96 years of economic insights for Indiana

The IBR is a publication of the Indiana Business Research Center at IU's Kelley School of Business

Where the innovations are: Patent making in the Great Lakes states

Co-Director, Indiana Business Research Center, Indiana University Kelley School of Business

Innovation is one of those “I know it when I see it, but I need to find my glasses” concepts. Innovation, a leading driver of economic and productivity growth, is difficult to measure—quantitatively or qualitatively.

The typical go-to measure for the innovation concept is counting patents. There is hard data on the number and technologies of patents, along with the geographic location of the inventors and their organizational affiliations. Patent technology often finds its way into the marketplace as new products or even start-up firms focusing on delivering products that meet the needs and wants of a heretofore unfulfilled customer niche.

A patent is the U.S. government’s stamp of approval that the technology described in the patent application is distinct from other technologies in previous patents—that is, this technology is unique. Cutting-edge patents propel productivity and economic growth because it benefits at least two members of the free-market system: consumers and producers. It benefits producers along the supply chain because the product/technology meets a need of another link in the value-added chain to reduce production costs (and raise profits). It may also fulfill a need of consumers to reduce their effort and time to do tasks (think washers and dryers), reduce their pain (think knee replacements), or increase their enjoyment (for example, better TV screens or fabrics). If a patent technology does not benefit either producers or consumers, it will not be commercialized at full scale. In other words, the product, and maybe the firm, will flop. When there are benefits to producers and consumers, profits and wages typically rise. And following, productivity (or value-added) per worker or worker-hours worked also increase.

For this reason, making patents in a region is a good proxy for innovation in a region.

By extension, in regional economic development, there is the concept and type of policy initiative called “technology-based economic development” or TBED. If a region is producing new technology and patents, there is a good chance that many of those technologies will find their way into products at commercial scale and, hence, juice the regional growth of profits, employment and wages. Not all patents, or entrepreneurs who look for the potential applications of technologies to market segments, will find a favorable, growth-stimulating product that can be commercialized. But many technologies will. Reducing the friction from technology discovery to technology development to market application and opportunity to financing and building production to scale are goals of several U.S. government programs. For example, the federal government has initiatives for national laboratories to take their technologies and, typically with partnerships with private enterprises, polish and perfect their technologies from birth to commercialization and eventual market scale. There is even an established progress ruler to monitor and evaluate the success of government-funded technologies as they make their way from pure science to the marketplace.

Patents not only provide a signal for future economic regional growth potential, but the patent data also have information and geographic specifics on who the inventors are, whether there are several collaborators and the organizations with whom they are associated. This matters for an economic development practitioner (EDP) who may be looking for partners with whom they may apply for TBED-related research and development funding. Or, it may also help EDPs understand their local web-of-science—e.g., network analysis and knowing whether a critical technology resource is connected within-region (at a local university or corporation) or dependent upon knowledge flows from outside the region (from collaborators and relationships across the nation or internationally). Knowing what technology matters, and who matters, for a particular innovation relationship in a region can help EDPs formulate their vision and strategies for future industry and economic development. This network information, along with other resource or asset inventories, pressure checks whether a proposed development policy and strategic direction has a high or low chance of success.

Before concluding this section on why patents matter in economic development, it is important to note that some profound benefits of technology may not show up in the productivity statistics as value-added (profits and compensation). Environmental benefits are one example. Simply put, if there is a technology that, without an increase in capital costs (that reduce profits), reduces smokestack emissions of mercury and other effluent nasties, while not reducing electricity output, increasing the cost per megawatt to consumers, or reducing employment and wages at the plant, that is a huge social and economic win. That environmental benefit will not register in the productivity statistics or GDP, however. In this simplistic case, the regional benefits of fewer cases of asthma or neurotoxin-related outcomes don’t register. (Explaining how these benefits can be measured is not within the scope of this paper. There are, however, established and well-accepted methods to put a price and benefit on mitigating the negative effects of pollution.)

In summary, this introduction tackles the question of why patents and technologies are important and tees up questions like:

  1. Which regions are more innovative?
  2. How does regional innovation impact the regional “bottom line”?

In this article, we address question one for the Great Lakes region’s states and counties. We’ll address question two another day in a future article.

Patent data

The U.S. Patent and Trademark Office (USPTO) employs some 12,000 people. So the following quick-and-dirty overview on patent data will lack in precision. By that, dear reader, I mean be grateful. Some of what follows has been known to make crusty tough guys weep.

For the following reckoning of regional technology concentration, the first key distinction for a patent is if it has been awarded—not just applied for. Many, if not most, of those 12,000 people determine whether a patent application demonstrates whether the technology expressed is new or just derivative. There is also a distinction between utility patents and design patents. A utility patent is the invention of a new and useful process, machine, manufacture, composition of matter, or a new and useful improvement thereof, according to the USPTO. A design patent is new, original and ornamental design embodied in or applied to an article of manufacture. If one has come up with a teapot that transfers heat more efficiently and quickly on a stove, thereby cutting the wait time by half, that sounds like a utility patent. Making the teapot round, square or oval, is design.

Likely the thorniest and most profound patent data change has been the transition from the U.S. patent classification (USPC) system to the Cooperative Patent Classification (CPC) system. According to USPTO, on January 1, 2013, this transition occurred completely in the U.S. As the reader will soon surmise, the transition was not and is not seamless. The USPC is based upon technology groupings and common subject matter (based on USPTO language). From the author’s point of view, the CPC is more use or application driven than technology driven. There are many close alignments with technology and use—the case of peptides, for example. The technology of making and applying peptides is rather tight. The classification of “brushes,” however, presents a problem. Brushes are everywhere, from cleaning teeth to actuating electrical motors to reading credit cards to cleaning bathrooms. The ubiquity of “brushes” in products and processes according to the CPC can be disorientating. “Brushes” as a technology intersect with many end uses. Consider this alternative, hypothetical example: In the CPC scheme, the same electronic thermometer technology or device that is placed on a ski jacket is considered apparel, placed on a wall in a house is considered construction, used in an automobile to regulate heat flow is considered vehicular, or used to measure temperature at a research station in Antarctica can be considered mitigating climate change. It is the same technology applied to different uses.

This makes the conversion from one system to another tricky.

Thus, reconciling USPC and CPC is akin to the NAICS-based framework of production—like a restaurant preparing meals—and consumer-based frameworks of consumption like the “tourism sector.” Tourism consists of multiple NAICS-based industries like hotels, eating and drinking establishments, buses, auto and boat rental, public transportation, gas stations and tchotchke stores, among other tourist end uses. (Consult the BEA tourism satellite account for the industries that comprise tourism.)

To repeat, these two classification systems are not always well-aligned and it depends upon the technology and application classification to assess the probability that a technology applies to the CPC. But more importantly for the EDPs and TBED policy advocates, the CPC may not be helpful in determining the technology and science foundation of a region.

We did our level best to translate the CPC to the technology categories of the USPC and, following, general technology classes that a newcomer to the technology work can easily assess and understand. We translated patent counts using the CPC after 2012, as well as patent counts from 2002 to 2012 for a double check, using a transformation matrix. In so doing, we are able to create a time series of patent making by county by USPC, for which there are 464 categories, and then aggregated those into 13 IBRC classes (see sidebar) from 2002 to 2019. After all, our goal is to see technology concentration by region and how that concentration may be changing over time.

How did we do this? First, based on contact with one of the authors1 who created multiple crosswalks/concordances between the CPC and USPC categories and NAICS industry technologies, we needed to create our own crosswalk between the CPC and the USPC. Given that there was no publicly established crosswalk between the two patent/technology classifications—CPC to USPC—we created our own. We combined two translation, or crosswalk, probabilities: 1) the probability of CPC technology classification matching the technologies association with the production-based NAICS categories, and 2) the probability of the NAICS-based technology and production categories matching the USPC patent classifications. These probabilities were developed by Goldschlag and colleagues (2016). This team compared the texts describing each classification description and, using a text mining scheme, assigned a probability based on textual similarity. We used 6-digit NAICS codes as the bridge within the CPC to USPC bridge. We applied the intersection of the 4-digit CPC categories to NAICS 6-digit probabilities and the NAICS to USPC probabilities. (The resulting probability matrix for CPC to USPC is the product of two matrices. Because some codes come and go over time—those who use and follow the NAICS code assignments for industries over the last few decades know what this means—using a fixed matrix of CPC to USPC assignments over time results in some slippage betwixt the cup and the lip.)

Finally, we express patent and technology concentration in two ways. One, by simple patent counts annually by county and by the 13 IBRC categories—an absolute measure. And two, by patent counts per capita annually by county and by the 13 categories—a relative concentration measure. In this way, we hope to tease out how regional scale, or population, affects patent making in contrast to relative patent productivity. If smaller regions are punching above their weight in patent making, that is an important characteristic to note for economic development potential, or, for that matter, determining whether economic development initiatives or private investment from the past have grown deep roots.

Patent-making patterns in the Great Lakes states

We use the Great Lakes states as a case study and pressure test of our work to show technology concentrations geographically and over time.

Table 1 shows the total patent concentration for the Great Lakes states of Illinois, Indiana, Michigan, Minnesota, Ohio and Wisconsin for the year 2017. While Illinois and Ohio have more residents, Michigan generated the greater number of patents. As we will see shortly, the type of patents produced in Michigan explains why it is the count leader—and helps provide the justification for differentiation by technology type. If adjusted for scale, i.e., the size of the state population, Minnesota rises to the top. Minnesota has about half the population of Ohio, but its total patent count is not far behind Ohio.

Table 1: Total patents and patents per capita ratios, Great Lakes states

State Total patents Population Patents per capita (000)
Illinois 5,203 12,778,828 0.407
Indiana 2,235 6,658,078 0.336
Michigan 6,680 9,973,114 0.670
Minnesota 4,356 5,566,230 0.783
Ohio 4,706 11,659,650 0.404
Wisconsin 3,460 5,790,186 0.598

Source: IBRC, using 2017 data from the U.S. Patent and Trademark Office and U.S. Census Bureau population estimates

Table 2 shows the top 25 counties in the Great Lakes states for both total patent counts and patents per capita in 2017. Indiana counties, like the state as a whole in terms of patent making, do not show well compared to their Great Lakes neighbors. Marion County, the leading patent making county in Indiana, ranks 20th. If adjusted for scale, Marion County doesn’t show up, which is true for many of the top 25 counties by patent count. Three out of the top five counties in patents per capita are in Minnesota, indicating that Minnesota punches well above its weight in patent production. Minnesota employees, firms and other institutions are highly productive in generating intellectual property. For Hoosiers, it should come as no surprise that Kosciusko County, and its neighboring Whitley County, are the Indiana counties with the greatest patenting rates when adjusted for population.

Table 2: Top 25 counties in total patents and patents per capita, Great Lakes states

Rank Total
patents
County, State
1 1,823 Oakland, MI
2 1,757 Wayne, MI
3 1,541 Cook, IL
4 1,328 Hennepin, MN
5 1,085 Washtenaw, MI
6 739 Lake, IL
7 681 Washington, MN
8 678 DuPage, IL
9 606 Summit, OH
10 594 Dane, WI
11 579 Hamilton, OH
12 572 Waukesha, WI
13 542 Ramsey, MN
14 538 Cuyahoga, OH
15 530 Warren, OH
16 456 Olmsted, MN
17 438 Ottawa, MI
18 398 Franklin, OH
19 394 McHenry, IL
20 367 Marion, IN
21 342 Scott, MN
22 313 Will, IL
23 272 Milwaukee, WI
24 270 Kane, IL
25 241 Kent, MI
Rank Patents per
capita (000)
County, State
1 2.947 Olmsted, MN
2 2.942 Washtenaw, MI
3 2.666 Washington, MN
4 2.350 Scott, MN
5 2.317 Warren, OH
6 2.224 Woodford, IL
7 2.089 St.Croix, WI
8 1.772 Kosciusko, IN
9 1.634 Sheboygan, WI
10 1.528 Ottawa, MI
11 1.483 Whitley, IN
12 1.451 Oakland, MI
13 1.443 Redwood, MN
14 1.428 Midland, MI
15 1.427 Waukesha, WI
16 1.422 Highland, OH
17 1.309 Pierce, WI
18 1.280 McHenry, IL
19 1.225 Renville, MN
20 1.215 Winnebago, WI
21 1.169 Steele, MN
22 1.156 Moultrie, IL
23 1.154 Tazewell, IL
24 1.129 Berrien, MI
25 1.119 Summit, OH

Note: Bold cells indicate Indiana counties.
Source: IBRC, using 2017 data from the U.S. Patent and Trademark Office and U.S. Census Bureau population estimates

Figures 1 and 2 provide maps of these counties with the highest concentration of patent making.

Figure 1: Map of top 25 counties in total patents, Great Lakes states

Map of leading counties in Indiana, Illinois, Ohio, Michigan, Minnesota and Wisconsin. More than 1,000 = 5 counties; 600 to 1,000 = 4 counties; 400 to 599 = 8 counties; Less than 400 = 8 counties.

Source: IBRC, using 2017 data from the U.S. Patent and Trademark Office

Figure 2: Map of top 25 counties in patents per capita, Great Lakes states

Map of leading counties in Indiana, Illinois, Ohio, Michigan, Minnesota and Wisconsin. More than 2.0 = 7 counties; 1.501 to 2.0 = 3 counties; 1.3 to 1.5 = 7 counties; Less than 1.3 = 8 counties.

Source: IBRC, using 2017 data from the U.S. Patent and Trademark Office and U.S. Census Bureau population estimates

Figures 3 and 4 are interactive maps showing the patenting landscape for all counties in the Great Lakes states.

Figure 3: Map of total patents by county, Great Lakes states

Figure 4: Map of patents per capita by county, Great Lakes states

The question then becomes: What sort of patents are these states and counties making? Or, put another way, what are we good at?

Tables 3 and 4 present the region’s patenting from 2002 to 2019. Table 3 shows the patent counts by IBRC category over time. Almost across the board, the patent counts have been on a consistent rise, with the exception of most—but not all—categories during the Great Recession. This is particularly interesting as it shows how research and development outcomes and filing for patent protection based on recent work may be sensitive to current market conditions instead of future opportunity. It may also hint that certain types of patent filings and awards are less abundant in consumer, construction and industrial technologies, while computer and communications R&D, together with design, are not as sensitive to the business cycle. This may reflect the technology development cycle duration between one classification and another. These are simply initial questions that these data elicit. (Answering those questions is not within the scope of this article.)

Table 3: Annual patent count by technology class, Great Lakes states, 2002-2019

Class title 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
Chemicals, including coatings, except pharma 1,646 1,484 1,381 1,013 1,122 965 920 985 1,382 1,350 1,421 1,567 1,682 1,609 1,498 1,531 1,455 1,582
Communications 638 596 727 668 918 735 800 845 1,005 969 1,115 1,185 1,254 1,240 1,323 1,419 1,456 1,567
Computers, information technology and data processing 1,130 1,227 1,303 1,137 1,485 1,273 1,366 1,443 1,888 2,010 2,264 2,451 2,582 2,371 2,216 2,246 2,258 2,666
Bio-tech and pharma 637 557 573 500 575 553 524 650 741 812 1,021 1,090 1,157 1,057 1,134 1,091 1,008 993
Electronics and electrical 1,048 1,033 1,030 925 1,077 932 910 968 1,064 1,082 1,069 1,230 1,278 1,350 1,407 1,616 1,418 1,612
Mechanical 2,782 2,876 2,610 2,259 2,588 2,268 2,132 2,081 2,754 2,795 3,069 3,178 3,365 3,260 3,341 3,493 3,338 3,881
Transportation, material transfer and storage 2,877 2,848 2,592 2,370 2,600 2,282 2,025 2,002 2,607 2,632 3,000 3,288 3,426 3,639 3,736 4,091 4,071 4,480
Agriculture and natural resources 517 520 430 338 390 296 284 321 421 372 413 484 501 430 465 522 487 548
Building, construction materials and methods 1,040 977 934 856 898 730 688 699 990 950 953 1,109 1,223 1,241 1,179 1,399 1,340 1,551
Power generation and other industrial, including armaments 729 761 712 630 692 600 589 671 845 865 916 1,034 1,135 1,200 1,270 1,378 1,367 1,586
Consumer goods, including furniture 1,420 1,385 1,263 1,065 1,179 1,034 973 945 1,332 1,415 1,405 1,498 1,586 1,601 1,592 1,644 1,521 1,793
Medical devices and medical practices 1,286 1,303 1,050 933 1,146 1,082 982 1,085 1,714 1,807 2,039 2,342 2,486 2,359 2,363 2,559 2,403 2,892
Design 2,118 2,481 2,138 1,705 2,766 3,128 3,304 2,737 2,866 2,821 2,852 2,975 3,086 3,253 3,451 3,598 3,521 4,086
Total 17,869 18,050 16,744 14,399 17,436 15,878 15,497 15,432 19,610 19,882 21,538 23,429 24,762 24,611 24,976 26,588 25,644 29,238

Source: IBRC, using data from the U.S. Patent and Trademark Office

Table 4 shows the proportion or percentage of a technology class over time. This is the table, or the data aggregation, that hints at the robustness of the CPC to USPC transformation matrix and whether it can be trusted to estimate technological concentrations across geographic space. The proportions appear stable—rising and falling with smooth transitions—but also show the ripple effects of some technologies harder hit by the Great Recession while others remained constant or even rose, as was the case with design patents. The table also shows that over time, technologies in chemicals and coatings, agriculture, and consumer goods have declined in relative importance in the Great Lakes states; meanwhile, communications, computers and IT, and medical devices have risen in the share of patent technology in the region.

Table 4: Annual patents by technology class, Great Lakes states, 2002-2019

Class title 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
Chemicals, including coatings, except pharma 9.2% 8.2% 8.2% 7.0% 6.4% 6.1% 5.9% 6.4% 7.0% 6.8% 6.6% 6.7% 6.8% 6.5% 6.0% 5.8% 5.7% 5.4%
Communications 3.6% 3.3% 4.3% 4.6% 5.3% 4.6% 5.2% 5.5% 5.1% 4.9% 5.2% 5.1% 5.1% 5.0% 5.3% 5.3% 5.7% 5.4%
Computers, information technology and data processing 6.3% 6.8% 7.8% 7.9% 8.5% 8.0% 8.8% 9.3% 9.6% 10.1% 10.5% 10.5% 10.4% 9.6% 8.9% 8.4% 8.8% 9.1%
Bio-tech and pharma 3.6% 3.1% 3.4% 3.5% 3.3% 3.5% 3.4% 4.2% 3.8% 4.1% 4.7% 4.7% 4.7% 4.3% 4.5% 4.1% 3.9% 3.4%
Electronics and electrical 5.9% 5.7% 6.2% 6.4% 6.2% 5.9% 5.9% 6.3% 5.4% 5.4% 5.0% 5.3% 5.2% 5.5% 5.6% 6.1% 5.5% 5.5%
Mechanical 15.6% 15.9% 15.6% 15.7% 14.8% 14.3% 13.8% 13.5% 14.0% 14.1% 14.3% 13.6% 13.6% 13.2% 13.4% 13.1% 13.0% 13.3%
Transportation, material transfer and storage 16.1% 15.8% 15.5% 16.5% 14.9% 14.4% 13.1% 13.0% 13.3% 13.2% 13.9% 14.0% 13.8% 14.8% 15.0% 15.4% 15.9% 15.3%
Agriculture and natural resources 2.9% 2.9% 2.6% 2.3% 2.2% 1.9% 1.8% 2.1% 2.1% 1.9% 1.9% 2.1% 2.0% 1.7% 1.9% 2.0% 1.9% 1.9%
Building, construction materials and methods 5.8% 5.4% 5.6% 5.9% 5.2% 4.6% 4.4% 4.5% 5.0% 4.8% 4.4% 4.7% 4.9% 5.0% 4.7% 5.3% 5.2% 5.3%
Power generation and other industrial, including armaments 4.1% 4.2% 4.3% 4.4% 4.0% 3.8% 3.8% 4.3% 4.3% 4.4% 4.3% 4.4% 4.6% 4.9% 5.1% 5.2% 5.3% 5.4%
Consumer goods, including furniture 7.9% 7.7% 7.5% 7.4% 6.8% 6.5% 6.3% 6.1% 6.8% 7.1% 6.5% 6.4% 6.4% 6.5% 6.4% 6.2% 5.9% 6.1%
Medical devices and medical practices 7.2% 7.2% 6.3% 6.5% 6.6% 6.8% 6.3% 7.0% 8.7% 9.1% 9.5% 10.0% 10.0% 9.6% 9.5% 9.6% 9.4% 9.9%
Design 11.9% 13.7% 12.8% 11.8% 15.9% 19.7% 21.3% 17.7% 14.6% 14.2% 13.2% 12.7% 12.5% 13.2% 13.8% 13.5% 13.7% 14.0%

Source: IBRC, using data from the U.S. Patent and Trademark Office

So far, we have explored our region’s high-level technology metrics and hinted what was to come. Namely, detailing the IBRC technology classes for the leading patent-making counties in the Great Lakes region. We present the region’s top 10 counties plus Marion County, Indiana, by IBRC patent class counts for 2017 in Table 5. Left to right, the table is organized by state and county alphabetically (otherwise known as FIPS codes) with Marion County on the far right because it ranks 20th in absolute patent making in 2017. This table shows that the leading patent-making counties in the Great Lakes region each have a much different technology portfolio. Cook County, Illinois, is particularly strong in communications, computers and IT, and consumer goods. Oakland County, Michigan, just outside of Detroit, is strong in a wide range of technologies: electronics, mechanical, transportation, building/construction, and power generation and industrial.

Table 5: Patent counts by technology classes for top 10 counties (plus Marion), Great Lakes states

Class title Total patents Cook, IL DuPage, IL Lake, IL Oakland, MI Washtenaw, MI Wayne, MI Hennepin, MN Washington, MN Summit, OH Dane, WI Marion, IN
Chicago Chicago MSA Chicago MSA Detroit MSA Ann Arbor Detroit Minneapolis Minneapolis MSA Akron Madison Indianapolis
Chemicals, including coatings, except pharma 796 75 51 36 39 29 281 59 91 87 34 14
Communications  694 205  69 107 95 61 1 69 21 23 30 13
Computers, information technology and data processing 956 240 79 85 87 90 3 189 53 26 89 16
Bio-tech and pharma 432 56 29 40 13 30 9 23 19 17 126 71
Electronics and electrical 571 68 52 37 124 75 24 93 30 25 21 23
Mechanical 1,735 176 90 105 236 133 526 179 109 75 48 57
Transportation, material transfer and storage 1,548 87 58 30 595 371 150 71 36 85 26 41
Agriculture and natural resources 174 18 13 14 7 3 56 25 15 4 15 4
Building, construction materials and methods 418 45 29 18 83 41 66 59 19 28 16 14
Power generation and other industrial, including armaments 497 41 33 21 117 98 36 59 24 24 22 21
Consumer goods, including furniture 755 116 41 61 99 53 172 99 56 20 25 16
Medical devices and medical practices 972 138 42 70 56 65 22 286 151 37 71 33
Design 1,359 271 88 112 264 34 156 114 56 154 68 44

Note: Highlighted cells indicate the two highest percentages for each technology class.
Source: IBRC, using 2017 data from the U.S. Patent and Trademark Office

Table 6 presents the percentage of patents by technology class based on a county’s total patents. The picture changes slightly as we consider relative concentration or specialization in technology types. Compared to patent counts, Cook County maintains its relative strength as a percentage of R&D output in communications, as well as computers and IT, but not in consumer goods. Oakland County is still strong in building/construction and power generation and industrial—and even more transportation-heavy relative to other technology classes. As expected, Marion County, Indiana, and Dane County, Wisconsin, have a strong bio-pharma presence. Summit County, Ohio, has relatively strong chemical and building/construction concentrations. Considering that Summit County’s largest city is Akron and the region has a long history of tire and rubber production, these specialties are not surprising. One might speculate that much of the material and chemical science that goes into tires can also be applied to membranes for roofs or for building foundation water barriers.

Table 6: Percentage of patent technology classes for top 10 counties (plus Marion), Great Lakes states

Class title Cook, IL DuPage, IL Lake, IL Oakland, MI Washtenaw, MI Wayne, MI Hennepin, MN Washington, MN Summit, OH Dane, WI Marion, IN
Chicago Chicago MSA Chicago MSA Detroit MSA Ann Arbor Detroit Minneapolis Minneapolis MSA Akron Madison Indianapolis
Chemicals, including coatings, except pharma 4.9% 7.6% 4.9% 2.2% 2.7% 18.7% 4.4% 13.4% 14.3% 5.7% 3.9%
Communications 13.4% 10.3% 14.5% 5.3% 5.6% 0.1% 5.2% 3.0% 3.8% 5.0% 3.5%
Computers, information technology and data processing 15.6% 11.7% 11.5% 4.8% 8.3% 0.2% 14.3% 7.7% 4.4% 15.0% 4.2%
Bio-tech and pharma 3.7% 4.3% 5.5% 0.7% 2.8% 0.6% 1.7% 2.8% 2.8% 21.4% 19.3%
Electronics and electrical 4.4% 7.7% 5.1% 6.8% 6.9% 1.6% 7.0% 4.4% 4.1% 3.6% 6.4%
Mechanical 11.4% 13.4% 14.3% 13.0% 12.3% 35.0% 13.5% 16.0% 12.4% 8.1% 15.6%
Transportation, material transfer and storage 5.6% 8.6% 4.1% 32.7% 34.2% 10.0% 5.3% 5.3% 14.1% 4.4% 11.2%
Agriculture and natural resources 1.2% 1.9% 1.9% 0.4% 0.3% 3.7% 1.9% 2.2% 0.7% 2.6% 1.0%
Building, construction materials and methods 3.0% 4.3% 2.5% 4.6% 3.8% 4.4% 4.4% 2.7% 4.6% 2.7% 3.8%
Power generation and other industrial, including armaments 2.7% 4.9% 2.9% 6.5% 9.1% 2.4% 4.5% 3.5% 4.0% 3.7% 5.7%
Consumer goods, including furniture 7.5% 6.0% 8.3% 5.5% 4.9% 11.4% 7.4% 8.2% 3.3% 4.2% 4.2%
Medical devices and medical practices 9.0% 6.3% 9.5% 3.1% 6.0% 1.5% 21.6% 22.3% 6.0% 12.0% 9.0%
Design 17.6% 13.0% 15.2% 14.5% 3.1% 10.4% 8.6% 8.2% 25.5% 11.5% 12.0%

Note: Highlighted cells indicate the two highest percentages for each technology class.
Source: IBRC, using 2017 data from the U.S. Patent and Trademark Office

An important takeaway here is that industry concentration as measured by employment is one way to gauge the presence of economies of agglomeration, but one may be able to also use patent technologies. Agglomeration economies, commonly known as industry clusters, indicate the co-location of related firms and industries that share a workforce, have mutual supply chains and enjoy knowledge spillovers. These technology data by county show similar patterns, albeit at a high level. The question then becomes: Why aren’t patent technologies baked into the descriptions and measures for how industry clusters develop?

Even this high-level gaze at the technology data has generated several questions that may warrant exploring. In the next section, we turn our attention to Indiana in particular. (This is, after all, the Indiana Business Review.)

Patent making in Indiana and select counties

In order to evaluate the value of these intellectual property and technology data, we focus on a map of Indiana and highlight three counties. As noted above, in the technology and patent-making competition with our Great Lakes neighbors, Indiana does comparatively poorly. That does not mean that Indiana is devoid of cultivating and developing technologies. Figure 5 shows county-level patents for Indiana and highlights our three selected counties: Allen, Kosciusko and Marion.

Figure 5: Map of Indiana total patents by county

Indiana map. 50 or more = 12 counties; 20 to 49 = 11 counties; 10 to 19 = 11 counties; 1 to 9 = 47 counties; Zero = 11 counties.

Source: IBRC, using 2017 data from the U.S. Patent and Trademark Office

Table 7 shows Allen County’s patent counts by IBRC technology class in five-year increments, 2002 to 2017. Figure 6 plots the percentages in the same time frame to get a sense of the shifts in relative importance of the technology classes. Allen County isn’t a patent-making powerhouse, but just the same, it appears that technology development has undergone several shifts. In Allen County, both transportation and mechanical technologies have declined in relative importance. Medical devices, as well as power generation and industrial, have enjoyed an upswing. It is important to remember that the total number of patents awarded to inventors in Allen County is modest, a mere 75 in 2017—a significant drop from 119 total patents in 2002. In the transportation technology category, there were 30 patents in 2002 but only 10 in 2017. It appears that the air is going out of patent making in Allen County.

Table 7: Patent counts by technology class, Allen County, five-year increments 2002 to 2017

Class title 2002 2007 2012 2017
Chemicals, including coatings, except pharma 4 1 1 1
Communications 5 2 2 6
Computers, information technology and data processing 2 2 9 2
Bio-tech and pharma 1 1 0 3
Electronics and electrical 9 2 6 8
Mechanical 21 9 12 5
Transportation, material transfer and storage 30 13 19 10
Agriculture and natural resources 5 2 1 1
Building, construction materials and methods 12 6 8 2
Power generation and other industrial, including armaments 9 2 6 11
Consumer goods, including furniture 7 5 5 3
Medical devices and medical practices 7 3 15 16
Design 8 16 5 8

Source: IBRC, using data from the U.S. Patent and Trademark Office

Figure 6: Patent percentage by technology class, Allen County, five-year increments 2002 to 2017

Line chart from 2002 to 2017 showing percentage of total patents for Allen County's 13 technology classes

Source: IBRC, using data from the U.S. Patent and Trademark Office

Table 8 shows Kosciusko County’s patent counts by IBRC technology class in five-year increments, 2002 to 2017. Figure 7 plots the percentages in the same time frame to get a sense of the relative importance of technology classes in Kosciusko County. In contrast to Allen County, the patent counts are on the upswing, with a total count of 27 in 2002 and 140 in 2017. This is attributed to what can only be described as a surge in medical device patents from 2002 to 2017. The five-year increments of the medical device totals jumped from 7 in 2002 to 57 in 2017. Mechanical patents also grew but to a lesser extent. The growth in percentage of the patent portfolio in Kosciusko County in these categories did not come at the expense of other patent categories. Most other categories were stable in patent counts as Table 8 shows. It is simply a matter of robust patenting growth of mechanical, consumer goods and medical devices.

Table 8: Patent counts by technology class, Kosciusko County, five-year increments 2002 to 2017

Class title 2002 2007 2012 2017
Chemicals, including coatings, except pharma 3 6 13 15
Communications 0 0 1 0
Computers, information technology and data processing 0 1 0 2
Bio-tech and pharma 0 0 1 3
Electronics and electrical 1 1 1 3
Mechanical 5 8 21 25
Transportation, material transfer and storage 1 2  1 2
Agriculture and natural resources 0 1 1 1
Building, construction materials and methods 1 0 2 3
Power generation and other industrial, including armaments 2 3 8 9
Consumer goods, including furniture 3 4 11 15
Medical devices and medical practices 7 13 44 57
Design 5 3 9 5

Source: IBRC, using data from the U.S. Patent and Trademark Office

Figure 7: Patent percentage by technology class, Kosciusko County, five-year increments 2002 to 2017

Line chart from 2002 to 2017 showing percentage of total patents for Kosciusko County's 13 technology classes

Source: IBRC, using data from the U.S. Patent and Trademark Office

Table 9 shows Marion County’s patent counts by technology class in five-year increments, 2002 to 2017. Figure 8 plots the percentages in 2002, 2007, 2012 and 2017 to observe changes in the relative importance of particular technologies. Total patents were up and down (down especially in 2007), but closed strongly in 2017.

Table 9: Patent counts by technology class, Marion County, five-year increments 2002 to 2017

Class title 2002 2007 2012 2017
Chemicals, incl. coatings, not pharma 11 11 11 14
Communications 11 15 15 13
Computers, IT & data processing 11 10 25 16
Bio-tech and pharma 51 29 37 65
Electronics and electrical 15 12 18 23
Mechanical 47 46 42 57
Transportation & material transfer 28 18 30 41
Agriculture & natural resources 15 8 10 10
Building/construction materials 9 4 9 14
Power generation & industrial 9 6 15 21
Consumer goods, incl. furniture 7 7 12 16
Medical devices & practices 34 19 40 33
Design 26 35 24 44

Source: IBRC, using data from the U.S. Patent and Trademark Office

Figure 8: Patent percentage by technology class, Marion County, five-year increments 2002 to 2017

Line chart from 2002 to 2017 showing percentage of total patents for Marion County's 13 technology classes

Source: IBRC, using data from the U.S. Patent and Trademark Office

The uptick in patent making cannot be ascribed to any one or two technology classes. Bio-tech and pharma, along with mechanical, closed 2017 stronger than in 2002, but bio-tech and pharma had a bumpy road, with significant dips in 2007 and 2012. Power generation and industrial, consumer goods, and design patents experienced a modest uptick, and together with transportation edged up the total patent count in 2017. In terms of measuring the shifts in the relative importance of one technology over the other, the picture is inconsistent. In the case of Marion County, patenting patterns may be more evident using patent counts in contrast to shifts in the portfolio balance using percentages.

Conclusion  

We presented the motivation and a method for measuring the concentration of technologies at a state and county level. We applied these methods to the Great Lakes states to assess the relative patenting performance of these states and their constituent counties. In several respects, this article is a trial run for a national data and visualization tool that may be of value to economic development practitioners (EDPs) who are searching for means to spark business and economic growth. Building on regional strengths is one way EDPs may use this tool to apply the TBED model.

It is our hope at the IBRC that our center and the websites we host—Hoosiers by the Numbers and StatsAmerica, for example—could provide greater technology detail to equip EDPs to focus on narrower technology classes that are strong in their region. Because such a tool would be built using patent data, complete with organizations and inventors holding the patents, EDPs and policymakers would be able to plot the networks of collaborators and resources for a type of technology within their region in which they have an advantage. In so doing, EDPs could help to combine the necessary ingredients for business formation, job growth, and rising profits, wages and standards of living.

We learned that Indiana is not a leader in patent/technology development compared to our Great Lakes neighbors. Among those neighbors, Marion County ranked only 20th in 2017 in total patent making (and that was a pretty good year for the county). We also saw a decline in patent making over the last almost 20 years in Allen County. This may forebode a longer-term, slow devolution from higher value-added jobs to lower-paying jobs in the region. At a workshop several years ago, I recall a person from Fort Wayne who accepted the notion that his town was patent-using, not patent-making. Given the productivity-enhancing benefits of the innovation that correspond with making patents, there may be good reasons to re-assert regional activities in research, development and engineering.

Patent making may not be the be-and-end-all of innovation, but patent and technology concentration and growth in a region suggests a brighter economic future, especially when the future is more technology dependent. Just ask the residents of Kosciusko County and its surrounding region.

Notes

  1. An ‘algorithmic links with probabilities’ crosswalk for USPC and CPC patent classifications with an application towards industrial technology composition by N. Goldschlag, T. J. Lybbert and N. J. Zolas (2016). Email reference available upon request.

References

  • Goldschlag, N., Lybbert, T. J., & Zolas, N. J. (2016). An ‘algorithmic links with probabilities’ crosswalk for USPC and CPC patent classifications with an application towards industrial technology composition. Working Papers 16-15, Center for Economic Studies, U.S. Census Bureau.
  • United States Patent and Trademark Office. (2021, May 1). Patent classification. www.uspto.gov/patents/search/classification-standards-and-development.
  • Hall, B. H., Jaffe, A. B., & Trajtenberg, M. (2001, October). The NBER patent citation data file: Lessons, insights and methodological tools (No. w8498). National Bureau of Economic Research.