Skip to Content

In Search of Innovation

You may know it when you see it, but before you can foster innovation in a company or nation you need to be able to get a grip on it. Meet the researchers who are trying to quantify this crucial but elusive process.
November 1, 1999

A bright light shone on the United States as business executives, academics and policy leaders-including Al Gore-converged on MIT’s glitzy Tang Center for the National Innovation Summit. It was the spring of 1998, and as the country savored one of its best economies ever, Harvard Business School professor Michael Porter sobered the powerful audience by pointing to dangers beyond the horizon.

True, U.S. companies had succeeded brilliantly in getting their houses in order, Porter said. But he and MIT professor Scott Stern were analyzing innovation in 25 nations-and preliminary projections from their “National Innovation Index” were not rosy. Indeed, based on such parameters as international patents filed, R&D spending and share of gross domestic product spent on higher education, it appeared the United States would tumble from the top spot to sixth place by 2005 (see companion article “National Numbers Game”). Porter delivered the news again in January at the World Economic Forum’s annual meeting in Davos, Switzerland. As a summary of the controversial report warned: “All the good macroeconomic news may be distracting us from looming threats to long-term U.S. economic strength.”

The Porter-Stern study addresses vital issues about the future of American competitiveness. But for those interested in predicting future trends in productivity, there’s one lingering problem: The index doesn’t measure innovation.

Inside the Black Box

Innovation is a tough nut to crack because it involves much more than invention (see “Invention Is a Flower, Innovation Is a Weed” by Bob Metcalfe). Truly measuring the process of innovation means going beyond traditional analysis of big inputs such as R&D spending and outputs like patents to shine a light inside the black box of national and corporate competition. It’s a problem that spans everything from investment in basic science to research and development tax credits, education systems, corporate programs, inter-industry dynamics and the entrepreneurial culture, all of which combine to fire the innovation process (see “Four Pillars of Innovation” by Michael Dertouzos).

Since many of these factors include qualitative as well as quantitative attributes, measuring innovation’s inner stuff is enough to give any management guru fits. Indeed, executives have struggled with such issues since science steamrolled into business in the 19th century, primarily in the American and German chemical and electrical industries. What’s new, though, are the intensity and magnitude of current efforts to track innovation.

The dawn of this modern era is sometimes traced to Harvard University economics professor Zvi Griliches’ 1979 paper, “Assessing the contributions of research and development to economic growth.” In it, Griliches proposed a theoretical framework for defining and measuring R&D, identifying and tracking its outputs-and charting their economic impact. Not content with theory, Griliches put his ideas into practice in the form of the Productivity Program at the National Bureau of Economic Research in Cambridge. It took years to build a foundation for deeper studies, notes program member and Harvard Business School professor Josh Lerner. But in the past decade, with the rise of powerful desktop computing systems that crunch hundreds of equations in seconds, he notes, “there’s been a whole flowering of research around these issues.”

The most significant initial finding was a strong link between research and development spending and overall productivity and growth. The connection prevails across every level of analysis: national, industrial and corporate. More recently, as the picture has been refined, much attention has centered on the most verifiable output of R&D, i.e. patents. In particular, researchers have devised ever more sophisticated ways to rank patents and evaluate them. On the level of individual organizations and firms, such measures are being used to pick potential marketplace winners and losers and even identify shifts in strategic focus.

Who Walks the Walk?

For instance, some cutting-edge researchers believe this approach can be used to pinpoint companies that will outperform their competitors. Francis Narin, president of CHI Research of Haddon Heights, N.J., has emerged as a pioneer in assessing firms’ technological strength. Narin believes that, in looking at the patents a high-tech firm generates, three variables are crucial: “citation intensity,” “science link” and “technology cycle time.”

Citation intensity gives an indication of how influential a particular patent is based on how often it’s cited in other patents. Science link represents the number of scientific papers cited in each patent, and technology cycle time is the median age of the patents cited in all of a company’s recent patents; these last two measures form a way of quantifying the widely held assumption that the most innovative patents are closest to the forefront of science and technology.

Narin has long combined these three factors to rank firms in key high-tech industries. Now, however, he’s going beyond just measuring to try to prove that high marks on these scales pay off financially. In a study published this spring in Financial Analysts Journal, he teamed with New York University finance expert Baruch Lev and doctoral student Zhen Deng to find out whether technological strength itself is an indicator of a firm’s future financial performance.

The trio’s hypothesis was that a company whose portfolio contains highly cited, science-rich patents is likely generating innovative technology, providing a marketplace advantage that will show up in future stock prices and market-to-book ratios-the stock price in relation to the value of a firm’s hardcore financial assets. All told, they studied 388 companies in four high-tech sectors: chemical, drugs, electronics and “others.”

Narin, Lev and Deng compared company results against industry averages, then gauged each firm’s market performance over the next year or longer. In line with their hypothesis, the market-to-book ratios of companies with high citation intensity and high science linkage outdid those of companies ranked low in both categories by 25 percent. These “high-high” companies also tended to do better in the stock market, though results were more mixed.

Since just about every company these days claims to be “innovative,” more sophisticated methods are needed to find out who walks the walk and who just talks the talk. Narin thinks his research can help. Armed with refined stock-picking parameters, in July he launched Investor Tech-Line, a $15,000-a-year database that provides monthly rankings of 250 firms to institutional investors.

By comparing actual market-to-book values to his technology-based ratios, Narin classifies companies as overvalued or undervalued. To get an idea of the power of the approach, consider this: A portfolio of equal dollar investments in Narin’s 20 most undervalued stocks from last December 31 would have increased 50 percent over the next six months-a time when the S&P 500 rose only a few points. Beating the market by such a large factor “blows your mind away,” Narin says.

“We Show the Money”

Part of the value of this analysis is that it can go beyond picking winners and losers to offering insight on a company’s technology strategy. Take Dow Chemical. The Midland, Mich., giant slipped from 455 patents in 1989 to 263 in 1995 (although it still patented at twice the industry rate). Meanwhile, Dow was laying out substantially more to come up with each patent than its competitors, while its citation intensity ranked below the industry norm.

But that was only part of the story, because another Narin-Lev measure shows that Dow was getting much closer to the cutting edge: Science linkage had grown significantly, from below the industry average to almost double it. Technology cycle time, meanwhile, ran at the industry average. These two measures taken together indicated that Dow was patenting developments close to the cutting edge of science, as fast as competitors turned out more standard inventions.

The study tracked firms as a group and not individually, and therefore the researchers didn’t check with Dow officials to find out whether these findings did indeed reflect changed internal thinking. But TR contacted Dow R&D vice president Richard M. Gross, who confirms that around 1990 his company did alter its strategy.

The strong science linkage, for instance, stems from a concerted push into new growth arenas that hinged on developing a more fundamental understanding of the science behind both its core existing businesses and promising emerging areas. For example, research into metallocenes-a new class of catalysts used to make plastics with more precisely tailored properties-allowed the company to develop improved polymers vital to everything from cushioning for high-end athletic shoes and sports equipment to thinner but stronger garbage bags.

This science focus was linked to a broader push to raise patent quality. “We had encouraged quantity historically, as a lot of companies had,” Gross acknowledges. Indeed, the since-dismantled patent honor roll at the main R&D building in Midland, Mich., spanned an entire wall, with brass plates bearing the names of top inventors. But as competitive pressures forced Dow’s Inventions Management group-now called Intellectual Asset Management (IAM)-to examine the costs and returns associated with this vast portfolio, it got a shock. “We had a $40 million liability on just patent costs,” relates Sharon Oriel, an IAM manager. “And yet we couldn’t show that other part-what’s the value.”

That wake-up call triggered a vigorous campaign to drop, spin off or license patents that didn’t fit the needs of Dow’s 14 global business groups, and target future patenting more closely on the company’s objectives. Working with Cambridge, Mass.-based consulting firm Arthur D. Little, IAM has developed proprietary methods of tracking patent status, costs and benefits, which it uses to facilitate business group planning. Oriel says this provides a new dimension to measuring innovation that resonates with what the “show-me-the-money” business groups have always wanted from research. Now her group wears T-shirts that say: “We show the money.”

Intellectual Capital

Moving beyond a broad-brush approach to dissecting the relationship between patents and innovation marks a big step forward. But patents certainly don’t tell the whole story. In today’s knowledge economy, many innovations are simply not patentable. Some of the most creative manufacturing and distribution companies-think Dell-won’t appear on any yardstick that tracks intellectual property. As a result, a lot of attention is shifting to other scales. Perhaps most notable is what’s often called “intellectual capital”-the sum total of the seemingly “intangible” knowledge, processes, culture, customer and supplier networks and other factors that can ignite industrial creativity.

Intellectual capital might be the hot zone of gauging innovation today. NYU’s Lev notes that in the manufacturing economy of the 1970s, the market-to-book ratio of S&P 500 corporations ran about 1:1. These days, it’s around 6:1, while high-flyers like Microsoft come in around 25:1-largely on the basis of their intellectual capital.

“It’s really mind-boggling when you think about this whole accounting machinery that ends up with a balance sheet that explains less than one-sixth of a company’s real value,” says Lev. “So we are talking here about an enormous asset which really is the engine of the success of companies and growth.”

Earlier this year, Lev compiled a Knowledge Capital Scoreboard for CFO magazine ranking 47 large chemical and pharmaceutical firms. He and portfolio manager Marc Bothwell of BEA Credit Suisse Asset Management analyzed each firm’s earnings for the past three years, as well as its projected earnings for the next three years. From the weighted average, or “normalized” earnings, they subtracted expected returns on tangible assets such as bond portfolios and physical equipment; the remainder, they assumed, represented knowledge-capital earnings.

For companies that are knowledge-rich, this kind of capital is a huge part of their success. Take Merck, considered one of the scientific leaders in drug development. In this analysis, Merck’s normalized earnings were $5.5 billion. Its physical assets were expected to generate $343 million annually, while some $624 million in financial assets were projected to earn $28 million. That left a whopping $5.1 billion a year attributed to knowledge earnings. According to Lev’s formula, this represents the return on intellectual capital assets totaling nearly $50 billion-which moved Merck to the top of the pharmaceutical pack, ahead of others with greater sales.

Intangible Assets

Studies like these are themselves on the cutting edge of their own field of research. For that reason, they entail a large number of assumptions that remain to be tested in subsequent work. Still, Lev’s studies are helping fuel a growing movement to get companies to produce such data in quarterly and annual reports, much as they generate traditional financial information. Indeed, following a three-day intellectual capital symposium in Amsterdam last June, the Organization for Economic Cooperation and Development reported widespread international support for the voluntary reporting of knowledge assets.

On this front, the clear pioneer is Skandia Life Insurance, a Stockholm-based global financial services powerhouse, which has published an intellectual capital supplement to its annual report since 1994. The supplement has tracked turnover rates, the number of contracts generated per employee and the percentage of women managers. It also includes stories meant to illustrate trends behind the figures.

A story in the 1998 supplement called “Female Potential,” for example, centered on a Skandia program aimed at bringing more women into sales leadership. The goal, according to the piece, was “to obtain more female leaders, and to actively eliminate obstacles to the realization of the full development potential of these women and add to the collective value-creating process within Skandia.” Toward that end, each woman was paired with a “personal mentor” from a different business area. “The mentor’s role is to be a dialogue partner and to be on hand for support and advice, while facilitating the candidate’s continued development and career in the company.”

If it all seems a bit touchy-feely, it’s because innovation itself is still a somewhat mysterious combination of work, skill, inspiration and environment. But for companies concerned with staying on the leading edge, such factors have become paramount. Now corporations are taking fresh looks at the whole value-creation chain, seeking to maximize returns on everything from employee-training programs to the colleges of expertise that enable people to share knowledge and skills.

Mark B. Myers, senior vice president of Corporate Research and Technology for Xerox, concedes that astute managers have always paid attention to these issues. What’s different now, he argues, is that the efforts are more sophisticated and they’re starting to tease apart the deeper elements of innovation.

Over the past two years, Myers has worked with Harvard Business School’s Rosenbloom to develop methods for determining whether Xerox is investing correctly in key elements of innovation. This undertaking, which involves tracking decades of R&D expenses against revenue growth and benchmarking Xerox’s investments against those of principal competitors, is still confined to tangible measures. But Myers is hoping the effort will lead him to greater inroads on the intangibles as well. After all, he notes, “a higher portion of the value creation in a knowledge-based economy versus a manufacturing economy may be contained on the intangible side.”

To the extent that Xerox is onto something, it will still represent a rare victory for the measurement forces. Nevertheless, a host of corporate leaders, academics and others remain bent on prying the lid off innovation’s black box. Each success promises to help the gears inside turn more smoothly.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.