Home Finance Differences in Entrepreneurship Across the US: Can Industrial Policy Help?

Differences in Entrepreneurship Across the US: Can Industrial Policy Help?

by internationalbanker

By Dr. Swati Bhatt, Ph.D., Lecturer of Economics, Princeton University

 

 

 

 

Place-based industrial policy represents recent legislative efforts to promote growth in specific industries in selected regions. Together, the bills passed account for $3.8 trillion in total spending. These bills include the American Rescue Plan (ARP) of March 2021, the Infrastructure and Jobs Act (IIJA) of November 2021, the CHIPS (Creating Helpful Incentives to Produce Semiconductors) and Science Act of August 2022 and the Inflation Reduction Act (IRA) of August 2022.1

State-funded programs may have geographical implications, but targeted industrial policy is specific to certain sectors. The idea is to assist the local economy, “engaging with the local needs of individuals and industries and leveraging the ‘bottom-up’ energy of local talent, networks, clusters, institutions, and ecosystems”.2 The policy must consider the needs of regional and local communities, which may differ from a national perspective.

For instance, Title III(C), Section 3301, of the ARP of 2021 covers the State Small Business Credit Initiative (SSBCI), which allocates federal funds to support states’ small business initiatives.3 Businesses with fewer than 10 employees, which may include independent contractors and sole proprietors, are eligible for assistance. The funds may be used for legal, accounting and financial-advisory services, driving the growth of entrepreneurship in areas of particular concern.

Place-based industrial policy could smooth the variegated entrepreneurship landscape in the United States. My research on US startups, using data from the Business Dynamics Statistics (BDS) of the U.S. Census Bureau, suggests that states west of the Mississippi River are in the entrepreneurial vanguard, leaving high-density states such as New York in the tail. Defining startups as new establishments less than one year in existence with fewer than 100 employees, Bhatt (2022) found a trend of 1 percent per year in startup rates over the 1994-2022 period. This trend was negatively impacted by the 2008 financial crisis and the subsequent recession, when the unemployment rate peaked at 10 percent in October 2009, representing the worst economic downturn since World War II.4 (Stock and Watson wrote: Following the NBER-dated peak of 2007Q4, GDP dropped by 5.5% and nearly 8.8million jobs were lost.”5)

Splitting the 26-year stretch 1994–2020 into two structurally coherent periods, we define Period I from 1994 through 2007 and Period II from 2010 through 2020. In Period I, the growth rate of startups was 1.12 percent per annum, while in Period II, it increased to 4.3 percent. Period I followed the introduction of the personal computer, so we call it the “desktop digital revolution”, while Period II followed the introduction of the smartphone, so we named it the “mobility digital revolution”. See Figure 1.

The sharp upturn in the rate of startups during the mobility revolution (2010-20) reflects the hypothesis that it takes time to incorporate new inventions into trade and commerce, with multiple setbacks and barriers to smooth adoption. Daron Acemoglu and Pascual Restrepo’s research suggests a skills mismatch, as tasks displaced by automation are not yet complemented by task creation. Individuals who have lost jobs due to the application of new technologies to production processes do not have the necessary skill sets demanded by newer types of products and services.6 In the decade following the Great Recession of 2009, skill matching could have become more efficient due to the reshuffling of tasks as workers were being redeployed and suppliers and customers were brought in as partners in the matching process.

Data from the Business Formation Statistics (BFS) of the U.S. Census Bureau reinforces this surge in startup activity during the post-pandemic years. Individuals intending to start a new business must file for an Employer Identification Number (EIN) with the Internal Revenue Service (IRS). The number of seasonally adjusted business applications across all sectors jumped from 424,149 in September 2022 to 472,961 in September 2023, an increase of 11.5 percent in a year! The sector reporting the largest one-year increase in business applications was the retail trade sector, with a one-year increase of 36 percent. The healthcare and social assistance sector saw an increase of 14.5 percent, the educational sector increased by 13 percent, and the professional services sector jumped by 11.2 percent.7

Focusing on the pandemic years and using the same data (Business Formation Statistics), Haltiwanger found that after a decline in late March 2020 through May, new business applications started to surge” through May 2021.”8 New business applications, Haltiwanger discovered, predominated in Georgia, Florida and Texas but not in the eastern states, such as New York and New Jersey.

What is the geographic distribution of these digital revolutions, as manifested in entrepreneurial activity, across the United States? The startup growth rate across the United States can be represented by a multi-colored tapestry.9 (While my research defines startups by the number of new establishments less than one year in existence, with fewer than 100 employees, Karahan, Pugsley and Sahin examined the declining number of new employers as a fraction of all employers during 1979-2007.10 They found that this “startup deficit” can be attributed to sluggish labor-supply growth—the decline in labor supply impacts the extensive margin, the margin of entry of new firms or startups.) See Figure 2.

The Mississippi River is a significant geographical marker. States west of the Mississippi exhibit stronger entrepreneurial vitality than states east of it. (Since state laws and regulations play an important role in new business formation, I define the country by states rather than counties or metropolitan areas.)

Top-ranked states in startup activity across the entire period, 1994–2022, were North Dakota, Nebraska, Missouri, Idaho, Massachusetts, California, Texas, Wisconsin, Washington, Virginia, Nevada, Florida, Utah, South Dakota and South Carolina, in decreasing order. Ten of these fifteen states represent regions west of the Mississippi—only Massachusetts, Wisconsin, Virginia, Florida and South Carolina hail from the east. Interestingly, the fourth through tenth largest states by population—New York, Pennsylvania, Illinois, Ohio, Georgia, North Carolina and Michigan—are not included in the top 15 ranked by startup activity. See Figure 3.

Notably, the 15 weakest-performing states during 1994-2020 were Ohio, West Virginia, Arkansas, Mississippi, Indiana, Alabama, Michigan, Louisiana, Kansas, New York, Illinois, Minnesota, Maryland, New Mexico and New Hampshire, with Ohio being the weakest. All are east of the Mississippi.

Moreover, this geographical boundary is reemphasized when considering the second half of the 26-year span, the mobility digital revolution. During 2010-20, Idaho, Washington, Missouri, Wisconsin, Massachusetts, Nevada, Montana, Georgia, Utah, California, Texas, South Carolina, South Dakota and Colorado ranked among the top 15 states. Only four states are representative of regions east of the Mississippi.

State populations and population densities do not correlate with entrepreneurial activity. While 56 percent of the US population in 2022 resided east of the Mississippi, states with greater populations, such as New York andPennsylvania, are not among the highest-ranking in startup formations. Population density is also not predictiveof vibrant business cultures. Seven of the ten most densely populated states do not rank highly in business dynamism—all are east of the Mississippi. California and Texas, the two most populous states that together account for one-fifth of the American population (20.76 percent), are west of the Mississippi and among the top-ranking states in entrepreneurial activity.

However, population-growth rates strongly correlate with business formations: Six of the fastest-growing states also exhibit growing business communities: Idaho, Nevada, Utah, Texas, South Carolina and Washington. With the exception of South Carolina, these states are all west of the Mississippi. Importantly, the growth in these states is largely driven by Latinos, who account for a larger share of entrepreneurs in the mobility revolution compared with any other racial or ethnic group, including Asians, Blacks, Native Americans and Whites.11

While college education is an important explanatory factor, native skills acquired by experience and apprenticeship are also crucial: resourcefulness, practical intelligence, over-optimism and personal initiative. These skills are more likely to be correlated with a culture or belief system of risk-taking than with college degrees. Values such as collaboration, openness to new ideas, an awareness of the environment and the needs of people in one’s radius of interaction reflect a strongly embedded community whose values cannot be taught—they must be learned through experience at young ages.

What are the sectors that participated in this startup ecosystem? The professional and business services (PBS) sector, which includes technical services such as research and development, is the driver of not only the national economy but also the economies of the leading states in the mobility revolution. Measured by value added to national gross domestic product (GDP), the PBS sector has surpassed manufacturing since roughly 2012.12

The PBS sector began to dominate the economies (in terms of value added) of Massachusetts, California, Texas, Virginia and Georgia in 2015. Oil and gas extraction added the most value to the economies of North Dakota and South Dakota, while finance and insurance were strong in Nebraska. Manufacturing was dominant in Utah, Wisconsin, North Carolina and South Carolina, while real estate rental and leasing (RRL) was the principal driver of the economies of Missouri, Idaho, Nevada and Florida. RRL also includes industries such as professional employer organizations, which lease employees and manage all the human resources, payroll and employee-benefit functions for their clients. Employee leasing is a valuable organizational tool for firms with seasonal swings in their workforces. This format benefits small businesses that cannot provide worker benefits, such as unemployment insurance.

In Washington state, the information sector added the highest value to the state’s GDP. Information includes publishing industries (except software), motion picture and sound-recording industries, broadcasting and telecommunications, data processing, digital publishing and other information services. This sector has ranked second in California since 2020, after the PBS sector.

However, when one moves past nascent startups and considers mature startups, the picture changes. Defining mature startups as establishments that have existed between one and four years and have fewer than 500 employees, nascent startups exited at the rate of 0.38 percent per annum. Interestingly, states that represented rapid entrepreneurial activity also experienced the highest exit rates. For example, California, Missouri, North Carolina, Wisconsin, Massachusetts and Georgia were among the states with the highest number of establishment closings relative to openings.13

What could explain this?

Speed is a likely demon as startups look to expand their customer bases rapidly. The focus on growth at all costs suggests a misunderstanding of the concept of scaling, which is the response of economic organizations to changes in size. Young businesses are evolving systems, and the outcome of simply growing subscribers or customers is poorly understood.

Acquisition by larger, well-established firms is another likely reason. For example, Pfizer acquired the startup BioNTech in 2019, suggesting that the promise of distribution and marketing at scale could be compelling, in addition to sharing the associated administrative and labor costs.

Additional factors contributing to these exits include the decline in labor-force participation rates, increased mortality rates of white adults aged 35-54 and resistance to immigration.

Finally, the endurance of these new businesses beyond the early years requires timely counseling and networking. Founders may be passionate about their ideas but have little experience with execution. If recent legislative efforts could equalize startup rates and stem exit rates in states with high business creation, innovation and productivity could be enhanced.

To summarize, entrepreneurial activity was strong during the decade of 2010-20 in the US, but it has not been manifested in the densely populated financial centers in the Northeast.

Footnote

Dinlersoz et al. examined the spatial variation of entrepreneurship in the US over 2010-16 by focusing on startup rates defined by the number of new business applications (as a proxy for new ideas) multiplied by the fraction of those applications that turn into employee-hiring businesses within eight quarters of the applications. Twenty percent of the variation in startups is explained by education, income per capita, the employment-to-population ratio and the fraction of the local population that is foreign born. Curiously, they also found a negative relationship between startup rates and the fraction of African Americans, Asians and Hispanics in a local population.14

 

Note: This article has benefited greatly from data analysis by Melissa Woo, Princeton Class of 2024.

 

References

1 Brookings: “Breaking down an $80 billion surge in place-based industrial policy,” Mark Muro, Robert Maxim, Joseph Parilla and Xavier de Souza Briggs, December 15, 2022.

2 Ibid.

3 United States Congress: “American Rescue Plan Act of 2021,” H. R. 1319.

4 Entrepreneurship Today: The Resurgence of Small, Technology-Driven Businesses in a Dynamic New Economy, Swati Bhatt, 2022, Palgrave Macmillan.

5 National Bureau of Economic Research (NBER): “Disentangling the Channels of the 2007-2009 Recession,” James H. Stock and Mark W. Watson, May 2012, NBER Working Paper Series Number 18094.

6 Journal of Economic Perspectives: “Automation and New Tasks: How Technology Displaces and Reinstates Labor,” Daron Acemoglu and Pascual Restrepo, Spring 2019, Volume 33, Number 2, Pages 3-30.

7 United States Census Bureau/Business and Industry: Business Formation Statistics retrieved on 10/18/2023.

8 National Bureau of Economic Research (NBER): “Entrepreneurship During the COVID-19 Pandemic: Evidence from the Business Formation Statistics,” John C. Haltiwanger, June 2021, NBER Working Paper Series Number 28912.

9 Entrepreneurship Today: The Resurgence of Small, Technology-Driven Businesses in a Dynamic New Economy, Swati Bhatt, 2022, Palgrave Macmillan.

10 Federal Reserve Bank of New York: “Demographic Origins of the Startup Deficit,” Fatih Karahan, Benjamin Pugsley and Ayşegül Şahin, March 2021, New York Fed Staff Reports Number 888.

11 Entrepreneurship Today: The Resurgence of Small, Technology-Driven Businesses in a Dynamic New Economy, Swati Bhatt, 2022, Palgrave Macmillan.

12 Ibid.

13 Ibid.

14 Center for Economic Studies, U.S. Census Bureau: “The Local Origins of Business Formation,” Emin Dinlersoz, Timothy Dunne, John Haltiwanger and Veronika Penciakova, July 2023, CES Working Paper Number 23-34.

 

 

ABOUT THE AUTHOR
Dr. Swati Bhatt (Ph.D., Princeton University, 1986) has been on Princeton’s Economics faculty since 1992. Her research interests include the economics of digitisation and industrial organisation with a focus on the technology industry. Previously, she worked for the Federal Reserve Bank of New York and taught at the New York University Stern School of Business.

 

Related Articles

Leave a Comment

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.