Business Review

Executive Summary

Even companies that claim to have a long-term orientation worry about whether R&D is worth the investment. This is because they worry the resulting knowledge might walk out the door, as employees join other firms or start their own and because you can acquire firms that have the needed technology instead. If everyone followed that logic however, there’d be little innovation to walk out the door or to acquire. Fortunately, neither of these concerns is warranted, and research shows why companies, investors, and the nation will be better off if companies make long-term investments in R&D. In summary, there is no free lunch. Acquiring your way to innovation doesn’t work — the only way to stay innovative as a big company is to invest in R&D. Luckily for big companies, investing in R&D is a good strategy.

vincent tsui for hbr

Even companies that claim to have a long-term orientation worry about whether R&D is worth the investment. Sarah Williamson is the CEO of FCLTGlobal (formerly Focusing Capital on the Long Term), an organization cofounded in 2016 by BlackRock, CPPIB, Dow Chemical, McKinsey, and Tata Sons to encourage a longer-term focus in business and investment decision making. According to Williamson, a current concern among many institutional investors as well as corporations is that long-term investments in R&D have little payoff. This is both because the resulting knowledge might walk out the door, as employees join other firms or start their own, and because you can acquire firms who have the needed technology. If everyone followed that logic, however, there’d be little innovation to walk out the door or to acquire! Fortunately, neither of these concerns is warranted, and my research shows why companies, investors, and the nation will be better off if companies make long-term investments in R&D.

R&D Seldom Walks Out the Door

If an employee comes up with a great new product or a technical discovery during R&D, aren’t they likely to leave the company, either to found a startup or to join a competitor? In fact, this is unlikely. For one thing, most industries just don’t have many new startups, period. A recent study indicates that the average rate of new firm creation in an industry is 0.06%. This means in an industry of 100 firms, you can expect a new firm every 17 years. Related research looked at employee movement to both startups and existing firms and obtained similar results. Although the setting was law firms, rather than technology firms, all the assets in those firms reside in human capital. Even in that setting, where starting a firm is as simple as hanging a shingle, the researchers found that the probability of employees leaving to create new firms is 1%. Moreover, their work indicates that the employees you most want to keep are the ones most likely to stay. The likelihood of staying increases with both education and company tenure. Furthermore, if you treat these employees well, they are even more likely to stay. The likelihood of leaving decreases with pay (until salary reaches $5.2 million).

Small Firms Are Not More Innovative

What about the claim that companies can buy technology downstream through acquisitions? The first problem with this view is that it ignores where most innovation comes from. The prevailing view is that small, entrepreneurial companies are the source of innovation and growth in our economy. While it’s true that a small number of new firms are disproportionately innovative, big companies are the primary source of invention. Recent research I conducted under a grant from the National Science Foundation, and documented in my book, How Innovation Really Works, reveals that large companies invest more in R&D and have higher RQ (research quotient, a measure of the return on R&D investments). Companies with more than 500 employees not only do 5.75 times more R&D than small companies, but their R&D is 13% more productive — meaning large firms are the real engine of economic growth.

Given that large companies have more-productive R&D, why are small companies considered more innovative? For one thing, small-company innovation is often more visible. Small companies have to swing for the fences to attract market share from large companies, and home runs attract attention. The problem with swinging for the fences, however, is that the probability of hitting a home run is extremely low. Similarly, the probability of small company success is low: On average, only 25% of venture capital–backed startups ever return their invested capital. Further, VC-backed companies are a select group, as fewer than 5% of companies receive venture capital. So while hitting a home run attracts a lot of attention, it is the exception, not the rule.

A second reason people may think small companies are more innovative is that when they say “small,” they are typically thinking “young.” The general impression that small companies are more innovative stems from the fact we see the tremendous success of a handful of former startup companies: Amazon, Google, Apple. These former startups are still relatively young by corporate standards, and most have an advantage in that they still have their founders. While we don’t know all the reasons this is important, founders have stronger strategies (which is why they became successful) and better intuition about how all the pieces work and fit together (because they’ve managed the company from the seed stage). Both of these lead to better management and higher RQ. In addition, however, the company’s success allows the founder to maintain their strategy in the face of investor pressure for current profits. Thus, they exhibit less short-termism.

Startup Innovations Don’t Really Come from the Startup

So far, you might say, OK, there aren’t too many innovative startups. But shouldn’t big companies just try to acquire those few startups as a substitute for doing R&D internally? The answer is no. The reason why is that those startups appear more innovative than they really are, because their initial innovation typically comes from outside. It is well-known that Bill Gates didn’t develop DOS — he bought it for $50,000 from Seattle Computer Co. Similarly, Steve Jobs developed a graphical user interface after seeing one at Xerox PARC and hiring employees who had worked there. These stories are the rule rather than the exception. In a 2003 survey of Inc 500 companies, only 21% were based on ideas the founders researched on their own. The most likely source of ideas is the founder’s prior job. Fully 52% of venture ideas come from their prior company or industry, and another 14% come from a buyer or supplier to that industry. This isn’t to say that founders walked out the door with their employers’ intellectual property. This is merely saying that when all your expertise and contacts are tied to one industry, they likely have the greatest value as a new venture in that industry.

The late Steve Klepper conducted a series of industry studies examining how founders’ origins affected the likelihood that they would succeed in their new ventures. What he found across these industries is that the most successful startups came from founders who had worked for young, successful companies in the same industry. What these founders brought with them was a deep understanding of how the industry operates. They also brought with them a sense of how the product offered by the prior employer could be improved and where there was opportunity for new products that weren’t profitable for the former employer to pursue. The point of all this is that the innovations small companies are credited with often originate in large companies. So the best place for large companies to find innovation is to look inside. This is precisely what Xerox did when it created Xerox Technology Ventures. Whereas other corporate venture capital funds invest in external startups, Xerox’s invested in technologies emerging from Xerox PARC. The result was that its annual returns were four times that of the average VC fund.

What Happens When Companies Acquire Technology Firms

We’ve seen that when large firms have forgone developing a technology internally, it’s highly unlikely that it will be able to find a small company with that technology. Let’s assume for the moment, however, that such companies did exist. What’s the record on technology acquisitions?

Yazin Ozcan shared data on 2,378 technology mergers and acquisitions that he compiled for his paper “Innovation and Acquisitions.” His paper identified the top 16 technology acquirers (those with more than 10 technology acquisitions between 1986 and 2007). For each of the 17 companies, I compared their RQ with that of their industry. Seventy percent of these acquirers have lower average RQ than their industry, and all but three of them have below-average RQ. Thus these companies seem to be following the maxim “Those that can, do; those that can’t, acquire” — precisely the strategy captured in the ethos Sarah Williamson mentioned: “Instead of doing R&D, we’ll acquire companies who do.” The problem is that this ethos applied to the new acquisitions kills the target’s R&D as well — thus the lower RQ of acquirers.

In addition to this simple check of the 16 “high acquirers,” I conducted more formal analysis for the full set of technology acquisitions. In the analysis, I controlled for the initial quality of the acquiring company and examined what happened to their RQ as they assimilated their targets. I found that an increase in three-year acquisition intensity (dollars of acquisitions divided by acquirer’s revenues) corresponds with a significant decrease in company RQ. On average, a company with acquisitions whose value is 50% of its revenues sees a nine-point decrease in RQ following the acquisitions. At the very least, technology acquisitions fail to turn around troubled companies. More likely, they actively harm them.

There’s No Easy Alternative to R&D

In summary, there is no free lunch. Acquiring your way to innovation doesn’t work — the only way to stay innovative as a big company is to invest in R&D. Luckily for big companies, investing in R&D is a good strategy. Your inventors will seldom walk out the door with your ideas. And recognizing that you can’t buy the technology downstream will force you to be more productive with your R&D investment. Higher R&D productivity yields higher profits, growth, and market value per dollar of R&D.

But these conclusions make big companies’ reluctance to invest in R&D all the more concerning. If the large companies currently responsible for the innovation engines in this country shut down those engines in anticipation of acquiring technology startups downstream, they won’t just hurt themselves. With less big-company R&D to exploit, there will be fewer or less innovative startups. Most important, the entire nation will experience lower growth.