Fairlie, Fossen and Lyu, "A tale of two startups: the loss and gain of startups in the U.S. economy in the pandemic," Small Business Economics 66 (2026)
Summary
A research article by Robert Fairlie, Frank M. Fossen and Ke Lyu, 'A tale of two startups: the loss and gain of startups in the U.S. economy in the pandemic,' Small Business Economics 66:1665–1672, received 31 October 2025 and published online 16 January 2026. Using the Comprehensive Startup Panel, the authors give what they call the first official numbers on the pandemic's effect on employer and nonemployer startups. Nonemployer startup formation declined substantially in 2020 while employer startups unexpectedly increased; in 2021 nonemployer formation rebounded and almost offset the decline. Survival diverged: nonemployer survival dropped markedly while employer startup survival remained largely stable, and the contraction hit nonemployers with and without an Employer Identification Number alike.
Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used
Full text
Small Bus Econ (2026) 66:1665–1672
https://doi.org/10.1007/s11187-025-01171-7
RESEARCH
A tale of two startups: the loss and gain of startups
in the U.S. economy in the pandemic
Robert Fairlie · Frank M. Fossen · Ke Lyu
Received: 31 October 2025 / Accepted: 30 December 2025 / Published online: 16 January 2026
© The Author(s) 2026
Abstract The COVID-19 pandemic delivered an employer and nonemployer startups and underscore the
unprecedented shock to business entry, with sharply importance of designing policies tailored to nonem-
contrasting effects on different types of startups. Using ployer startups in economic crises.
newly constructed administrative data from the Com-
prehensive Startup Panel covering the universe of
U.S. startups, we provide the first official numbers Plain English Summary COVID-19 exposed a divide
of the pandemic’s impact on employer and nonem- between employer and nonemployer startups in the
ployer startup dynamics. Nonemployer startup forma- United States. Using new administrative data that track
tion declined substantially in 2020, while the number every startup, we show that nonemployer formation
of employer startups unexpectedly increased. In 2021, fell sharply in 2020 as employer startups unexpect-
nonemployer formation rebounded, almost offsetting edly increased. In 2021, nonemployers rebounded and
the 2020 decline. Survival outcomes also diverged: nearly offset the decline. Survival patterns diverged:
nonemployer startup survival dropped markedly dur- nonemployers were less likely to survive, whereas
ing the pandemic, whereas employer startup survival employer startups remained largely stable. Within
remained largely stable. Within nonemployer startups, nonemployers, more formal, growth-oriented ventures
the contraction was broad-based, affecting nonemploy- were hit as much as side businesses. The takeaway
ers with and without an Employer Identification Num- is that headline figures centered on employer startups
ber, an indicator of growth orientation. These find- miss large and different movements among nonem-
ings reveal a pronounced compositional shift between ployer startups. Policymakers and researchers should
recognize the distinct needs of nonemployer entrepre-
neurs and ensure that measurement, training, and pro-
R. Fairlie grams track and serve both types of startups to foster a
University of California, Los Angeles, CA, USA resilient, inclusive recovery.
R. Fairlie (B)
NBER, Cambridge, MA, USA Keywords Entrepreneurship · Startup · Employer ·
e-mail: rfairlie@ucla.edu Nonemployer · COVID-19 · Survival
F. M. Fossen · K. Lyu
University of Nevada, Reno, NV, USA JEL Classification L26 · E32 · M13
123
1666 R. Fairlie et al.
1 Introduction extended through 2021. Statistics on startup numbers
and survival that include the universe of business enti-
COVID-19 and the widescale policy response created ties tracked by the U.S. Census Bureau are not pub-
the largest sudden disruption to health, work, educa- lished. Instead, we generated these statistics using the
tion, and commerce possibly ever. Large losses to busi- restricted-access version of the CSP at UC Berkeley’s
ness activity, earnings, and sales have been documented Federal Statistical Research Data Center. During the
in survey, industry, and state administrative data (Bar- pandemic, we find that the U.S. economy missed out
tik et al., 2020a; Bohn et al., 2020; Crane et al., 2022; on roughly 700,000 business startups without employ-
Desai & Looze, 2020; Fairlie, 2020; Fairlie & Fossen, ees, and unexpectedly gained roughly 80,000 business
2022; Fairlie et al., 2023a; Farrell et al., 2020; Kim startups with employees in the first year of the pan-
et al., 2020). Historically, macroeconomic crises have demic.
been associated with increases in business closures and
declines in revenues, but impacts on new business for-
mation are less clear (e.g., Fairlie, 2013; Parker, 2018; 2 Methods
Perry, 2025). These patterns raised concerns that the
pandemic could lead to a collapse in business activity. We analyzed startup numbers and survival rates based
Early evidence confirmed a sharp contraction. In April on the universe of business entities tracked by the U.S.
2020, the number of active business owners in the USA Census Bureau using the Comprehensive Startup Panel
dropped by 22%—the largest decline on record (Fairlie, (CSP), a novel restricted-access administrative dataset,
2020). Business applications for new employer firms which was developed by Fairlie et al. (2023b) and can
fell by 27% in the 5 weeks from mid-March to mid- be accessed at the Federal Statistical Research Data
April relative to the previous year (Wilmoth, 2020). Centers. The CSP integrates two core U.S. Census
The full implications for business creation and Bureau sources: (i) the Longitudinal Business Database
startup survival, however, are not well known. Most (LBD), which covers all non-farm employer estab-
early indicators—such as business applications and lishments from 1976 onward, and (ii) the Integrated
surveys—captured intent and short-run activity, but not Longitudinal Business Database (ILBD), which cov-
the realized creation and survival of startups across ers nonemployer establishments (quinquennially from
the entire economy. In particular, a gap remains in 1977 to 1992 and annually thereafter). The integration
understanding how the pandemic affected different uses revenue activity to identify business births and pro-
types of startups. New businesses can take the form vides harmonized measures of job creation and survival
of employer startups (businesses that hire employees) for all startups. A key innovation of the CSP is that it
or nonemployer startups (businesses without employ- links the employer and nonemployer universes, allow-
ees). Nonemployer firms account for roughly 89% of all ing us to observe nonemployer startups and to capture
startups in the USA (Fairlie et al., 2025), yet published the pre-employment years of employer startups. This
statistics using the Longitudinal Business Database linkage yields more accurate startup birth dating and
(LBD) or Business Dynamics Statistics (BDS) focus enables each annual cohort to be followed longitudi-
only on employer businesses and ignore nonemployer nally,1
startups. Moreover, for employer startups, these statis- We report formation and survival for employer and
tics often misdate the true birth year because they do nonemployer startups. The CSP removes the concern
not account for a prior nonemployer phase. that any changes in the number of employer startups
Using newly created administrative panel data com- in other datasets are partly capturing transitions from
piled from underlying U.S. Census Bureau business nonemployer businesses. The data on employer star-
registry information, we provide the first official num-
bers on the impacts of COVID on startup formation and 1 One limitation of the CSP is that it does not provide demo-
survival that include all business entities. The Com- graphic information on startup owners. There might be interest-
prehensive Startup Panel (CSP) covers the universe of ing differences in who owns startups prior to the pandemic and
post pandemic (see, for example, Cowling and Dvouletý 2023).
startups in the U.S. economy and was only recently Future research could link the CSP to the new Census NES-D,
created to study startup job creation and survival from which includes demographic information on nonemployer busi-
1995 to 2018 (Fairlie et al., 2023b), and it has now been ness owners.
123
A tale of two startups: the loss and gain of... 1667
tups from the LBD refer to March of a given year, expected to have negative impacts on many busine-
whereas the data on nonemployer startups from the sses.
ILBD are based on tax records for a given calendar The fears of consumers, owners, and workers over
year. Although cases of COVID in the USA had occured catching the contagious virus at stores, restaurants,
earlier, the social distancing restrictions that shut down offices, warehouses, plants, and anywhere else people
many businesses did not start until the end of March interacted face-to-face were also a major concern in
2020. Therefore, we refer to 2020 as before the pan- the beginning of the pandemic. Additional strategies
demic for employers and as during the pandemic for to slow the spread of COVID beyond social distanc-
nonemployers. Survival by cohort is measured as year- ing restrictions that might have impacted businesses
to-year survival rates from year t-1 to year t conditional included but are not limited to mask mandates, vaccine
on survival until year t-1. rollouts, vaccine mandates for certain occupations and
Many nonemployer business entities are consult- enrollments, school closures, and remote learning.
ing, contracting or side-gig activities instead of more
growth-oriented startups. To address the potential con-
cern about overcounting the number of nonemployer
startups, in additional analyses, we follow Fairlie et
3 Results
al. (2023b) and remove all nonemployers that do not
have employer identification numbers (EINs) from the COVID led to both a major bust and boom in busi-
nonemployer counts. Nonemployers with EINs typi- ness formation in the country depending on the type of
cally plan to hire or incorporate in the future, so this startup (Fig. 1). Total startups initially contracted dur-
attribute can serve as an indicator of more formal and ing the pandemic, but the aggregate downturn masked
potentially growth-orientated startups. a divergence between nonemployer and employer star-
tups. The number of nonemployer startups plummeted
by 693,000 from 5,021,000 in 2019, the year before the
Timing of pandemic policies pandemic, to 4,328,000 in 2020, the first year of the
pandemic. Using average growth rates in prior years,
The novel coronavirus arrived in the USA in December the expected number of nonemployer startups lost off
2019 and was starting to spread rapidly in the next few prior trends is even larger at 855,000 (adjusted number).
months. The governmental response to the disease was However, the decline did not last. The U.S. economy
swift and included an effort to try to limit face-to-face experienced a major rebound in nonemployer startups
interactions especially in crowded events or spaces, the following year. In 2021, this number increased by
but the response mostly included voluntary or scat- 627,000, which nearly wiped out all of the losses in the
tered recommendations or restrictions. But, these early first year of the pandemic.
attempts to slow the spread of the new disease were not Next, we distinguish between nonemployer startups
enough. On March 16, 2020, the San Francisco Bay with and without EINs. We find that nonemployer star-
Area imposed the first social distancing restrictions that tups with EINs decreased by 265,000, while those
instructed residents to stay at home and not go to work, without EINs decreased by 428,000 in the first year
school, or stores if possible. By March 17, all states in of the pandemic (Fig. 2). Thus, even after exclud-
the USA had reported COVID cases, and most were dis- ing less growth-oriented nonemployers without EINs,
cussing whether to also implement shelter-in-place or there remains a substantial decline in nonemployer
social distancing restrictions. The states of California startup activity, highlighting that the pandemic also
and New York soon followed the Bay Area by impos- severely affected more formal and potentially growth-
ing state-wide shelter-in-place restrictions on March oriented nonemployer startups. By 2021, both EIN and
19, 2020 and March 20, 2020, respectively. By April 5, non-EIN nonemployers rebounded, with those without
2020, almost all, if not all, states imposed at least some EINs driving much of the resurgence.
form of shelter-in-place or social distancing restriction In contrast, startups with employees experienced
or recommendations. The mandated shutdowns were a surprising boom during the first year of the pan-
123
1668 R. Fairlie et al.
Fig. 1 Number of employer and nonemployer startups by year
Fig. 2 Number of nonemployer startups with and without EINs by year
123
A tale of two startups: the loss and gain of... 1669
Fig. 3 Annual conditional
survival rates by follow-up
year and startup cohort:
employer and nonemployer
startups
demic.2 As shown in Fig. 1, the number of employer cohort, the second-year conditional survival rate, which
startups increased by 77,000 from prior to the pan- is through 2020, was 67% compared with an average
demic to roughly 1 year after the pandemic, as there second-year conditional survival rate of roughly 79%
were 483,000 employer startups in March 2020 grow- for previous cohorts before the pandemic. For all previ-
ing to 560,000 employer startups in March 2021, rep- ous nonemployer startup cohorts, there is a clear drop in
resenting a 16% increase. The projected gain in the conditional survival rates in 2020 from pre-pandemic
number of employer startups adjusting for pre-trends levels.
is 100,032 (adjusted number). This surge is remark- This overall pattern remains consistent when we
able, as employer startup formation typically falls dur- separate survival rates for nonemployer startups with
ing severe economic contractions. and without EINs (Fig. 4). While both groups expe-
The pandemic also had a large impact on survival rienced sharp declines during the pandemic, the sur-
rates among new businesses started prior to the pan- vival rate dropped more significantly for nonemployers
demic. Figure 3 shows conditional year-to-year sur- with EINs. The conditional survival rate averaged over
vival rates by cohort. The U.S. economy lost many the six youngest cohorts fell for nonemployer startups
new nonemployer businesses due to COVID. Survival without EINs from 82% in 2019 to 77% in 2020. In
rates dropped precipitously for every startup cohort. comparison, the survival rate for nonemployer startups
For the cohort of nonemployer startups that began in with EINs declined from an average of 87% in 2019
2019, only about 45% survived into 2020. By compar- to 71% in 2020. This result highlights that even more
ison, roughly 55% of nonemployer startups survived formalized nonemployer businesses were substantially
1 year in pre-pandemic cohorts. For the 2018 startup affected by the pandemic.
Employer startups exhibited more resilient survival
2 Early studies found large negative effects of COVID-19 on a
patterns through the pandemic. COVID only reduced
wide range of business outcomes across a wide range of business
scales and data sources. The data sources included the Current
survival rates for the cohort starting just prior to the
Population Survey, Business Formation Statistics, the Census beginning of the pandemic; it did not reduce condi-
Small Business Pulse Survey, the Alignable business network tional survival rates across all cohorts (Fig. 3). The
survey, the Facebook business survey, Homebase data, Google 2020 employer startup cohort had a 86% survival
Places data, SafeGraph data, Yelp data, and Womply data (Alek-
seev et al., 2023; Bartik et al., 2020a, b; Bohn et al., 2020; Chetty
rate to 2021. All previous employer startup cohorts
et al., 2020; De Vaan et al., 2021; Fairlie, 2020; Kurmann et al., had a higher first-year survival rate averaging roughly
2021; U.S. Census Bureau, 2020; Yelp, 2020). 88%. Conditional survival rates from 2020 to 2021 for
123
1670 R. Fairlie et al.
Fig. 4 Annual conditional
survival rates by follow-up
year and startup cohort:
nonemployer startups (with
vs. without EINs)
previous startup cohorts are on par with similar follow- example, 55% of nonemployer startups typically sur-
up year survival rates prior to the pandemic.3 vive through the following year, but only 45% of 2019
nonemployer startups made it through the following
year.
4 Conclusion In contrast, the number of startups with employ-
ees increased from 483,000 to 560,000 (an increase
The importance of startups for economic growth, of 77,000 or 16%) in the first year of the pandemic.4
knowledge spillovers, and innovation is well docu- Survival rates among businesses starting with employ-
mented (Audretsch et al., 2006; Baumol & Strom, ees just prior to the pandemic dropped slightly, but for
2007; Kerr & Robert-Nicoud, 2020; Reynolds, 2007). all other cohorts of employer startups, the pandemic
We find that there was a “tale of two startups” in which did not reduce survival rates. Although the causes of
nonemployer startups were hammered by COVID, but the divergence of patterns between nonemployer and
employer startups were mostly unaffected. We docu- employer startups in the pandemic are largely unknown
ment that the number of nonemployer startups dropped and beyond the scope of this short descriptive paper,
from 5,021,000 to 4,328,000 (a loss of 693,000 or 14%) one potential cause requiring further study is the Pay-
in the first year of the pandemic but rebounded quickly. check Protection Program (PPP). The PPP, which was
In the second year, the number of nonemployer star- launched at the start of the pandemic, provided capi-
tups increased by 627,000, nearly covering the previ- tal to retain employees. The first round of PPP funds
ous year’s losses. We also find that the initial loss and was rushed out to small businesses with employees in
rebound show the same pattern when we focus on only April 2020, and funds were exhausted within 2 weeks
more formal and potentially growth-oriented nonem- (Mills, 2024; Rouse, 2024). The PPP likely reduced
ployer startups. Survival rates also dropped sharply closure rates among new businesses with employees,
in the first year of the pandemic for all cohorts. For but nonemployer startups did not qualify for PPP funds
in the first year of the pandemic. By early 2021, how-
3 The conditional survival rates of employer startups are slightly ever, changes were made to the PPP to include existing
larger in the first than in the second year in most cohorts, nonemployer businesses (U.S. Small Business Admin-
consistent with the liability of adolescence, which emphasizes
that newly founded organizations often have a stock of initial 4 This increase in the number of employer startups does not
resources, helping them to survive initially until these resources capture transitions from nonemployer startup to employer startup
are depleted (Fichman & Levinthal, 1991; Mahmood, 2000). in the CSP as it would in the BDS and LBD data.
123
A tale of two startups: the loss and gain of... 1671
istration, 2025). Future research is needed on the differ- References
ence by linking PPP loan receipt microdata to the CSP
data used here. The analysis here also only scratches Alekseev, G., Amer, S., Gopal, M., Kuchler, T., Schneider, J.
W., Stroebel, J., & Wernerfelt, N. (2023). The effects of
the surface, whereas the CSP offers the opportunity for COVID-19 on US small businesses: Evidence from owners,
more granular insights (e.g., sector-specific impacts, managers, and employees. Management Science, 69(1), 7–
regional variations, or the role of policy interventions). 24
Further research using the CSP could explore all of Audretsch, D. B., Keilbach, M. C., & Lehmann, E. E. (2006).
Entrepreneurship and Economic Growth. Oxford Univer-
these areas.
sity Press, 1st edition
Bartik, A. W., Bertrand, M., Cullen, Z., Glaeser, E. L., Luca,
Acknowledgements The authors are grateful to the Federal M., & Stanton, C. (2020a). The impact of COVID-19 on
Statistical Research Data Center in Berkeley, CA, for provid- small business outcomes and expectations. Proceedings of
ing access to the microdata. Any views expressed are those of the National Academy of Sciences, 117(30), 17656–17666
the authors and not those of the U.S. Census Bureau. The Cen- Bartik, A. W., Bertrand, M., Lin, F., Rothstein, J., & Unrath,
sus Bureau has reviewed this data product to ensure appropriate M. (2020b). Measuring the labor market at the onset of the
access, use, and disclosure avoidance protection of the confiden- COVID-19 crisis. Working Paper 27613, National Bureau
tial source data used to produce this product. This research was of Economic Research
performed at a Federal Statistical Research Data Center under Baumol, W. J., & Strom, R. J. (2007). Entrepreneurship and eco-
FSRDC Project Number 2936 (CBDRB-FY25-P2936-R11747 nomic growth. Strategic Entrepreneurship Journal, 1(3–4),
and CBDRB-FY25-P2936-R12020). Frank Fossen thanks the 233–237
Ewing Marion Kauffman Foundation for funding the research Bohn, S., Cuellar Mejia, M., & Lafortune, J. (2020). The
project “RG-202204-12283”. The contents of this paper are economic toll of COVID-19 on small business. Public
solely the responsibility of the authors. Policy Institute of California, https://www.ppic.org/blog/
the-economic-toll-of-covid-19-on-small-business/
Author contributions All three authors contributed to the Chetty, R., Friedman, J. N., Hendren, N., & Stepner, M. (2020).
manuscript with the analysis and writing. Real-time economics: A new platform to track the impacts
of COVID-19 on people, businesses, and communities using
Data Availability The administrative data used for this research, private sector data. Working Paper 27431, National Bureau
the Comprehensive Startup Panel, is provided by the U.S. Census of Economic Research
Bureau and available at Federal Statistical Research Data Cen- Cowling, M., & Dvouletý, O. (2023). Who is brave enough to
ters. Interested authors can apply to the Census Bureau for access start a new business during the COVID-19 pandemic? Baltic
to the data. Journal of Management, 18(3), 402–419
Crane, L. D., Decker, R. A., Flaaen, A., Hamins-Puertolas, A., &
Declarations Kurz, C. (2022). Business exit during the COVID-19 pan-
demic: Non-traditional measures in historical context. Jour-
nal of Macroeconomics, 72, 103419
Ethics approval and consent to participate Not applicable. De Vaan, M., Mumtaz, S., Nagaraj, A., & Srivastava, S. B. (2021).
Social learning in the COVID-19 pandemic: Community
Conflict of interest Frank Fossen thanks the Ewing Marion establishments’ closure decisions follow those of nearby
Kauffman Foundation for funding the research project “RG- chain establishments. Management Science, 67(7), 4446–
202204-12283”. We have no other declarations to make. 4454
Desai, S., & Looze, J. (2020). Business owner perceptions
Open Access This article is licensed under a Creative Com- of COVID-19 effects on the business: Preliminary find-
mons Attribution 4.0 International License, which permits use, ings. Ewing Marion Kauffman Foundation. Trends in
sharing, adaptation, distribution and reproduction in any medium Entrepreneurship, 10
or format, as long as you give appropriate credit to the original Fairlie, R. W. (2013). Entrepreneurship, economic conditions,
author(s) and the source, provide a link to the Creative Com- and the Great Recession. Journal of Economics & Manage-
mons licence, and indicate if changes were made. The images or ment Strategy, 22(2), 207–231
other third party material in this article are included in the article’s Fairlie, R. (2020). The impact of COVID-19 on small busi-
Creative Commons licence, unless indicated otherwise in a credit ness owners: Evidence from the first three months after
line to the material. If material is not included in the article’s Cre- widespread social-distancing restrictions. Journal of Eco-
ative Commons licence and your intended use is not permitted by nomics & Management Strategy, 29(4), 727–740
statutory regulation or exceeds the permitted use, you will need Fairlie, R., & Fossen, F. M. (2022). The early impacts of the
to obtain permission directly from the copyright holder. To view COVID-19 pandemic on business sales. Small Business
a copy of this licence, visit http://creativecommons.org/licenses/ Economics, 58, 1853–1864
by/4.0/. Fairlie, R., Fossen, F. M., Johnsen, R., & Droboniku, G. (2023a).
Were small businesses more likely to permanently close
in the pandemic? Small Business Economics, 60(4), 1613–
1629
123
1672 R. Fairlie et al.
Fairlie, R., Fossen, F. M., Johnston, A. C., & Lyu, K. (2025). The Parker, S. C. (2018). The economics of entrepreneurship. Cam-
effects of local taxes and incentives on entrepreneurship: bridge University Press, 2nd edition
Evidence from the universe of U.S. startups. SSRN Working Perry, A. M. (2025). Entrepreneurs’ resilience in New
Paper 5262559, https://doi.org/10.2139/ssrn.5262559 Orleans shined after Hurricane Katrina and the COVID-
Fairlie, R., Kroff, Z., Miranda, J., & Zolas, N. (2023b). The 19 pandemic, but small businesses now require sus-
promise and peril of entrepreneurship: Job creation and tained support. Brookings Data Center. https://www.
survival among US startups. MIT Press brookings.edu/wp-content/uploads/2025/09/Entrepreneurs
Farrell, D., Wheat, C., & Mac, C. (2020). Small business finan- -resilience-in-New-Orleans.pdf
cial outcomes during the onset of COVID-19. JP Morgan Reynolds, P. D. (2007). Entrepreneurship in the United States:
Chase Institute, https://www.jpmorganchase.com/institute/ The future is now, volume 15 of International Studies in
all-topics/business-growth-and-entrepreneurship/small-bus Entrepreneurship. Springer New York
iness-financial-outcomes-during-the-onset-of-covid-19 Rouse, C. E. (2024). Lessons for economists from the pan-
Fichman, M., & Levinthal, D. A. (1991). Honeymoons and demic. NBER Reporter, 3. https://www.nber.org/reporter/
the liability of adolescence: A new perspective on dura- 2024number3/lessons-economists-pandemic
tion dependence in social and organizational relationships. U.S. Small Business Administration (2025). New report
Academy of Management Review, 16(2), 442–468 shows consequential impacts of SBA pandemic relief.
Kerr, W. R., & Robert-Nicoud, F. (2020). Tech clusters. Journal https://www.sba.gov/article/2025/01/15/new-report-shows
of Economic Perspectives, 34(3), 50–76 -consequential-impacts-sba-pandemic-relief. Archived
Kim, O. S., Parker, J. A., & Schoar, A. (2020). Revenue collapses News Release 25-26
and the consumption of small business owners in the early U.S. Census Bureau (2020). Small Business Pulse Survey.
stages of the COVID-19 pandemic. Working Paper 28151, https://www.census.gov/data/experimental-data-products/
National Bureau of Economic Research small-business-pulse-survey.html. Accessed: 2025-04-09
Kurmann, A., Lale, E., & Ta, L. (2021). The impact of Wilmoth, D. (2020). Small business facts: Early data
COVID-19 on small business dynamics and employ- show severe disruptions. US Small Business Administra-
ment: Real-time estimates with homebase data. SSRN tion. https://advocacy.sba.gov/wp-content/uploads/2020/
Working Paper 3954262. https://doi.org/10.2139/ssrn.3954 04/Small-Business-Facts-Early-Pandemic-Data.pdf
262 Yelp (2020). Local economic impact report. https://www.
Mahmood, T. (2000). Survival of newly founded businesses: A yelpeconomicaverage.com
log-logistic model approach. Small Business Economics,
14(3), 223–237
Mills, K. G. (2024). Fintech, small business & the American Publisher’s Note Springer Nature remains neutral with regard
Dream: How technology is transforming lending and shap- to jurisdictional claims in published maps and institutional affil-
ing a new era of small business opportunity. Palgrave iations.
Macmillan, 2nd edition
123
File and source
- File
- fairlie-2026-a-tale-of-two-startups-the-loss-and-gain-of-startups-in-the-u-s-economy-in-th.pdf
- Size
- 1,093,963 bytes
- SHA-256
- d05523d048251f274037fd41697d6d8c778dd97dd36b3f335f426fd87d9d2f80
- Our copy
- fairlie-2026-a-tale-of-two-startups-the-loss-and-gain-of-startups-in-the-u-s-economy-in-th.pdf
- Original
- link.springer.com