Popular in Course
verified elite notetaker
Popular in Business
This 64 page Document was uploaded by an elite notetaker on Sunday December 20, 2015. The Document belongs to a course at a university taught by a professor in Fall. Since its upload, it has received 9 views.
Reviews for University-of-Chicago---What-Do-Small-Businesses-Do-
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 12/20/15
1 What Do Small Businesses Do? Erik Hurst University of Chicago email@example.com Benjamin Wild Pugsley University of Chicago firstname.lastname@example.org August 2011 Abstract In this paper, we show that substantial differences exists among U.S. small businesses owners with respect to their ex-ante expectations of future performance, their ex-ante desire for future growth, and their initial motives for starting a business. Specifically, using new data that samples early stage entrepreneurs just prior to business start up, we show that few small businesses intend to bring a new idea to market. Instead, most intend to provide an existing service to an existing customer base. Further, using the same data, we find that most small businesses have little desire to grow big or to innovate in any observable way. We show that such behavior is consistent with the industry characteristics of the majority of small businesses, which are concentrated among skilled craftsmen, lawyers, real estate agents, doctors, small shopkeepers, and restaurateurs. Lastly, we show non pecuniary benefits (being one’s own boss, having flexibility of hours, etc.)playafirst-orderroleinth e business formation decision. We then discuss how our findings suggest that the importance of entrepreneurial talent, entrepreneurial luck, and financial frictions in explaining the firm size distribution may be overstated. We conclude by discussing the potential policy implications of our findings. 1 We would like to thank Mark Aguiar, Fernando Alvarez, Jaroslav Borovicka, Augustin Landier, Josh Lerner, E.J. Reedy, Jim Poterba, David Romer, Sarada, Andrei Shleifer, Mihkel Tombak, Justin Wolfers and seminar participants at Boston College, the 2011 Duke/Kauffman Entrepreneurship Conference, the Federal Reserve Bank of Minneapolis, Harvard Business School, the Institute for Fiscal Studies, the 2011 International Industrial Organization Conference, London School of Economics, MIT, 2010 NBER Summer Institute Entrepreneurship Workshop, Penn State, Stanford University, and the University of Chicago for comments. Hurst and Pugsley gratefully acknowledge the financial support provided by the George J. Stigler Center for the Study of Economy and the State. Additionally Hurst thanks the financial support provided by the University of Chicago's Booth School of Business and Pugsley thanks the financial support from the Ewing Marion Kauffman Foundation. Certain data included herein are derived from the Kauffman Firm Survey release 3.1 public-use data file. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the Ewing Marion Kauffman Foundation. 1. Introduction Economists and policy makers alike have long been interested in the effects of various economic policies on business ownership. Infact,theU.S.SmallBusi ness Administration is a federally funded agency whose sole purpose is to help Americans “start, build, and grow businesses.” Researchers and policy makers often either explicitly or implicitly equate small business owners with “entrepreneurs.” While this association could be tautological, we show the typical small business owner is often very different than the entrepreneur that economic models and policy makers have in mind. For example, economc i theory usually considers entrepreneurs as individuals who (1) innovate and render aging technologies obsolete (Schumpeter, 1942), (2) take economic risks (Knight (1921); Kihlstrom and Laffont (1979); Kanbur (1979), and Jovanovic (1979)), or (3) are considered jacks-of-all-trades in the sense that theyhaveabroad skill set (Lazear, 2005). Policy makers often consider entrepreneurs to be job creators or the engines of economic growth. In this paper we shed light on what the vast majority of small businesses actually do and, further, what they report ex-ante wanting to do. The paper proceeds in six parts. We begin by highlighting the industrial breakdown of small business within the US. When referring to small businesses, we primarily refer to firms with between 1 and 19 employees. However, throughout our analysis, we also define alternative classifications such as firms with between 1 and 100 employees. As we show in this section, over two-th irds of all small businesses are confined to 2For example, recent academic work has evaluated the implications of various tax regimes on business formation See Cullen and Gordon (2007) and Cagetti and De Nardi (2009). Just recently, policy makers advocating legislation to overhaul the U.S. health care system in part justified the reform as promoting entrepreneurial activity and economic growth by “reducing the [health care] burden on small firms and their workers.” (U.S. Council of Economic Advisers Report (2009)) 3 Within theU.S.,twentypercentand thirtyfivepercen t of the private sector workforce works in businesses with between one and twenty employees and between one and one-hundred employees, respectively. In section 2, we also discuss the importance of non-employers. 1 a e r c b o j y d u t s e w , r e p a p e h i d s e t u b i r t t a h c r a e s e r g n i (1989), Hopenhayn (1992)), or differences in entrepreneurial ability of the firms owners (e.g., Lucas (1978)). In Section 4, we use new data whichsamplesnascent small business owners about their expectations for the business in the future to show that these stories are incomplete. When asked at the time of their business formation, most business owners report having no desire to grow big and no desire to innovate along observable dimensions. In other words, when starting their business, the plumber and lawyer do so while expecting to remain small well into the foreseeable future and with little expectation to innovate by developing a new product or service or even enter new markets with an existing product or service. If most small businesses do not want to grow ordonotwanttoinnovate,whydothey start? We address this question in Section 5. Again, we use a new data set that samples nascent business owners at the time they were starting their business that specifically asks about motives and expectations. We find that over 50 percent of new businesses reported that non pecuniary benefits were the primary reason as to why they started their business. Non pecuniary benefits included answers such as “wanting flexibility over schedule” or “to be one’sownboss”. By comparison, only 34 percent of respondents reported that they were starting the business to generate income and only 40 percent indicated that they were starting a business because they 5 wanted to create a new product or because they had a good business idea. Usingthepanel nature of the data, we show that those small businesses that started for other than innovative reasons were much less likely to subsequently grow, were much less likely to report wanting to grow, were much less likely to subsequently innovate, and were much less likely to report wanting to innovate. 5 The sum of the percentages exceed one hundred percent because respondentscould provide up to two reasons why they started their business. We discuss this data, the nature of the question, and other reported motivations in subsequent sections. 3 Collectively, these results suggest that there are other first order reasons why small businesses form aside from the innovation or growth motives which are embedded in most theories of entrepreneurship. For example, non pecuniary benefits of small business ownership may be an important driver ofwhy firms start and remain small. Additionally, some industries may have a natural size of production at an estab 7 agent). In Section 6 of the paper, we discuss how our results challenge much of the existing work on entrepreneurship and small firm dynamics. In particular, we highlight how our findings suggest that the importance of entrepreneurial talent, entrepreneurial luck, and financial frictions in explaining the firm size distribution may be overstated. discuss the policy implications of our results. The work discussing the diversity of motives and expectations among small businesses in developing economies is more extensive than for developed economies. Porta and Shleifer (2008) and Banerjee and Duflo (2011) show that most small businesses in developing economies do not grow or innovate in any observable way. In the latter sections, we also discuss how the qualitatively similar outcomes we observe are driven by different forces than in developing economies. w o h s s t l u s e r r u o , l l a r e v O within the U.S both in actual and expected growth and innovation behavior. Most small 6 The existence of non-pecuniary benefits as being important for small businesses has been suggested by Hamilton (2000) and Moskowitz and Vissing-Jorgensen (2002). Both papers find there is a compensating differential for small business ownership. We discuss these papers in greater depth in section 6. 7Furthermore, there may be interactions between these two motives in that those who receive large non-pecuniary benefits from small business ownership may gravitate towards industries where the natural scale of production is quite low. See Pugsley and Hurst (2011) for a formalization of this claim. 8TwonotableexceptionsincludeBhide(2000)andArdagnaandLusardi(2008). Bhide(2000)examinesthe attributes of the founders of many successful firms and concludes that the actions and behaviors of the founders are an important determinant of firm growth. Entrepreneurship Monitor (GEM) to show that there are demographic differences between those individuals who report starting a business because they had a good business opportunity or other business owners. t a h t s e s s e n i s u b l l a m s f o s e r a s e s s e n i s u b l l a m s , n w o n k l s u b l l a m s f o n o i t a r t n e c n o c l p m e l l a f o e s r e v i n u e h t g n i f o m u s e v i t a l u m u c e h t t r o f p o t e s s t u c e v i t a n r e t l a o t t s u b o r s t l u s e r e h t h t i w e s i r a y a m h t r o f t p e c x e 1 e l b a T o t s u o t a h t s i t h g i l h g i h o t h s i w e a l e r e h t t n e i r t s u d n i t a h t w o h s 3 e r u g i F i F t u o b a e d a m e b n a c s t n e m m o employers collectively represent less than 4 percent of all sales or receipts within the U.S. during 16 a given year. Becausemanyoftheexistingdatase ts, exclude the non-employers from their analysis, it is hard to systematically analyze their composition. Recently, however, the U.S. Census has released data that segments the non-employer firms both in numbers and receipts by broad industry classifications. 17 We summarize this data for 2007 in Appendix Table A1. The patterns documented in Tables 1 and 2 seem to carry through to non-employers. Most non- employer firms are in a handful of industries where the bulk of production takes place in small firms. As a result, we feel our broad results extend to the inclusion of the non-employer firms. The major take away from this section is that most small businesses are from a limited set of narrowly defined industries where most of the industries’ economic activity takes place in small firms. As we discuss in later sections of the paper, th ese industries usually do not match the theoretical models of "entrepreneurship" that is usually put forth in the literature. 3. Ex-Post Small Business Growth and Innovation A. Small Business Growth It is well documented that there is heterogeneity in the extent to which small businesses grow across observable factors such as firm size or firm age. Most recently, Haltiwanger et al. (2010) find, for example, that there is little relationship between firm size and firm growth conditional on firm age. Employment growth is driven by young firms, who also happen to be small. In this section, we use some new and existing data sets to illustrate some additional facts about the distribution of growth propensities across both small and young firms. Specifically, we show that even among young firms and conditional on survival, growth is still rare overall. 16Even though they are currently smathe non-employers are an important source of future paid employee firms. Many eventual employer firms start outs and non-employers. See Davis et al. (2007) for a more detailed discussion. 17See http://www.census.gov/econ/nonemployer/index.html. 12 u B 5 0 0 2 e h t m o r f a t a d w o h s b Second, similar to the results in the previous section, there is substantial variation among industries. Relative to construction, very little of employment of maturefirmsisinsmall businesses within the manufacturing industry (16 percent). Additional industries that include a high concentration of the employment of mature firms being in small businesses include the FIRE, wholesale trade, retail trade and service industries. Again, this is consistent with the results from Tables 1 and 2 and Figure 3. The heterogeneity in the firm sizedistributionacross sectors implies differences in dynamics by sector. To shed light on employment dynamics for firms of different ages and industries, we use data from a variety of additional sources. We start by using data from the 2003 Survey of Small 19 Business Finances (SSBF). The SSBF is a random sample of businesses with fewer than 500 employees and was conducted by the Board of Governors of the U.S. Federal Reserve. The survey is designed to measure the financial position of these businesses. However, the survey also contains other background questions. In 2003, firms were asked to state whether in the past year the total employees within their business grew, remained the same, or contracted. Firms were also asked the same question over a three-year horizon. The responses to these questions by small firms are shown in Table 4. Like above, we define small firms as those firms with fewer than 20 employees. We break down the responses by firm age to try to highlight differences between newer businesses and more established businesses. The SSBF asks businesses to report how long the business has be en in existence. As seen from the table, the overwhelming majority of small firms do not grow by adding employees 19The SSBF was formerly known as the National Survey of Small Business Finances. It was a quinquennial survey that began in 1983 and was last conducted in 2003. 14 20 year to year or even over three-year periods. Not conditioning on firm age, only 14 percent of surviving small businesses added an employee between 2002 and 2003 and only 21 percent added employees between 2000 and 2003. Taking the converse, roughly 80 percent of surviving small firms did not grow at all over a relatively long three year period. The percentages are slightly higher among newer firms. However, even among small firms which have been in existence between 1 and 10 years, only 19 percent grew between 2002 and 2003 and only 28 percent grew between 2000 and 2003. These data show that while most aggregate employment growth may come from small (new) firms growing big, the vast majority of small (new) firms do not grow, even over longer horizons. The SSBF data does show that while most firms do not grow at all over multiple years, some firms did grow. The SSBF data does not tell us by how much they grew. To assess this question, we turn to the Kauffman Firm Survey (KFS). The KFS is a panel study of 4,928 21 businesses that were newly founded in 2004 administered by the Kauffman Foundation. As shown in Haltiwanger et al. (2010), it is the new firms that contribute, on average, to job growth. Yet, as we have just shown, this is rare for the typical small businesses. While much employment growth is due to new firms, it is not true that most new businesses generate employment growth. To create the KFS sample, researchers began with a sample frame of nearly 250,000 new businesses started in 2004 provided by the Dun and Bradstreet database. From this data, the KFS oversampled businesses in high tech industries and businesses for whom research and development employment in the primary business industry was high. The final sample admits 20 We exclude newly founded firms that are unable to answer the employment change question because they did not exist in the base year. The firms responding to the 1 year changequestionareatleast1yearold,andthefirms responding to the 3 year change question are at least 3 years old. 21TheKauffmanFoundationisanorganizationwhosegoalsaretostudyandunderstandentrepreneurship. Information about the organization can be found at http://www.kauffman.org/. 15 4,928 firms, which are re-surveyed annually in follow up interviews. Currently, public use data is available on these firms up through 2009. For the work below, we only focus on those firms that have survived through 2008. There were 2,617 such firms in the data. When using the KFS data, we use the survey weights proved which are designed to make the firms in the sample representative of all new firms in the economy. Because the KFS is a four year panel, we can assess the growth rate of employment for new businesses within the KFS over four years. In each wave of the survey, the KFS asks firms to report the number of their employees. Column I of Table 5 shows that between 2004 and 2008, 41.9 percent of the surviving firms in the KFS reported growing the total number of employees within their business. In columns 2 and 3 of Table 5, we show the fraction of new surviving businesses who added more than 5 employees (column 2) and 10 employees (column 3) between 2004 and 2008. While about forty percent of the surviving new firms within the KFS added employees, very few added more than one or two employees. Specifically, 60 percent of all new firms in this sample did not add an employee, 90 percent added fewer than 5 employees, and 97 percent added fewer than 10 employees. The results from the KFS hold more broadly in the U.S. We find that industries important for small businesses (i.e, the ones documented in Tables 1) have lower than average job creation rates. To see this we pool employment change data from the SUSB from the years 2003 to 2006. These data are released as a companion to the levels reported in the SUSB annual data. Using the same administrative data, the Census Bureau measures the number of jobs created(eitherfrom expanding or new establishments) or destroyed (either from contracting or exiting establishments) at the establishment level and aggregates these into annual measures of gross job 16 r o r e h t e h w k s a o t s e t a r h t w o Zg x ▯ ▯▯▯ ▯▯▯ ▯ s jt 0 1 j j t jt Ms where g jt takes one of three different measures, depending on the regression, representing either the gross job creation rate, the gross job birth rate, or the gross job destruction rate for firms of small firms in industry j. These measures are define above. Likewise,asabove, xj represents the share of small businesses in industry i out of all small business across all industries. This measure is the same as what was summarized in Figure 2 and Table 1. Z jis a vector of industry level controls and ▯ is atvector of year dummies. The industry level controls include industry wide measures of gross job creation rate, the gross job birth rate, and the gross job destruction rate. The sample for this regression is all 4-digit industries with non-missing measures of M s during the 2003-2006 period. This gives us 929 observations for the small jt business gross job creation regressions, 666 observations for the small business gross job birth rate regression, and 656 observations for the small business gross job destruction regression. The difference is sample sizes is due to more missing data for the measures of births and job destruction relative to job creation at the 4-digit industry level. Table 6 reports the estimation results. We estimate each specification first where each industry is equally weighted and second where each industryisweightedinproportiontoits share of small businesses. The weighted estimation is similar to a grouped data estimator and would deliver the same point estimates as firm level data if each small firms employment share 23 within an industry were equal. Theresultssupportourearlierclaimsthatthe"typical"small business does not create jobs. The small business share of an industry has little to say about small business job creation through new small businesses or small business job destruction (columns 4 23Thisisareasonableapproximationsinceallfirmshavefewerthan20employees,sotherewouldbeverylittle variation in the employment share within an industry if this were estimated with the underlying administrative micro data. 18 and 5 of Table 6). However, it is a powerful predictor of weaker than average small business job creation for existing firms (columns 1-3 of Table 6). The most common small businesses (those with a high x sjown in Table 1) grow slower than average. These results hold even controlling for each industry’s overall characteristics (comparing column 1 vs. column 2 of Table 6). One may be concerned that that thedifference between the strong effects for job creation relative to the no effects found for job births and job destruction could stem from differences in the samples across the regression. Column 3 of Table 6 shows such concerns are unwarranted. In this regression, we restrict the job creation regression to only include observations that had non- missing job births and job destruction. The job creation results are unaltered with this additional sample restriction. According the to the weighted results, for each percentage point increase in the share of small businesses, an industry's small business job creation rate falls by a little less than three- quarters of a percentage point. To provide greater context, a one-standard deviation increase in x (1.1percentagepoints)reducesthejobcreati on rate by roughly 0.8 percentage points. The j average weighted job creation rate for the sample was 14.6 percent. So, a one-standard deviation increase in the industry share of smallbusinessesreduces thesmallbusinessjob creation rate by about 6 percent (0.8 divided by 14.6). When industries are treated equally, a one standard deviation increase in x jreduces the industry small bu siness job creation rate by roughly 8 percent. All the results are robust to alternative specifications of industry controls. Additionally, similar results hold if we re-estimate the equation with x▯jreplacing x. j It may initially be surprising that so little job creation comes from the industries that most small business owners are likely to enter. However it is consistent with an understanding of the important heterogeneity among small businesses. Mostsmallbusinesses(thosehighlightedin 19 Tables 1 and 2) start small and stay small throughout the life of their business. Collectively, we can conclude three things from the results in Tables 3-6. First, there is substantial skewness across firms in the extent to which they grow over time. While some firms do grow (in terms of the number of employees) over time, most do not. Onlyasmallportionofsmallfirmsadda more than ten employees over the life of their business. To this end, the bulk of employment in mature firms is still concentrated in firms with fewer than 20 employees. Second, even among new or young firms, most firms do not grow by any meaningful amount, even conditional on survival. Finally, a portion of theheterogeneityinemployme nt growth for small firms is explained by industry. While many mature businesses in manufacturing are quite large, most mature businesses in other industries like construc tion remain quite small. The industries that tend to remain small are the industries that tend to comprise the bulk of small businesses. B. Small Business Innovation In this sub-section, we document that there is also substantial heterogeneity across firms in the extent to which they successfully innovate along observable measures. Again, while some authors have shown that a large share of measured innovation (patent applications for example) is attributed to small businesses, the converse is not true.24 Mostsmallfirmsdonotseemto innovate along those observable margins. Before proceeding, we want to stress that it is hard to measure all aspects of potential small business ninovation via the surveys we are analyzing. As a result, we focus on some broad measures of innovation that are asked of firms within the surveys. 24See Acs and Audretsch (1990) and the cites within. 20 We begin by documenting the fact that very few new firms innovate via patent, trademarks, or copyrights during the first 4 or 5 years of their existence using two data sources. First, we continue our use of the Kaufman Firm Survey focusing on the same sample as above. The KFS survey asks respondents to reportseparatelywhetherthey havealreadyappliedorare in the process of applying for any patents, copyrights, or trademarks. We focus on the responses in 2008 when the firms have been in business for four years already. These results using the 2008 data from the KFS are shown in Table 7. Within the first four years of business, only 2.7 percent of the businesses in the sample had already applied or were in the process of applying for patents. Copyright and trademark usage is slightly higher but still most firms do not innovate at least according to these crude observable measures. According to the KFS, nearly 85 percent of small businesses did not acquire a patent, trademark or copyright during their first four years of existence. We augment our analysis using data from the Panel Study of Entrepreneurial Dynamics II (PSED). 25 The PSED started with a nationally representative sample of 34,000 individuals during the fall of 2005 and the early winter of 2006. An initial screening survey identified 1,214 "nascent entrepreneurs". To be considered a nascent entrepreneur, individuals had to meet the following four criteria. First, the individual had to currently consider themselves as involved in the firm creation process. Second, they had to have engaged in some start up activity in the past twelve months. Third, they had to expect to own all or part of the new firm. Finally, the initiative, at the time of the initial screening survey, could not have progressed to the point that it could have been considered an operating business. The goal was to sample individuals who were in the process of establishing a new business. 25There was an early wave of the PSED (PSED I) that was a test run for the bigger PSED II. We do not use the initial data in our analysis. All data and documentation for the PSED can be found at http://www.psed.isr.umich.edu/psed/data. 21 r a m e d a r t d n a , s t h g i r y p o c , s t company name without doing any real innovation. Wefocusfirstonthesemeasuresbecause they are easily observable in both the KFS and the PSED. The PSED, however, also has broader measures of innovation. In a separate set of question, businesses were asked directly whether they have "developed any proprietary technology, processes, or procedures". This is a slightly broader measure of innovation than patent, trademark and copyright applications in that it conceivably covers a more fluid set of activities that the business owner could relay about the innovation in production or business model that is takingplacewithintheirbusiness.Yet,only between 6 and 8 percent of new businesses (depending on the sample) reported than they had developed any proprietary business practices or technology during their first few years of business. Even conditional on survival five years later, 80 percent of firms still report not 26 developing any proprietary technology, process or procedure. The PSED asks one last broad question about the potential innovation taking place within the firm. This question asks about how the product or service produced by the businesses compares with the products and services of other producers within the market. Specifically, PSED respondents were asked: “Right now, are there many, few, or no other businesses offering the same products or services to your [intended] customers?” Respondents were allowed to provide one of the following answers: many, few, or noother. Thisquestionisinformativein the sense that it states whether the firm is providing a new product or service to existing customers or an existing product or service to potentially new customers. Across the three samples, between 36 and 43 percent of new business owners report providing a similar service to 26Weshouldbewaryofputtingtoomuchemphasisonselfreportsofinnovativebehaviorbysmallbusinesses. However, most behavior stories of how the business owners would respond to such questions would likely lead us to believe that the innovation numbers are upper bounds on actual behavior. This would occur if the respondents were more likely to report that theyre innovative even if there was no actual innovation taking place within the business. 23 an existing customer base as existing firms in the market. These businesses, more often than not, provide a standardized service (e.g., plumbing) to existing local customers. Conversely, Table 8 also shows that fewer than 20 percent of respondents reported that no one other business was provided their expected product or service to their expected customer base. There was substantial variation in the response to this question across business owners in different industries. For example, owners who reported starting a business in the professional, health, construction and real estate industries, were between 7.5 and 9.5percentage points more likely to report saying that they were staring their business in an area where there were many current providers of the service to their expected customer base. Owners in these same industries were nearly 10 percentage points less likely to report that they were providing a new product or service or were targeting an underserved customer base. 4. Ex-Ante Expectations About Growth and Innovation In this section, we document that many business owners have no expectation or desire to grow or innovate when they start their business. One of the strengths of the PSED data is that it asks the nascent business owners about their expectations for the business, theirdesiredfuturebusiness size, and for their motivations for starting the business. For example, all new firms were asked the following: “Which of the following two statements best describe your preference for the future size of this new business: ‘I want this new business to be as large as possible’ or ‘I want a size I can manage myself or with a few key employees’”. The top row of Table 8 shows the response to this question across our three different PSED samples. For the sample of those businesses who lasted to 2010, we report their expectations when they were first asked in 2006. Nearly three quarters of all respondents, regardless of sample, reported they wanted to keep their business small. 24 a f o g n i n a e m e t u o b a s k s a y e h t f i d e k s a o s l a e r e w D E S P n o i t s e u q e h t o t s e s n o p s e r w s r i f f o n o i t u b i r t s i d e h t s e t a h t s i 0 1 e l b a T m o r f r o l p x e e w , n o i t c e s s i h t f o r e s u a c e b s s e n i s u b r i e h t t r a t w e s o h t , 1 1 e l b a T n i n e i d d a w e f a t h g i l h g i h o t h s i r e h t t a h t w o h s n o i t c e s s i h t the results we document in the prior sections suggest that these stories are, at best, incomplete. It is not only differences in luck, talent, or credit market access are the only determinants of firm size. As we show above, there is also substantial ex-ante heterogeneity in the desires and expectations of new business owners with respect to their growth pr ocess. In other words, some firms do not grow or innovate simply because they do not want to grow or innovate. What drives these differences in ex-ante expectations and desires across owners of newly formed firms with respect to their desire to grow or innovate? The results in the prior sections point to at least two potential channels. First, many small business owners start their businesses, in part, because of the non pecuniary benefits associated with small business ownership. As seen from the PSED data, many small business owners report starting their business because they value the control and flexibility providedbysmallbusinessowne rship. If these benefits diminish with firm size, individuals who start for these reasons will prefer to keep their businesses small. We do find evidence of such correlations in the data: those business owners that report starting their business in part for non-pecuniary reasons were much more likely to want to keep their firm size small well into the future. Second, some businesses may stay persistently small because they are in industries which have low natural efficient scales. Many small businesses are dentists, plumbers, real estate and insurance agents, small shop keepers, and beauticians. Within these industries, the productivity of the firm is directly linked to theindividual'sskillset. Giventhefixedcostsofproduction may be small relative to the variable costs, optima l firm size may be quite small. As a result, 32 firms may start with no expectations of growth given that their natural scale is quite low. 31 These firms may be particularly attractive to business owners driven by non pecuniary motives. Pugsley (2011) and Pugsley and Hurst (2011) formalize the insights put forth in this paper by writing down models of small business formation and small business dynamics where individuals are allowed to have differential utility from small business ownership and industries differ in their natural returns to scale. In these models, they show that many of the predictions of the standard models of firm dynamics can be replicated in a model with no differences across firms in entrepreneurial ability and no difference across firms in their financing constraints. There are two important results from these papers. First, Pugsley and Hurst (2011) show that the existence of non pecuniary benefits can generate a positive relationship between wealth and starting a business, by making business ownership a normal good, where wealthier individuals “purchase” these benefits as their marginal utility of consumption diminishes. Second, Pugsley (2011) shows that there is not a one-for-one mapping between the distribution of firm size and productivity draws (like the ones emphasized in much of the literature outlined above) when industries differ in their fixed costs and owners haveapreferenceforkeepingtheirbusiness small, which cautions against using unconditional firm level dynamics to estimate a process for entrepreneurial productivity. Finally, much of the empirical work on firm dynamics proceeds by studying either the universe of firms, or focuses specifically on a sector thought to be representative of that universe. Typically, this is the manufacturing sector where micro-level administrative data have historically been the most available. It is in this empirical context where the applicability of 31This idea is consistent with recent research by Holmes and Stevens (2010) which attributes the variation in firm size within narrowly defined manufacturing industries to differences between large plants who produce standardized goods and small plants that make custom or specialty goods. 33 “Gibrat’s” law, which states firm growth rates are on average independent of size, or why the distribution of firm sizes appears to follow a particular power law(“Zipf’s”law)arefrequently demonstrated. Whytheseempiricalregularitiesappearat theaggregatelevelisaninteresting question. However, consistent with Pugsley (2011) it does not suggest that imposing this structure on a particular industry, or assuming a representative industry typified by manufacturing, is appropriate. The concentration of small businesses in industries varies considerably, and the heterogeneity we consider is especially important for industries we highlight in this paper. Preliminary work done bytheauthorssuggestthatZipf’slawdoesnot hold at detailed industry levels. There is considerable cross industry variation in the distribution 33 of firm sizes, even conditional on average firm size. B. Understanding the Risk-Return Tradeoff There is a separate literature assessing the risk-return trade off of small business owners. For example, Moskowitz and Vissing-Jorgensen (2002) document that the returns to investing in private equity (business ownership) are no higher than the returns to investing in public equity despite the poor diversification and higher risk. Their focus is only on the pecuniary returns of private business investment. This spans a large class of businesses, many of which are the small businesses we study here. However, even among venture-backed startups, which are a tiny fraction of small businesses, the risk-return tradeoff looks poor. Hall andWoodward(2010) perform a careful study of entrepreneurs backed byventurecapital,and find the risk adjusted return to entrepreneurs to be low relative to their outside options in paid employment. 32See, for example, Sutton (1997) and Gabaix (2008). 33For example, Figure 3 highlights these patterns at broad industry groupings. 34 Not surprisingly, a model with non pecuniary benefits can help to explain these findings. If there are private benefits to small business ownership (relative to allocating effort to the labor market), the measured pecuniary return could belower than the total return. Our results above suggest that for many individuals, non pecuniary benefits are an important motive for starting their small business. While the results above are based on survey reports, they are consistent with the work of Hamilton (2000) that shows the median small business owner receives lower 34 accumulated earnings over time relative to paid employment. Overall, our results suggest that for many individuals, non pecuniary benefits could be an important factor driving their small business formation. Incorporating such preferences into our models of small business formations can alter our assessment of the risk-return tradeoff of small business ownership. C. Misallocation or Motives There is also a growing literature about the within sectormisallocationofresourcesatthefirm level. Recently, in dynamic models calibrated to match US establishment level data, Moll (2010) and Midrigan and Xu (2010) have studied the effects of firm level financing constraints on aggregate TFP. In a similar setting, Restuccia and Rogerson (2008) find firm level wedges in marginal products and prices produce substantial e ffects on aggregate TFP. In empirical work, these firm level distortions appear to play a prominent role. Hsieh and Klenow (2009) estimate firm level wedges between marginal products and factor prices in India, China, and the US, which they use to estimate the aggregate effect on TFP. They find substantial (30 to 60 percent) 34Hamilton (2000) does not take into account of income underreporting by the self employed when performing his analysis. Hurst et al. (2011) show that such income underreporting by the self employed is important. Although, the results of Hamilton are mitigated when income underreporting is account for, it still appears that the median self employed individual takes a pecuniary earnings loss when becoming self employed. 35 o m t a h t t c a f e h t d e z i s a h p m e l l a m s t s o m t a h t g n i w o h s s t l u studied in the endogenous growth literature(e.g.Audretschet al. (2006), Acs and Audretsch (2009)). If a substantial portion of R&D occurs in small firms, the social returns to entrepreneurship could far exceed the private returns. Jones and Williams (1989), for example, find the optimal level of investment in R&D to be2 to 4 times the observe d level of investment. Additionally, subsidizing small businesses may be appropriate if liquidity constraintsorother financial market imperfections prevent small businesses from securing the financing they need to bring their innovations to market (Evans and Jovanovic (1989); Evans and Leighton (1989)). Given the belief that there are social spillovers from small business innovation or that small businesses face liquidity constraints, many developed economies enact policiesthatfavor small businesses relative to established firms. Within the United States, for example, small business subsidies include providing subsidized or guaranteed loans to small businesses, providing small businesses with access to special lending programs, exempting small businesses from various regulations, providing small businesses preferential treatment when awarding government contracts, and providing small businesses with preferential treatment through the tax 36 code. Whilepolicymakersand researchers often invoke the potential benefits of small business subsidies, very few discuss the costs. The results in our paper suggest that such costs may be nontrivial. The potential costs associated with small business activity come from two sources. First, as we show above, the bulk of small businesses report ex-ante that they do not want to grow nor do they want to innovate. And, as anticipated, most small firms do not grow or innovate. We show that these small business owners represent the bulk of the distribution of small businesses. Linking small business subsidies to firm size may support the handful of firms that eventually 36See De Rugy (2005) for a detailed discussion of the various ways the U.S. government provides subsidies to U.S. small businesses. 38 turn into Google or Microsoft, but they also stimulate real estate agents, small law firms, and construction workers where the social spillovers and growth potential is much smaller. Further, with substantial private rewards to innovative entrepreneurship (see for example Hall and Woodward (2003)), as well as the skewness we describe in the distribution of small businesses, our work suggests the marginal small business owner is more likely to be partofthelatter group. Second,wealsoshowthatnonpecuniary benefits are a first order reason small businesses form. The non pecuniary consumption flow from small business ownership adds an additional layer of potential distortion in response to small business subsidies. In a companion paper, Pugsley and Hurst (2011) illustrate the potential costs of small business subsidies in a simple general equilibrium static model of small business formation and occupational choice. Within the model, industries differ by their natural return to scale. Households differ by the size of the non pecuniary benefit they receive (in flow utility) from starting their own business. To highlight the potential costs of subsidizing small business activity in the model, we assume that there are no differences across individual in their talent, there are no social spillovers from small business formation and there are no liquidity constraints preventing firm formation. These extreme assumptions allow us to focus on potential costs of subsidizing small businesses in a world where individuals get non pecuniary benefits from small firm ownership. Individuals in the model can either allocate their labor to running a business or working for some other business. Household run businesses cannot grow to theirefficientscale without forfeiting the utility flow. 37Hall and Woodward (2003) show extremely fat tailed returns to venture capital investment in start up companies. Much of the return accrues to financial capital; in a follow up paper Hall and Woodward (2010) show the average risk adjusted return to the entrepreneur tooften be low relative to paid employment. 39 The model makes many predictions which can informresearchersandpolicymakers about the potential costs of small business subsidies. First, subsidizingsmallbusiness(funded with taxes on labor income) can distort the allocation of production within the economy to smaller scale businesses. Individuals choosing to start a small business trade-off the size of their non-pecuniary benefits from owning a small business with the loss in wages they would incur from foregoing the benefits of agglomeration. When small business activity is directly subsidized, the economy as a whole becomes less productive given the response of individuals to work in small (subsidized) self-owned firms as opposed to establishing larger firms which can produce at lower average costs by taking advantage of the returns to scale. Notice, such distortions could occur even in a world where there are no direct subsidies to small businesses if taxes are only levied on the pecuniary returns from working. Moreover, in such a model where the non-pecuniary benefits of small business ownership is a normal good, the subsidies to small businesses are highly regressive. The reason for this is that the high wealth individuals aregoingtobetheonesthatar e much more likely to start a business in the no subsidy world because they are more able to afford (in utility terms) the foregone benefits of agglomeration when they start their business. For the wealthy, the small business subsidy is simply a transfer tied to activity that they were more likely to do anyway. Not only do the existence of non pecuniary benefits to small business ownership result in subsidies being welfare reducing, lower wealth households suffer more from the subsidy than do higher wealth households. To our knowledge, there is no empirical work that evaluates whether subsidizing small businesses is a positive net present value venture. Our work, however, suggests subsidies would be most productively deployed to subsidize expansion rather than entry. This policy addresses 40 the concerns raised by our results in at least two ways. First, we show that most small businesses operate in industries with naturally small scales. Business owners with little intention to grow or innovate may select into these industries for that very reason. By focusing the subsidy on the intensive margin, the benefit is more likely to be taken up by a business owner focused on growth or innovative activity. Subsidies could lower the cost of credit for existing firms, and by increasing their value entice productive entrepreneurs with high wage employment opportunity costs. Second, if non pecuniary compensation is independent of the scale of the firm, the incidence of an expansion subsidy would be undistorted by non pecuniary benefits. If anything, non pecuniary benefits may help separate businesses that want to grow from businesses that would prefer to remain small. Of course there may be other social virtues to non-innovative small businesses, such as supporting communities and neighborhoods, which are aided by subsidizing the entry and exit margins. However, when targeting job creationor innovativerisk taking, our findings suggest caution when supporting businesses purely by size. At a minimum, future research is necessary to better understand both the costs and benefits of subsidizing small business activity. 8. Conclusion In this paper, we have shown that there is substantial skewness in the desires and expectations of individuals who start small businesses. Specifically, the vast majority of small business owners do not expect to grow, report not wantingtogrow,neverexpect toinnovatealongobservable dimensions, and report not wanting to innovate along observable dimensions. We also show that there is also substantial heterogeneity in the reported reasons for why individuals start their business. In particular, only about one-third of new businesses (on the eve of their start up) reported that they were starting their business because they have a product or service that they 41 n o m m o c s c i t s i r e t c a r a h c e s e h t and innovate. In fact, the US Small Business Administration already partners with venture capitalists whose high powered incentives are aligned with finding these small businesses with a desire to be in the tail of the firm size distribution. Lastly, we conclude that our results suggest that it isofteninapproprte for researchers to use the universe of small business (or self employment) data to test standard theories of entrepreneurship. Most small businesses dos uoc entrepreneurship which focuses on the desire to innovate or grow. Researchers interested in testing such specific theories of entrepreneurship may need to use more specialized data sets like the ones that track small businesses seeking venture capital funding. 43 References Acs, Zoltan and David Audretsch (1990). Innovation and Small Firms, MIT Press: Cambridge, Ma. Acs, Zoltan and David Audretsch (2009). y e"nl Entrepreneurship." Small Business Economics (32), 15-30. Audretsch, David, Max Keilbach, and Erik Lehmann (2006). Entrepreneurship and Economic Growth. Oxford University Press, Inc. New York, New York. Ardagna, Sivia and Annamaria Lusardi (2008). "Explaining International Differences in Entrepreneurship: The Role of Individual Characteristics and Regulatory Constraints", NBER Working Paper 14012. Asoni, Andrea (2011). "What Drives Entrepreneurship?", University of Chicago Mimeo. Banerjee, Abhijit and Ester Duflo (2011). Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty. PublicAffairs. Bhide, Amar (2000). The Origin and Evolution of New Businesses. Oxford University Press. Cagetti, M. and M. De Nardi (2009). "Estate Taxation, Entrepreneurship, and Wealth", American Economic Review, 99(1), 85-111. Clementi, Gian Luca and Hugo Hopenhayn (2006). "A Theory of Financing Constraints and Firm Dynamics", Quarterly Journal of Economics, 121(1), 229-65. Cullen, Julie and Roger Gordon (2007). "Taxes and Entrepreneurial Risk Taking: Theory and Evidence for the U.S.", Journal of Public Economics, 91(7-8), 1479-1505. Davis, Steven, John Haltiwanger, and Scott Schuh (1996). Job Creation and Destruction MIT Press. Davis, Steven, John Haltiwanger, Ron Jarmin, C.J. Krizan, Javier Miranda, Alfred Nucci, Kristin Sandusky (2007). "Measuring the Dynamics of Young and Small Businesses: Integrating the Employer and Nonemployer Universes." NBER Working Paper 13266. De Rugy, Veronique (2005). "Are Small Businesses The Engine of Growth?". American Enterprise Institute Working Paper #123. Evans, D. and B. Jovanovic (1989). "An Estimated Model of Entrepreneurial Choice Under Liquidity Constraints." Journal of Political Economy, 97(4), 808-. Evans, D. and L. Leighton (1989). "Some Empirical Aspects of Entrepreneurship", American Economic Review, 79(3), 519-35. 0 Executive Office of the President Council of Economic Advisers (2009). "The Economic Effects of Health Care Reform on Small Businesses and Their Employees", http://www.whitehouse.gov/assets/documents/CEA-smallbusiness-july24.pdf. Gabaix, Xavier (2009). "Power Laws in Ec onomics and Finance", Annual Review of Economics, 1, 255-93. Hall, Robert and Susan Woodward (2003). “Benchmarking the Returns to Venture.” NBER Working Paper 10202. Hall, Robert and Susan Woodward (2010). "The BurdenoftheNondiversifiableRiskof Entrepreneurship", American Economic Review, 100(3), 1163-1194. Haltiwanger, John, Ron Jarmin, and Javier Miranda (2010). "Who Creates Jobs? Small vs. Large vs. Young", NBER Working Paper 16300. Hamilton, B. (2000). "Does Entrepreneurship Pay? An Empirical Analysis of the Returns to Self Employment". Journal of Political Economy, 108(3), 604-31. Holmes, Thomas and John Stevens (2010). "An Alternate Theory of the Plant Size Distribution with an Application to Trade", NBER Working Paper 15957. Hopenhayne, Hugo (1992). "Entry, Exit, and Firm Dynamics in Long Run Equilibrium", Econometrica, 60(5), 1127-50. Hurst, Erik, Geng Li, and Benjamin Pugsley (2010). "Are Household Surveys Like Tax Forms: Evidence from Income Underreporting of the Self Employed", NBER Working Paper 16433. Jones, Charles and John Williams (1989). “Measuring the Social Return to R&D” Quarterly Journal of Economics, 113, pp 1119-1135. Jovanovic, B. (1982). "Selection and the Evolution of Industry", Econometrica, 50(3), 649-70. Jovanovic, B. (1979). "Job Matching and the Theory of Turnover", Journal of Political Economy, 87(5), 972-. Kanbur, S (1979). "Of Risk Taking and the Personal Distribution of Income." Journal of Political Economy, 87(4), 769-97. Kaplan, S. and J. Lerner (2009). "It Ain't Broke: The Past, Present, and Future of Venture Capital", University of Chicago Working Paper. Kihlstrom, Richard and Jean-Jaques Laffont (1979). "A General Equilibrium Entrepreneurial Theory of Firm Formation Based on Risk Aversion", Journal of Political Economy, 719-48. Knight, Frank (1921). Risk, Uncertainty, and Profit, Houghton Mifflin. 1 La Porta, Rafael and Andrei Shleifer (2008). "The Unofficial Economy and Economic Development", Brookings Papers on Economic Activity, Fall. Lazear, Edward (2005). "Entrepreneurship", Journal of Labor Economics, 23(4), 649-80. Lucas, Robert (1978). "On the Size Distribution of Business Firms", Bell Journal of Economics, 9(2), 508-523. Moll, Benjamin (2010), “Productivity Losses from Financial Frictions: Can Self-Financing Undo Capital Misallocation?” Princeton University, Mimeo. Moskowitz, Tobias and Annette Vising-Jorgensen (2002). "The Returns to Entrepreneurial Investment: A Private Equity Premium Puzzle?" American Economic Review, 92(4), 745 - 778. Pakes, Ariel and Richard Ericson (1989). "Empirical Implications of Alternate Models of Firm Dynamics", NBER Working Paper 2893. Pastor, Lubos, Lucian Taylor, and Pietro Veronesi (2009). "Entrepreneurial Learning, the IPO Decision, and the Post-IPO Drop in Firm Profitability", Review of Financial Studies, 22, 3005- 3046. Pugsley, Benjamin (2011), “Small Bus
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'