The CHIPS and Science Act Will Boost Competitiveness and Promote Inclusive Growth

The United States is poised to sharpen its competitive edge in science and technology while addressing disparate economic growth across the nation’s geography, provided in part by the passage of the CHIPS and Science Act of 2022.¹ Properly implemented, the CHIPS and Science Act could jump-start innovation and ensure that the benefits of the science and high-tech industry are spread coast to coast, creating lasting advantages for regional economies.

A particular CHIPS and Science Act component will provide a major boost toward that goal: The Regional Technology and Innovation Hub Program authorizes $10 billion to establish at least 20 large-scale technology hubs across the country.² These hubs would invest millions of dollars into regional economies, catalyzing much-needed growth for smaller to midsized cities and neighboring rural communities, while bolstering the science and technology industry for years to come.

The legislation sets forth several requirements to ensure that the program benefits are geographically diverse and shared among rural communities. Beyond these requirements, however, the U.S. Department of Commerce and the National Science Foundation (NSF) will be tasked with determining the right locations for 20, or more, technology hubs. In this report, the Center for American Progress offers an additional set of factors to consider, which will ensure that the technology hubs are situated to maximize success while promoting strategic and inclusive growth across America. Specifically, CAP recommends that the awarding body consider metrics concerning education and skills, housing costs, and lifestyle factors such as commute times in certain regions, as well as preexisting research capacity.

This set of combined criteria led to CAP producing a short list of regions³ with underutilized capacity and that could thrive with an increase of federal stimulus. First, this list meets the requirements of the legislation. Second, metrics across education and skills, housing, and lifestyle demonstrate that these regions have the capacity to support an influx of funding and development needed to create successful technology hubs.

Organized by the six U.S. Economic Development Administration (EDA) regional offices, CAP recommends the below 30 localities as excellent locations for the placement of regional technology hubs. The five recommended areas within each EDA region all perform well on the aforementioned metrics. CAP believes each should be given equal consideration by the awarding body, without particular concern for the ordering presented here, in tandem with the specifics of their actual proposals.

Figure 1

The CHIPS and Science Act at large, but particularly the Regional Technology and Innovation Hub Program, has two principal aims: supercharge America’s science and technology sector and rectify what has been disparate economic growth in the United States. The legislation recognizes that there is a need to spur increased technological advancement and that, thus far, the innovation sector has been heavily concentrated in large cities, creating a spatial equity problem that hinders regional economies. The CHIPS and Science Act aims to simultaneously solve these two challenges.

Regaining or sharpening America’s competitive edge

The United States has historically been a global leader in scientific and technological advancements, but recently, it has started to lose ground. As other countries—particularly those in Asia, with China at the forefront—significantly increase investments in research and development (R&D) and make critical advancements, the United States has not kept pace. According to a recent report by the NSF, the United States’ global share in R&D has declined: From 2000 to 2019, the United States contributed 23 percent growth to global R&D performance, whereas China contributed 29 percent.⁵ This is at least partially attributable to the stagnation of government funding in the United States; the share of total R&D funded by the federal government declined from 31 percent in 2010 to an estimated 21 percent in 2019.⁶

This lack of R&D investment creates considerable vulnerabilities in the United States’ national and economic security efforts. The CHIPS and Science Act, a long-awaited and much-needed competitiveness and innovation package, is a step in the right direction to address these challenges. The package invests $54 billion to promote domestic semiconductor manufacturing capacity, which will decrease reliance on foreign production, free up supply chains, and improve America’s competitive edge. But the law also authorizes even larger investments in other scientific and technological areas critical for America’s long-term economic success, such as artificial intelligence and machine learning; quantum and other advanced computing; robotics and advanced manufacturing; climate solutions and new energy technologies; biotechnology and genomics; material science; and cybersecurity. The CHIPS and Science Act also bolsters educational opportunities in STEM, ranging from elementary to higher education, and promotes the advancement of underrepresented groups.

The law is a large-scale investment and, accompanied by other key transformative legislative accomplishments, most particularly the Inflation Reduction Act and the Infrastructure Investment and Jobs Act, represents an ambitious use of industrial policy strategy to boost U.S. competitiveness and domestic economic outcomes. Industrial policy is not a new tool, but its use fluctuates; it is most often employed in response to dramatic changes in the global economy or perceived international threats, most notably following periods of war. At the time President Joe Biden took office in January 2021, COVID-19 had heavily and adversely affected the global economy and international supply chains. In addition, the previous administration had ignored and worsened the climate change crisis, and U.S. manufacturing output was declining. Substantial strategic industrial investment can work to resolve these challenges, with the CHIPS and Science Act poised to transform the science and technology sector.

Some of this transformation is already visible. Micron Technology, for example, pledged up to $100 billion to build a new semiconductor facility in New York.⁷ Beyond these more immediate impacts, however, the policy goals of the CHIPS and Science Act will require significant implementation efforts to successfully reshape the science and technology industry and deliver broad economic benefits. Past industrial strategy efforts can offer lessons here. A recent report by the Peterson Institute for International Economics,⁸ which analyzes 50 years of U.S. industrial efforts, found that public and private R&D is more effective than several other industrial policy tools and can make the designated industry more competitive, create jobs, and produce important advancements. The report identifies the Pentagon’s Defense Advanced Research Projects Agency (DARPA) as the “model for frontier industrial policy.” DARPA awards grants in a range of categories and has a mandate to fund high-risk, high-reward R&D. Since its founding, it has created jobs, improved U.S. competitiveness, and contributed to a long list of achievements—including, as some argue, paving the way for the invention of the internet.⁹ With proper implementation, the CHIPS and Science Act and the Regional Technology and Innovation Hub Program could be as ambitious and impactful as DARPA.

Other research offers international comparisons and insightful analysis of localized, place-based policy interventions.¹⁰ For implementing the Regional Technology and Innovation Hub Program specifically, developing strong state partnerships, building upon preexisting capacity and infrastructure, promoting human capital development within the community, and ensuring careful coordination will all be essential for the program’s long-term success.

Addressing the geographic divide

Remedying the divergence between economically dominant, highly populous cities and economically stagnant, less populated regions by creating shared prosperity should be a fundamental goal driving federal economic policy decisions. And while the growth and prosperity brought by the rise of technology has been unequal and concentrated in a handful of cities, it does not have to continue this way. Federal policy can push back on this trend by establishing new technology hubs in geographically diverse areas in ways that both advance the industry and deliver benefits to smaller communities. The enactment of the CHIPS and Science Act represents a strong opportunity to meet this moment.

Over the past few decades, but especially in the decade following the Great Recession, economic growth in urban areas has far outpaced growth in rural areas.¹¹ This divergence has created a crisis of geographic imbalance as jobs concentrate in some of the densest urban areas, leaving many small cities and rural communities lagging behind.¹² This trend is partially due to the rapid growth of the technology industry, which has significantly altered the national landscape. While it has allowed for impressive technological advancements and generated substantial national wealth, it has also created an acute spatial equity problem. According to research from the Brookings Institution, the innovation sector has almost exclusively produced jobs in a handful of cities. From 2005 to 2017, five cities accounted for more than 90 percent of the nation’s innovation sector growth.¹³ This trend remains true even when considering overall employment growth—not just within the innovation sector: Since the financial crisis, the largest metropolitan areas accounted for 72 percent of the nation’s employment growth, while smaller cities and towns have contributed less than 6 percent.¹⁴

The CHIPS and Science Act is intended to address this gap by investing in geographically diverse regions. The legislation has the potential to bring essential economic and educational opportunities to regions that have long experienced economic stagnation. The enactment of the CHIPS and Science Act can help to reverse this trend and spread the benefits of the technology sector coast to coast. Beyond geographic diversity, however, it will be critical for the legislation to be implemented in a way that delivers broad benefits to historically marginalized and underserved populations. Research indicates that the technology sector has struggled to achieve a diverse workforce across race, gender, and socioeconomic background;¹⁵ therefore, it is critical that the Regional Technology and Innovation Hub Program is implemented in a way that addresses the lack of diversity and shares investments equitably, promoting an inclusive, prosperous economy.

The Regional Technology and Innovation Hub Program provides a critical opportunity to bolster the science and technology industry; and at the same time, it works to close the urban-rural gap and promote growth in smaller, regional economies. Although the emphasis must be on smaller regions, the awarding body should select localities that will allow for a technology hub to enjoy sustained success. Most likely, those will be localities with a decent-size population and enough infrastructure to support the development of the hub. The awarding body should account for certain factors that go beyond the minimum legislative requirements in order to strike a balance between promoting regional economic development and creating impactful technology hubs. Yet some small to midsized areas are likely to have the necessary ingredients—such as underutilized human capital and lower cost of living—for sustainable success while offering an attractive alternative to denser regions with a high cost of living. For example, Des Moines, Iowa, has a highly educated population—37.1 percent of the adult population has a bachelor’s degree or higher—and enjoys lower housing costs. This is comparable to Kansas City, Missouri, which is a much larger city¹⁶ with similar educational attainment rates—37.5 percent of the adult population has a bachelor’s degree or higher—but higher housing costs.¹⁷ Targeting these regions can also help to uplift surrounding rural communities, providing them access to increased educational and employment opportunities. It will help correct for the urban-rural divergence that has reshaped the national economic landscape and ensure that human talent is leveraged coast to coast.

Beyond the base size of a region, as captured by its working-age population, other factors will be critical for the success of a new technology hub. These include the region’s educational attainment rate, investment in STEM training programs, entrepreneurial ecosystem, and various quality-of-life indicators such as housing affordability and commute time that will be needed to attract and retain businesses and high-skilled human capital. If these necessary ingredients are in place, the placement of a new technology hub would likely succeed and start to close the gap between urban and rural prosperity.

Other researchers have arrived at similar conclusions. Jump-Starting America: How Breakthrough Science Can Revive Economic Growth and the American Dream, published in 2019 by Massachusetts Institute of Technology (MIT) professors Jonathan Gruber and Simon Johnson, argues for federally funded technology hubs to advance U.S. competition and spur regional economic growth.¹⁸ Gruber and Johnson developed an index that ranks localities across the country based on population, education, and lifestyle metrics. Their index closely aligns with the factors CAP identified as necessary for a regional technology hub’s success, and thus CAP’s analysis builds on their research.¹⁹

In addition to the above factors, CAP briefly examined preexisting research capacity within the recommended regions. This was done to meet the inclusion of key technology focus areas in the CHIPS and Science Act, which the law states “will enhance the competitive advantage and leadership of the United States in the global economy.”²⁰ The Regional Technology Hub and Innovation Program calls for the new technology hubs to relate to at least one of the following areas:²¹

1. Artificial intelligence, machine learning, autonomy, and related advances
2. High-performance computing, semiconductors, and advanced computer hardware and software
3. Quantum information science and technology
4. Robotics, automation, and advanced manufacturing
5. Natural and anthropogenic disaster prevention or mitigation
6. Advanced communications technology and immersive technology
7. Biotechnology, medical technology, genomics, and synthetic biology
8. Data storage, data management, distributed ledger technologies, and cybersecurity, including biometrics
9. Advanced energy and industrial efficiency technologies, such as batteries and advanced nuclear technologies, including but not limited to for the purposes of electric generation
10. Advanced materials science, including composites 2D materials, other next-generation materials, and related manufacturing technologies

Applicants to the Regional Technology Hub and Innovation Program do not necessarily need a preestablished program that excels in one of these areas. But metropolitan areas that have already received funding in these areas are worth examining. CAP briefly analyzed grants that were awarded by the NSF in 2019 and identified a few cities worth highlighting that were included in the recommendations:

• Pittsburgh, Pennsylvania, received approximately $25 million for research related to data storage, data management, distributed ledger technologies, and cybersecurity, including biometrics. The city also received approximately $15 million for research related to robotics, automation, and advanced manufacturing.
• Madison, Wisconsin, received approximately $20.4 million for research related to data storage, data management, distributed ledger technologies, and cybersecurity, including biometrics. The city also received approximately $9.8 million for research related to natural and anthropogenic disaster prevention or mitigation.
• Atlanta, Georgia, received approximately $10.5 million for research related to artificial intelligence, machine learning, autonomy, and related advances.
• Albany, New York, received approximately $9.2 million for research related to high-performance computing, semiconductors, and advanced computer hardware and software.
• Oklahoma City, Oklahoma, received approximately $4.2 million for research related to biotechnology, medical technology, genomics, and synthetic biology.

Over the past few decades, the technology industry has brought accelerated growth to the United States and has changed the landscape of the national economy—but that growth has not been equally shared. Thus far, the technology industry has been concentrated in a handful of large, densely populated metropolitan areas, while smaller and more rural economies have not had the opportunity to share in this growth. The CHIPS and Science Act represents a chance to realign and deliver key economic benefits to regional economies. Federal funding for new technology hubs can start to pull the concentration away from the largest, most economically dominant cities and promote the benefits of technology development and innovation for all Americans. This will help to spur regional economic growth and create a more inclusive national economy.

To be sure, the passage of the CHIPS and Science Act is just the first step. More work will need to be done to ensure that the provisions of the bill are properly funded and implemented. Moreover, regional economic policy needs to remain a priority in federal economic policymaking. Establishing 20 new technology hubs can be a pivotal shift in regional economies and, if successful, can offer lessons to future place-based economic innovation mechanisms.

The authors would like to thank our colleagues at the Center for American Progress, including Mara Rudman, Laura Rodriguez, Trevor Sutton, Marc Jarsulic, Jean Ross, Maggie Jo Buchanan, Nicole Ndumele, and Seth Hanlon, as well as Mark Muro at the Brookings Institution, for all their insightful comments and suggestions on this report.

Click here to download the full database. 

A.  Building the index

1. Preexisting index

As mentioned above, CAP drew inspiration from the index developed by professors Gruber and Johnson in their book Jump-Starting America.²³ Their research considers a total of eight variables that they examine across each metropolitan statistical area (MSA) across the contiguous United States. They then rank each variable and create category bundles to measure how an MSA performs across education, lifestyle, and population metrics.

CAP found their research to be compelling and built on their methodology, with some key changes. First, CAP updated the data where possible. Gruber and Johnson published their book in 2019, meaning that their data is now somewhat outdated. With the release of the 2020 census, CAP updated the variables to 2020, unless otherwise noted. Second, Gruber and Johnson measured housing affordability in each MSA with the variable “average house price.” This approach does not account for average income in that MSA. Instead, CAP replaced this variable with the “median home price-to-median income ratio,” produced by Harvard’s Joint Center for Housing Studies (JCHS). CAP’s change better measures the affordability of homeownership in a given locality. Take, for example, Napa, California: In Gruber and Johnson’s index, it is within the top five most expensive MSAs—and therefore down-ranked—because of its very high average house price ($759,725). However, ranking it instead on the JCHS’ ratio boosts it slightly, by five values, because average incomes are higher in Napa. Additionally, CAP recognizes that homeownership does not reflect the entire housing market and so added a new variable to reflect the affordability of rental units.

The final difference between CAP’s index and the Gruber-Johnson index is how the variables were bundled and weighted. This change was made to better reflect the goals of the legislation. For example, Gruber and Johnson gave the population variable a 33 percent weight, which rewarded larger cities. The CHIPS and Science Act was specifically designed to deliver benefits to smaller communities, so CAP significantly reduced this weight, to 10 percent. The below section details how CAP bundled and ranked each variable; to learn more about how Gruber and Johnson weighted and ranked each of their variables in their index see Jump-Starting America.

2. Variables

CAP’s index includes a total of 10 variables, which were collected by MSA. The U.S. Census Bureau defines MSAs as “a core area containing a substantial population nucleus, together with adjacent communities having a high degree of economic and social integration with that core.”²⁴ MSAs contain at least one urbanized area with a population of 50,000 or more, and there are 392 MSAs in the United States. However, CAP excluded from its analysis any MSAs that had a population of fewer than a 100,000, which left 354 MSAs for consideration.²⁵

The 10 variables are as follows:

1. Total population: Data are taken from the 2020 American Community Survey (survey number: S0101).²⁶
2. Working-age population: Data are taken from the 2020 American Community Survey (survey number: S0101).²⁷ This variable is calculated by subtracting the 65+ age category from the 21+ age category.
3. Commute time: Data are taken from the 2020 American Community Survey (survey number: S0801).²⁸ The authors collected data on the percentage of the sample size that spent fewer than 10 minutes commuting, 10–14 minutes, 15–19 minutes, 20–24 minutes, and 24–29 minutes. They then added them together to get the percentage that spent fewer than 30 minutes commuting.
4. Violent crime rate per 100,000 inhabitants: Data are taken from the 2019 FBI violent crime statistics.²⁹ Violent crime involves force or threat of force and includes four offenses: murder and nonnegligent manslaughter, rape, robbery, and aggravated assault.
5. Median-home-price-to-median-household-income ratio: Data are taken from the Joint Center for Housing Studies’ 2022 “The State of the Nation’s Housing” report.³⁰ The variable measures the ratio of the median sale price for existing single-family homes to the median household income for the year 2021. Income data for 2021 are based on Moody’s Analytics forecasts.
6. Rent burden: Data are taken from the 2020 American Community Survey (survey number: B25070).³¹ This variable measures gross rent as a percentage of household income in the past 12 months for renter-occupied housing units. CAP collected the data on the number of people paying 30.0 percent to 34.9 percent, 35 percent to 39.9 percent, 40 percent to 49.9 percent, and more than 50 percent of their household income on rent. CAP then added them together to get the total number of people paying more than 30 percent of their household income in rent and divided that by the survey sample size.
7. Percentage of adults with a bachelor’s degree or higher: Data are taken from the 2020 American Community Survey (survey number: S1501).³² This variable includes adults who are 25 or older.
8. Patents per inhabitant: Data are taken from the U.S. Patent and Trademark Office’s “Patenting In Technology Classes Breakout by Origin” table.³³ This variable measures the total number of utility patent³⁴ grants awarded from 2000 to 2015. It is organized by MSA based on the primary inventor’s place of residence. CAP divided the total number of patents granted per MSA from 2000–2015 by the total population of the MSA in the year 2015, using the 2015 American Community Survey.³⁵
9. Graduate programs: Data are taken from Jump-Starting America.³⁶ This variable captures the top 20 graduate science departments ranked according to the National Academy of Sciences from a 2005 survey.
10. Undergraduate programs: Data are taken from Jump-Starting America.³⁷ This variable measures the total number of undergraduates who subsequently graduated from a top 20 Ph.D. program, from 2005 to 2015. This is sourced from the NSF’s Survey of Doctorate Recipients.

3. Ranking

The variables were then ranked,³⁸ bundled into three categories, and weighted as follows:

1. Lifestyle

   Total weight: 30 percent

   1. Working-age population weight: 10 percent
   2. Commute time weight: 15 percent
   3. Crime weight: 5 percent
2. Housing

   Total weight: 30 percent

   1. Home-price-to-income ratio weight: 15 percent
   2. Rent burden weight: 15 percent
3. Education and skills

   Total weight: 40 percent

   1. Educational attainment weight: 20 percent
   2. Patents weight: 10 percent
   3. Graduate programs weight: 5 percent³⁹
   4. Undergraduate programs weight: 5 percent⁴⁰

Bundling and weighting allowed CAP to assign greater importance to certain factors. For example, the housing market is a significant concern in a region’s economy, and when seeking to attract and retain human capital, high levels of federal funding should avoid the most expensive areas that make cost of living more burdensome. Additionally, CAP weighted the education and skills bundle more heavily than the other two because the main objective of the CHIPS and Science Act would seem to be benefited by utilizing regions with high educational and entrepreneurial outcomes. Lastly, crime, graduate programs, and undergraduate programs all had data that were outdated or inconsistent in age across regions. Crime data, especially at the broader MSA level, may be useful to identify trends but may not be as predictive for a region’s ability to attract talent. Graduate and undergraduate STEM programs, although somewhat out-of-date, still may offer in smaller weights a complement to educational attainment and patent statistics. With these concerns in mind, CAP attached a smaller weight to these variables. Once each variable was ranked and weighted, CAP calculated the overall average rank for each MSA.

B.  Legislative requirements of the CHIPS and Science Act of 2022

The above metrics help to identify a relatively small pool of MSAs that would be best suited for a new technology hub. Beyond that, however, the legislation sets further requirements for the distribution of new technology hubs. To ensure that the technology hubs are geographically distributed, the CHIPS and Science Act stipulates several conditions for the implementation of the program.⁴¹ Citing from the legislative text, those requirements include:

• “(A) seeking to designate at least three technology hubs in each region covered by a regional office of the Economic Development Administration, while—
  • “(i) ensuring that not fewer than one-third of eligible consortia so designated as regional technology hubs significantly benefit a small and rural community, which may include a State or territory described in clauses (ii) and (iii);⁴²
  • “(ii) ensuring that not fewer than one-third of eligible consortia so designated as regional technology hubs include as a member of the eligible consortia at least 1 member that is a State or territory that is eligible to receive funding from the Established Program to Stimulate Competitive Research of the National Science Foundation; and
  • “(iii) ensuring that at least one eligible consortium so designated as a regional technology hub is headquartered in a low population State that is eligible to receive funding from the Established Program to Stimulate Competitive Research of the National Science Foundation;⁴³
• “(B) seeking to designate an additional two regional technology hubs based on selection factors which shall include likelihood of success and may include regional factors such as the extent to which the regional technology and innovation hub significantly engages and benefits underserved communities in and near metropolitan areas.”

Additionally, the bill lays out other metrics that the program should aim to meet, although are not necessarily required. Citing from the legislative text, those include:

• “(C) encouraging eligible consortia to leverage institutions of higher education serving populations historically underrepresented in STEM, including historically Black Colleges and Universities, Tribal Colleges or Universities, and minority-serving institutions to significantly benefit an area or region; and
• “(D) encouraging proposals from eligible consortia that would significantly benefit an area or region whose economy significantly relies on or has recently relied on coal, oil, or natural gas production or development.”

To determine the list of recommended geographic areas, CAP cross-walked the index with the above legislative requirements. First, to meet the requirement that at least three new technology hubs need to be placed in each EDA region, the list is organized by EDA office: Atlanta, Austin, Chicago, Denver, Philadelphia, and Seattle.

Second, to ensure that the list of recommended MSAs achieves geographic diversity and that the technology sector is spread coast to coast, CAP looked at areas that were already enjoying sustained success in the sector. The Brookings Institution has conducted analysis to this effect and produced a list of “superstar” metro areas, which they argue have propelled a “winner-take-most” effect on the innovation sector.⁴⁴ They define these areas as those that have the largest absolute number of jobs in innovation industries, and the top 10 are:⁴⁵

1. New York, New York
2. Los Angeles, California
3. Washington, D.C.
4. Chicago, Illinois
5. Houston, Texas
6. Boston, Massachusetts
7. Dallas, Texas
8. San Francisco, California
9. Seattle, Washington
10. San Jose, California

CAP decided to set a minimum mile radius of 120 miles⁴⁶ from these preexisting technology hubs to ensure that one region will not be highly concentrated with multiple technology hubs. On top of that, to foster greater geographic diversity in siting, CAP decided that each state should only have two recommendations, accounting also for the presence of preexisting technology hubs. Texas and California, then, already hit their max and are excluded from recommendations. Washington, New York, Illinois, and Massachusetts each have one and thus can receive only one recommendation. CAP believes that this adjustment best meets the goal of the legislation to promote growth in a geographically diverse way and ensure that federal funds are directed away from economically dominant cities. This will better allow the legislation to reduce the gap between urban and rural prosperity and for human talent to be leveraged across the country—not just in the pockets that have a preexisting talent pool.

Put together, CAP organizes the index by EDA region, removes MSAs that are within 120 miles of a preexisting technology hub, limits each state’s representation as described above, and recommends the five highest-ranked MSAs per region. Ultimately, this yields a list of 30 MSAs that may be best suited for a regional technology hub.

These MSAs are then cross-referenced with the legislation to ensure that they meet the necessary requirements. More than half of the recommended MSAs are located in an EPSCoR-eligible state. Nine of the MSAs are small and rural, with populations of fewer than 250,000; 12 MSAs are midsized, with populations of 250,000–749,000; and nine MSAs are large cities, with populations of more than 750,000. Although most of the recommendations fall in the midsize category, 30 percent of the list is small and rural, and the neighboring regions of midsized cities should not be discounted. These neighboring communities will also benefit from educational and employment opportunities, as measured in part by commute time, and as previously argued, setting a minimum threshold of 100,000 is important to ensure that there are enough people to support the development of the technology hub. Lastly, one of the recommended MSAs is in a low-population state (North Dakota).

C.  National Science Foundation research grants

To supplement this research, CAP briefly examined NSF grants that were awarded in 2019 that fit the descriptions of the key technology focus areas designated in the CHIPS and Science Act. This analysis exclusively focused on regions that were recommended in this report and did not examine any regions that were not promoted. This means that the analysis is not comparative: It is entirely possible that a highlighted MSA received less funding in a certain focus area than another MSA that was excluded from the search. CAP then categorized the grants into 1 of the 10 focus areas, and any grants that did not fit into the 10 categories were excluded. The analysis calculated the total amount of funds awarded in a given MSA across the 10 focus areas and highlighted the top one or two focus areas that were well funded by the NSF in 2019. Additionally, due to research limitations, the analysis only looked at 2019 and thus is not a comprehensive analysis. More cities across more years could be examined to offer a fuller analysis.