by cross-border acquisitions. Again, this is at least consistent with credible academic research that while M&A generally is fraught with risk, cross-border M&A is an entirely different sort of difficult – one that the top performing firms do not appear actively to seek out, and might even avoid.
Cisco Systems: An archetypal triple crown winner
Averages and general tendencies are helpful but can be somewhat blood-less. To bring to life the contours of a triple crown winner, consider Cisco Systems. Founded in 1984 and going public on the Nasdaq exchange in 1990, Cisco is widely regarded as having enjoyed a dynamic and highly successful quarter century.
Cisco provides a useful starting point for understanding the defining attributes of a company that has performed well over time because not only does it clear our statistically defined benchmarks for superior performance, its performance has been impressive by even merely intuitive standards of excellence.
So how did Cisco do it? The definitive explanation remains to be written, but we can offer a few observations that illustrate some common features of a triple crown winner.
First, Cisco has generated 46 percent of its growth from geographic expansion, very near the average for the GRT category and higher than in the other great grower categories. Cisco has pursued its geographic expansion in a systematic manner by capturing significant (>25 percent) market share in the markets it has entered. A perhaps defining element of its success has been its tendency to expand the geographic footprint of its R&D and manufacturing functions as well. In effect, Cisco has globalized as a company, rather than merely seeking customers beyond its home base in the United States. And, consistent with our observation that geographic expansion and M&A strategies seem to be independent of each other, Cisco’s geographic expansion seems to have been largely organic: rarely has an acquisition been a driver of growth in new regions.
New products, however, are another story. Much has been written about Cisco’s M&A prowess, and with good reason. However, for all the attention Cisco’s deal-making gets, the company has generated about 6 percent of its growth from M&A, despite having done over 150 deals between 1993 and 2008. The acquired companies were generally small compared to Cisco’s revenue at the time of acquisition and tended to be privately held companies built around a single technology. This implies rather strongly that the success of its acquisitions seems to be in what it does with them, not what they are upon being acquired.
Finally, perhaps one of the most well-documented findings on corporate performance is the negative impact of diversification. Clearly, when viewed from the perspective of its products and markets in the early 1990s, Cisco has diversified significantly: new products, new geographies, even entirely new businesses. Yet it seems to have avoided the “diversification discount.” How?
One possible answer is the gradual nature of its transformation over the years. The company’s primary focus for growth seems always to have been on strengthening and expanding its core business, both domestically, by targeting new customers for its existing products, and in new geographies.
The company has simultaneously stayed ahead or abreast of the rapidly advancing innovation curves in its various product markets by spotting trends and, where necessary, making relevant acquisitions. The result has often been an astute move into adjacent markets that have had clear linkages to its core business, often where the company has had brand permission and could easily leverage its core capabilities.
Ascending the throne: A CoDE for growth
It is always worth exploring whether there are any meaningful commonalities among firms that have a shared and remarkable performance profile. The value of identifying the specific actions or even general rules that increase one’s chances of performing similarly would be extraordinary.
As is seemingly unavoidably the case, however, the unifying themes that emerge tend to be rather high level. For even though Cisco’s experience is representative of triple crown winners generally, there are many more unique aspects to each triple crown winner’s story than there are shared ones. The dynamic nature of the challenges and opportunities faced by different companies at different times makes it challenging to have a formula-based approach to growth.
Within these constraints, our efforts to identify useful common characteristics has surfaced three recurring, if not universal, themes.
First, clarity in vision seems essential, if not in jump-starting growth, then certainly in avoiding the many distractions that can derail long-term success. Firms that grow profitability while creating shareholder value rarely pursue a single-minded dedication to one market, one segment or one technology. These firms can enjoy material growth but typically end up seeing their growth stall and their profitability erode. At the other extreme, firms that press growth but sacrifice value or profits have frequently diversified too much. The triple crown winners, in classic Goldilocks fashion, very often end up diversifying in material ways, and the sorts of transformations that characterize Cisco are far from uncommon.
One characteristic that allows these companies to strike the right balance seems to be a clear and compelling vision that is communicated broadly and shared widely across levels and functions. This vision helps triple crown winners to look beyond the constraints of their existing business without being distracted by every passing flashy bauble.
Second, disciplined resource allocation is a defining feature of triple crown winners. This discipline manifests itself in some unsurprising ways: a rigorous and multidimensional assessment of the merits of each new opportunity. At the same time, part of this rigor serves to compartmentalize resources in ways that ensure resources flow to strategically differentiated opportunities. In other words, not all dollars are created equal: triple crown winners invest consistently in the business of today and the businesses of tomorrow – both those that are extensions of current capabilities and those that expand the company’s competency footprint.
Finally, there is a common emphasis on excellence in execution across all functions. There is little evidence that strengths in any one area can compensate for weaknesses in others. It would appear that breaking trade-offs among growth, profitability and shareholder returns also requires breaking trade-offs in execution: triple crown companies are typically nimble and committed, efficient and effective, long-term thinkers and short-term obsessed.
In the end, triple crown companies share the following attributes:
- Clarity of vision
- Disciplined resource allocation
- Excellence in execution
And that is our growth CoDE. It might not win you the Derby, Preakness and Belmont, but it just might get your company that much closer to exceptional value creation.
Read the book: The Three Rules: How Exceptional Companies Beat the Odds, by Raynor and Ahmed, by Portfolio, the business imprint of Penguin Group (USA).