Analysis supporting regional competitiveness in a knowledge-based economy.

• home
• tools
• reports
• maps
• about
• sitemap
• contact

Innovation Index Methodology

Developing the Innovation Index began by identifying possible indicators used in previous analyses, as well as researching additional indicators that were theoretically important and available at the county level. This page overviews the rationale behind the selected index components.

Human Capital Sub-Index

Variables included in the human capital sub-index suggest the extent to which a county’s population and labor force are able to engage in innovative activities. Counties with high levels of human capital are those with enhanced knowledge that can be measured by high educational attainment, growth in younger age brackets of the workforce (signifying attractiveness to younger generations of workers), and a sizeable number of innovation-related occupations and jobs relative to the overall labor force.

• Education: Educational attainment measures the skills and knowledge that contribute to a population’s capacity to innovate. The research team was particularly interested in individuals in the labor force with tertiary degrees. Thus, educational attainment was divided into two categories:

  1. Some college or an associate’s degree

  2. Bachelor’s degree or higher

  The distinction is made to capture the relative importance of a knowledge differential, together with regional distinctions in the types of degrees earned. In many states, educational funding mechanisms favor 4-year universities. Elsewhere state policy tends to favor 2-year community colleges and vocational schools. An important educational differential is also present within states and counties where higher concentrations of bachelor’s degrees tend to surround metropolitan areas, whereas associate degree concentrations tend to be elevated in more rural counties where fewer residents have the resources or ability to travel to distant four-year institutions. Community colleges and vocational schools are more widely dispersed and proximate to rural residents. They also tend to provide education at a lower cost, with easier access, and tend to offer more flexible course schedules, such as evening or weekend courses. Community colleges are also more likely to cater to a region’s economic development needs than larger universities.

• Population Growth Rate: A growing population is desirable. But growth in the number of newborns or retirees does little to suggest whether those persons most likely to engage in innovative activities are present. For this reason, population growth rates are confined in this study to ages 25 to 44. The lower bound ensures transient college students typically aged 18 to 21 become less of a factor in influencing the overall rate of growth, whereas the upper bound signifies a point at which a professional’s geographic location would likely remain more stable. The 25-to-44 age bracket is likely to be less risk averse and more entrepreneurial. Moreover, population growth in this age bracket suggests the possibility that new residents are likely to expand the innovative and entrepreneurial characteristics of the base community.

• Occupational Mix: Certain occupational mixes favor innovative behaviors. The research team defined six technology-based knowledge occupation clusters that are hypothesized to have a higher probability of developing new and innovative ideas, products and processes that drive economic growth:

1. Information technology
2. Engineering
3. Health care and medical science practitioners and scientists
4. Mathematics, statistics, data and accounting
5. Natural sciences and environmental management
6. Postsecondary education and knowledge creation

• High-Tech Employment: In addition to knowledge occupation clusters, there are other occupations linked to high-technology firms and activities that either retain opportunities for the home-grown, skilled and specialized labor force or attract similar workers that are complementary to technology-based knowledge occupations. High-tech firm employment and growth is overwhelmingly found in urban centers, producing a rural-urban technology gap. The high-tech sector is defined by Moody’s as comprised of such industries as telecommunications, Internet providers, computer manufacturing, and scientific laboratories, to name a few. Together, the high-tech industry employment and technology-based knowledge occupational data provide a reasonable estimate of the extent to which a county’s occupational and industry mix provide either the existing capacity to generate innovative products and processes or the ability to augment local innovative capacity by attracting new firms and new talent.

Economic Dynamics Sub-Index

The economic dynamics sub-index measures local business conditions and resources available to entrepreneurs and businesses. Targeted resources such as research and development funds are input flows that encourage innovation close to home, or that, if not present, can limit innovative activity.

• R&D Investment: Inputs to innovation can come in the form of fund or knowledge transfers that may originate outside a region but benefit firms and individuals inside a region. Investments targeted to a region provide capital to aid the economic dynamics of a region. In particular, direct research and development (R&D) investments in a given county are indicative of overall levels of research being conducted. While research itself may not always result in a marketable innovation, it is a vital precursor. R&D expenditures are thus an indicator of innovation—even if the funds go toward unsuccessful products. It is generally understood, however, that those spending more will have the greatest innovative results or outcomes. R&D also has a well-documented spillover effect where R&D can provide crucial knowledge and resources for third-party firms to further innovate.
  
  Unfortunately, private R&D data is coded in a manner based on the location of company headquarters and not the research sites. Thus, the data may not reveal the true location of the R&D activity. Be that as it may, the R&D concept was operationalized in this study for each county by dividing total R&D expenditures by total worker compensation. In this way, the volume of R&D expenditures is adjusted for the level of productive activity in the county (as opposed to high personal income values that may be due to a large presence of retirees) and, to some degree, the local cost of living or doing business.

• Venture Capital Investment: Venture capital (VC) funds are used to launch new ideas or expand innovative companies. In the United States, VC may be responsible for up to 14 percent of all innovative output activity. VC investment firms are highly selective with their investments to maximize the probability of high returns. The return on VC, and possibly the importance of VC, is diminished somewhat by the fact that the VC investments are typically management-intensive. Looking for VC funding may consume a considerable level of effort by the seeking firm’s management, just as VC firms exert considerable effort seeking suitable projects to invest in.

• Broadband Density: Broadband provides high-speed Internet connections to businesses and consumers. Several state-level studies have attempted to capture the effect of adding broadband capacity to a region’s infrastructure. These studies suggest that broadband capacity has an overwhelmingly positive effect on economic performance. High-speed Internet access ensures that businesses and individuals can collaborate from virtually any location.
  
  The Federal Communications Commission (FCC) collects data on the number of broadband providers, not users or broadband lines, at a ZIP code level. To create a measure—that is, a broadband density proxy at the county level—broadband penetration was estimated by using population densities of ZIP codes to transform the FCC ZIP code data into county-level data. This measure of broadband penetration does not state how many individuals in a region have access to broadband. Given the limitations of the data, but acknowledging the theoretical importance of the concept, the index uses two broadband indicators. The first is a measure of current density and the second is the rate of density change.

• Churn: Competition is crucial to innovation. Market structures can influence the degree to which innovation is even possible. Specifically, markets with high rates of firm entry have been linked to increased levels of innovation. Conversely, the rate at which businesses shut their doors or reduce their workforce indicates a decrease in economic deadwood. Together the growth and contractions along with births and deaths produce the notion of economic churn, which serves as an indicator of the extent to which innovative and efficient companies replace outdated firms unable to modernize techniques and processes. Churn has been linked to positive employment growth and is not subject to agglomeration effects that often distinguish urban and rural economic structures.

• Business Sizes: Small firms, it is thought, are highly adaptable and can easily change their processes to incorporate new ideas. In recent years, high merger rates between small and large firms have coincided with increased technological influence of small firms. Some evidence, however, suggests these acquisitions may not be significant sources of innovation for large firms. Theoretically, a higher proportion of large businesses would positively contribute to innovation through the increased availability of funds for research and development, as well as the resources to directly employ scientists rather than hire out research services. Available data, however, do not identify whether, or the degree to which, an establishment is engaged in innovation activities. Moreover, using data on large establishments, defined as establishments with 500 or more employees, may be of limited utility for explaining innovative capacities in rural counties with small economies. Just the same, because the variable has some theoretical merit, the number of large establishments per 10,000 workers remains in the index.

Productivity and Employment Sub-Index

The productivity and employment sub-index describes economic growth, regional desirability, or direct outcomes of innovative activity. Variables in this index suggest the extent to which local and regional economies are moving up the value chain and attracting workers seeking particular jobs.

• High-Tech Employment Share Growth: Just as the share of high-tech employment in a county was an important input, the extent to which that share is increasing relative to total employment is an important performance measure. Firms requiring a highly skilled and specialized workforce are drawn to innovative areas. In a similar way, this measure also registers the degree to which home-grown, high-tech firms have expanded their presence. Growth in the share of high-tech employment suggests the increasing presence of innovative activity and signifies that high-tech firms are growing in the county or region both in relative as well as absolute terms.

• Job Growth-to-Population Growth Ratio: High employment growth relative to population growth suggests jobs are being created faster than people are moving to a region. Even though the ratio measures the change in level between jobs and population and, therefore, can’t be used to compare rates of growth, it can rank order counties or regions in terms of employment performance. A high ratio between these two variables indicates strong employment growth. A negative value signifies that population is growing while employment is declining or vice versa. In cases for which population is declining while employment is increasing, the absolute value of the ratio is used as that would be considered favorable employment performance.

• Patent Activity: Newly patented technologies provide an indicator of individuals’ and firms’ abilities to develop new technologies and remain competitive. The number of patents produced is a commonly used output measure for innovative activities, but the data can mislead. Patent data are coded to distinguish between the residence of the filer and the recorded location of the employer (if the applicant is not a private inventor), but the recorded location of the employer may or may not correspond to the location of the work that produced the patent, especially if the employer is a large, diversified company with many locations. In addition, the available patent data cover only utility patents and not all patent types. Patent data are recoded from the raw data provided by the U.S. Patent Office and awards patents to any county from which one of the filers reported as their location. This means that for any single patent with more than one filer, a patent may be counted multiple times if filers are located in different counties. Patents can also be an inaccurate indicator of innovation outcomes, particularly in areas where a single firm overwhelms the total patent count, such as Eli Lilly in Indianapolis.

• Gross Domestic Product: The final component of the productivity and employment sub-index is the single most important measure of productivity available—gross domestic product (GDP). The index incorporates both the level of a county’s current-dollar GDP per worker today, and also growth in the value over the past decade.

Economic Well-Being Sub-Index

Innovative economies improve economic well-being because residents earn more and have a higher standard of living. Decreasing poverty rates, increasing employment, in-migration of new residents and improvements in personal income signal a more desirable location to live and point to an increase in economic well-being.

• Average Poverty Rate: Innovative economies have greater employment opportunities with higher compensation, thus lowering rates of poverty. Reduced rates of poverty will tend to lag growth in employment opportunities. As a result, the last three years of the most recent data are used. Since a high poverty rate is a negative outcome, the index uses the inverse of the average poverty rate.

• Average Unemployment Rate: Innovative economies have greater employment opportunities and lower unemployment rates. Since a high unemployment rate is a negative outcome, the index uses the inverse of average unemployment rate.

• Net Migration: Migration measures the extent to which a county or region is broadly appealing and excludes other elements of population dynamics such as fertility rates. While people may migrate into a region for a host of reasons, from employment opportunities to environmental amenities, migration out of a region almost certainly signals declining economic conditions and the inability to keep the innovative talent that will spawn economic growth in the future.

• Compensation: Compensation data convey how much workers make based on their place of work. Likewise, proprietors’ income is also based on place of work. Compensation and proprietor’s income, therefore, probably provide a strong relationship between the activities of innovation and the rewards of innovation based on the location of innovation.

• Growth in Per Capita Personal Income: As an alternative to measuring remuneration based on place of work, per capita personal income (PCPI) measure incomes by place of residence. Because PCPI includes other forms of income in addition to wages, salaries and fringe benefits, it is a more comprehensive measure of well-being. That said, the linkage between where innovation occurs (county of work) and the financial rewards of innovation (county of residence) is less direct.

State Context Sub-Index

A fifth category, state context, seeks to capture data that are theoretically important but available only at the state level. It is composed of science and engineering graduates from state institutions per 1,000 residents of the state and research and development spending per capita. The state context category is not given as much attention because it is not used for the index calculation and because the indicator becomes diluted if a region crosses state boundaries.

More Information

• Calculating the Innovation Index
• What the Innovation Index Research Shows
• How to Use the Innovation Index
• More detailed information about the innovation index can be found in the report, Crossing the Next Regional Frontier: Information and Analytics Linking Regional Competitiveness to Investment in a Knowledge-Based Economy
  • The Innovation Index (Chapter 4)
  • Innovation Index Definitions, Calculations and Models (Appendix C)