Financing through leasing: Evidence from the Kauffman Firm Survey

Transcription

1 Financing through leasing: Evidence from the Kauffman Firm Survey Carmen Cotei Associate Professor of Finance University of Hartford Joseph Farhat Professor of Finance Central Connecticut State University 2015 FMA Annual Meeting Orlando, FL October 15, 2105 Certain data included herein are derived from the Kauffman Firm Survey restricted-access data file. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Ewing Marion Kauffman Foundation. 1

2 Financing through leasing: Evidence from the Kauffman Firm Survey Abstract We analyze the impact of firm s asset uniqueness and its growth opportunities on the leasing decisions of U.S. startups. To test our hypotheses, we use a unique dataset provided by the Kauffman Foundation. Our results show that startups in the high-technology sector have a lower propensity to lease whereas startups with high R&D intensity have a higher propensity to lease. In addition, we examine the impact of owners characteristics on the decisions to lease and find that female and older entrepreneurs as well as highly educated owners are less likely to lease. Our findings confirm previous studies on leasing decisions that larger and more leveraged businesses have a higher propensity to lease their assets. Consistent with Ang and Peterson (1984) and Bathala and Mukerjee (1995) we report that leasing and debt are complements rather than substitutes to each other. Keywords: leasing, startups, asset specificity, R&D intensity, survey methodology JEL classifications: C83, G32, M13, O31, O32 2

3 1. Introduction The Equipment Leasing and Finance Association (ELFA) reports that the market size for equipment/asset acquisition is projected to be $1.48 trillion in 2015, of which $922 billion will be either leased or financed (July 21, 2015). Leasing offers very flexible terms and customized options to businesses in order to match the proposed capital budgeting with cash-flow patterns emerging from financial statements. Leasing is especially important to small, young businesses that typically have no credit rating, very little access to banking finance and/or venture capital funds. Due to severe informational asymmetry between small business owners and outside investors, small businesses have no access to public equity and/or debt markets. Leasing is an alternative to bank loans to finance the firms assets. Contrary to a bank loan, when firms lease an asset, the lessor remains the owner of the asset. Because of the ability to repossess, a lessor can extend more credit than a lender whose claim is secured by the same asset. Thus, leasing has a higher debt capacity than secured debt, which is very valuable to small, financially constrained firms. The separation of ownership and control of a leased asset gives rise to agency costs that have to be traded off against the benefits of higher debt capacity (Einsfelt and Rampini, 2009). The higher debt capacity of leasing predicts that financially constrained firms will lease the asset while less constrained firms buy the asset and borrow against it. In this study, we analyze the impact of asset uniqueness and growth options on leasing decisions in 1,630 U.S. startup businesses included in the Kauffman Firm Survey, the largest longitudinal data for newly formed firms. Unlike previous studies that analyze the leasing decisions of publicly traded firms, we focus on small, privately owned businesses that have higher probability of failure and are financially constrained. The literature on the use of leasing in the U.S. and Europe has shown that leasing decisions 3

4 are a function of firm size, leverage, probability of bankruptcy, profitability, ownership structure, investment opportunity set and tax variables (Einsfelt and Rampini, 2009; Adams and Hardwick, 1998; Lasfer and Levis, 1998; Sharpe and Nguyen, 1995). Most of the previous studies focus on medium size and large firms, neglecting small firms and young, newly established businesses. In this paper we focus on startups established in 2004 and analyze their leasing decisions since inception until 2011, the last year Kauffman Foundation surveyed these firms. Our major goal is to examine the impact of asset uniqueness and growth opportunities as well as the impact of owners socioeconomic and demographic characteristics on the propensity to lease. This study will contribute to the existing literature in two important ways: first, we provide evidence on the leasing decisions of very small businesses in the U.S. Our results show that firm characteristics such as asset uniqueness and growth opportunities are especially important for leasing decisions of startups. Second, we analyze for the first time how owners attributes affect the leasing decisions of startups in the U.S. Our findings show that some characteristics such as owners gender, education level and age have a significant impact on the propensity to lease. In addition, our results confirm prior findings on leasing decisions of large firms, as highly leveraged and larger firms are more likely to use lease financing. Also, firms in certain industries (agriculture, construction) tend to lease more, while in some other industries (finance/insurance/real estate) firms tend to lease less. The rest of the paper proceeds as follows: Section 2 provides a literature review and develops the hypotheses. Section 3 describes the Kauffman Firm Survey and reports the descriptive statistics. We provide the empirical results in section 4 and conclude the paper in section 5. 4

5 2. Literature review and hypotheses development Leasing represents a financial contract that gives entrepreneurs the flexibility to use an asset without the burdens of purchasing that asset; therefore, it allows business owners to preserve cash. Since most startups are financially constrained in the first few years of their activity, bootstrapping finance, or the ability to use various resources to generate sales, is crucial to business owners who try to develop and grow their ventures. In this study we focus on the role of asset uniqueness and growth opportunities in explaining the leasing decisions of U.S. startup firms. High-technology businesses are a special subset of startups engaged in securing growth and revenue from industry sectors characterized by new and rapidly changing technology. According to Inc. Magazine, advanced technology has come to be utilized in so many different industries that members of the business community now often regard it as its own unique industry subset. The unique feature of high-tech businesses is that their assets are predominantly intangible, i.e. intellectual property (IP) rights, new technologies that are unique to the startup and has very little value to other businesses. Smith and Wakeman (1985) show that specialized assets are less likely to be leased due to the fact that these assets are highly valuable to a particular user but have little or no value to alternative users. Since specialized assets cannot easily be repossessed, the asset uniqueness feature of high-tech startups may explain their propensity to lease among all startups. Therefore, we hypothesize the following: H1: Startups with unique assets should lease less. To proxy for asset uniqueness we use a dummy variable that takes value of 1 if the startup operates in a high-tech sector and 0 otherwise. An alternative proxy is the presence of intellectual property rights (IP). In the annual survey, entrepreneurs were asked whether they possessed any type of IP: 5

6 patents, trademarks and/or copyrights. We construct a dummy variable that takes the value of 1 if the startup had any type of IP rights in a given year, and 0 otherwise. Some startups are characterized by high growth opportunities in that they focus on research and development (R&D) activities to help them grow at a faster pace and ultimately be successful. Financing the R&D expenses is very difficult as many small businesses generate very little sales or no sales during the first few years of launching a new product. As a result, startups with high R&D expenses are less likely to secure debt financing. Many banks or other providers of debt capital may see these firms as too risky and too informationally opaque. As a result, for a given debt ratio, high R&D startups may use leasing more than other startups. Hall (2002) examined the financing of research and development and concluded that small and new innovative firms experience high costs of capital that are only partially mitigated by the presence of venture capital. In addition, Hall documented that large firms experience high costs of R&D capital and therefore, these firms prefer internal funds for financing these investments. Since leasing contract is considered similar to secured debt, we argue that small firms eliminate the underinvestment problem (Myers and Majluf, 1984) by using leasing instead of debt financing. Therefore, we hypothesize the following: H2: Startups with high growth opportunities should lease more. To proxy for growth options we construct a dummy variable that takes the value of 1 if the entrepreneurs reported that at least one employee was designated for R&D activities in a given year. The survey does not reveal the actual expenses made for R&D in a particular year; therefore, our variable is R&D intensity (or incidence) rather than R&D expenses. A notable difference between startups and publicly traded firms is that startups are ownermanaged and therefore the owners demographic and socioeconomic characteristics are likely to 6

7 impact the individual risk-taking behavior of each owner. In the case of small firms, probability of default and financing constraints depend not only on firm characteristics but also on factors such as owners experience, education, age, gender, and race. In the psychology literature (McInish, 1982; Bertrand and Schoar, 2003), it is generally believed that risk taking behavior decreases with age and increases with education, income, wealth, experience and sophistication. With respect to gender, male entrepreneurs tend to be more risk tolerant than female counterparts. The literature on loan access documents that owners characteristics impact the access to loans as well as loan terms. While the evidence on gender discrimination in the loan market is mixed (Cavalluzo et al., 2002; Blanchflower et al., 2003), there is clear evidence on discrimination in the loan market based on ethnicity (Bruder et al., 2011, Cavalluzo and Wolken, 2005) and age (Reifner, 2005). In addition, experienced and better educated owners are more likely to reduce informational asymmetries and secure financing in the first few years of operations (Cole and Sokolyk, 2013). Business owners preferences also play an important role in the financing decisions of small, privately-held firms. Ang et al. (2010) show that the owner s personal preferences account for 33 to 60 percent of the variation in the capital structure decisions in sole proprietorships. In light of the evidence on the impact of owner characteristics such as education, experience, etc on financial decision-making, we argue that such characteristics may have an impact on the probability of lease financing. Therefore, we hypothesize the following: H3: Owners who are perceived as risky borrowers are more likely to lease. The literature on leasing decisions for large corporations documented several firm characteristics that explain the decision to buy versus lease and also the determinants of the finance leases versus operating leases. In our study, we control for the variables that have been used in the 7

8 previous studies of large corporations and provide insights into the ways in which startup firms are different from large corporations in the way they use leasing financing. Leasing is a substitute/complement to debt financing: There is a large body of literature focusing on the decision to lease or to buy an asset. The empirical evidence is mixed as some authors documented a substitution effect (Myers et al., 1976; Mukherjee, 1991; Lasers and Levis, 1998; Yan, 2006 amongst others), while others found that leasing is a complement not a substitute to debt financing (Ang and Peterson, 1984; Krishnan and Moyer, 1994; Bathala and Mukherjee, 1995). The databases and various models used in these studies are not comparable and therefore the substitute/complement of leases versus debt is still an empirical question. In addition, except for one study (Bathala and Mukherjee, 1995), all other studies used data on large publicly traded corporations. In our study we control for the debt ratio as well as other debt financing variables (the use of personal debt, business debt, and debt injections) to explain the propensity of leasing of U.S startups. To control for the endogeneity problem, the debt ratio used in the Logit regressions is lagged (Yan, 2006). Financial constraints: From the lessor s perspective, one benefit of leasing is the ability to repossess an asset. In the U.S. bankruptcy code, secured loans and leasing are treated differently. It is easier for the lessor to regain control of an asset than it is for a lender of a secured loan to repossess that asset. As a result, lessors extend more credit against a leased asset than a lender would for a similar secured asset. Eisfeldt and Rampini (2009) develop a model of leasing versus buying capital and demonstrate that financially constrained firms should lease more. Empirically, they show that the fraction of capital the firms lease is significantly decreasing with firm size (46% for small firms versus 11% for large firms). Lin et al. (2013) examine the choice between debt and leasing decisions of publicly traded corporations and show that firms are not indifferent between 8

9 leases and debt but rather these decisions are a function of how financially constrained firms are. For the most financially constrained firms, leasing is negatively related to internal funds and for the less constrained firms, leasing is positively related to internal funds. The conclusion is that debt and leasing decisions are simultaneous capital structure decisions, regardless of the firm size. In the case of startup businesses that are financially constrained and informationally opaque, leasing reduces the expected bankruptcy costs compared to debt financing. Therefore, in our study we include a measure of the firms ability to secure financing ( credit risk score ) as well as a measure of information asymmetry ( Info ). We expect to find a higher propensity of leasing in financially constrained and informationally opaque startups. Firm size: The impact of size on the leasing decisions is important since size is related to the cost of obtaining external funds. The empirical evidence is mixed as some studies document a positive relationship between leasing and firm size (Lasfer and Levis, 1998; Mehran et al., 1999) while some studies show a negative relationship (Sharpe and Nguyen, 1995; Graham et al., 1998). However, Ang and Peterson (1984) and El-Gazaar et al. (1986) find no significant relationship between leasing and firm size. Smaller firms, especially startups, bear higher costs for getting external financing due to information asymmetry, and therefore they should lease more. In our study we use alternative definitions for startup size, such as: logarithm of total assets, logarithm of sales, and total number of employees. In addition, we classify firms in four categories based on total assets and examine the relationship between leasing and size in various asset size categories. Industry: Several studies document that leasing activity is more prevalent in some industries than in others, although Ang and Peterson (1984) show that lessee firms are not concentrated in few industries and leasing occurs in every industry group included in their analysis. The reasons as to why leasing activity might be prevalent in some industries is that there are 9

10 industry-wide differences in investment tax credit, asset structure, marginal tax rates and debt capacity. Finucane (1988) and Gosman and Hanson (2000) show that air transport and retailing industries have a higher leasing activity, while Adams and Hardwick (1998) document that service and utilities companies used more leases whereas construction companies used leases less. In our study we use standard industrial classification system (SIC) to classify startups in eight industry groups based on 2 digits SIC code and examine the impact of industry differences on leasing activity. Form of organization: Prior studies document a significant relationship between a firm s legal form of organization and leverage level; specifically, firms with limited liability consistently report higher level of leverage (Herranz et al., 2009) than firms with unlimited liability. Riskaverse entrepreneurs choose lower level of leverage when they are personally liable for the firms debt. On the other hand, Robb and Robinson (2010) argue that loans to corporations usually require personal guarantees (i.e. real estate property) from entrepreneurs; if this is the case, then there will be no difference to risk-averse entrepreneurs in the amount of leverage they use regardless of legal form of organization. However, it may make sense to them to use more leasing financing to reduce the potential of being personally liable for the business debt. Therefore, we include the legal form of organization (sole proprietorship, LLC, S-corporation) in our analysis to control for entrepreneurs limited liability. 3. Data The Kauffman firm Survey 3.1 The sample The Kauffman Firm Survey (KFS) is the largest longitudinal survey of startup businesses in the United States. The survey data contain detailed information on 4,928 businesses founded in 2004 and track their business status and performance annually until 2011, an eight year panel-data. 10

11 Detailed information on the business includes sources and types of capital in the startup year and over time, intellectual property rights (patents, trademarks, copyrights), R&D activity, physical location, whether the business provides a product, service or both, revenues, asset structure, sales, employment, and industry. Owners' characteristics (up to ten active-owner-operators per business) include: years of work experience, previous startup experience, time commitment and education, net worth, age, gender, race and ethnicity. The definitions of variables used in this study are presented in Table 1. In response to the Kauffman Foundation s interest in understanding the dynamics of hightechnology, medium-technology, and women-owned businesses, the KFS sample was stratified based on industrial technology level (High-Tech, Medium-Tech, and Non-Tech sectors) and gender, oversampling businesses in high- and medium-tech sectors. It is important to emphasize that women-owned businesses were not oversampled in the KFS. The analysis in this study is based on the imputed data set. Survey weights are applied to all statistical analyses in this paper. Details of the data imputation procedures and the use of proper weights are available in Farhat and Robb (2014). Insert Table 2 here Table 2 shows the sample distribution for lessee and non-lessee firms by year. We require that sample firms have eight years of data. Thus, our data do not include firms that sold, merged or closed their operations during the period. There are 1,630 firms established in 2004 that meet our sample selection criteria. To avoid survival bias and sample selection problems, we use a subpopulation analysis with the proper weights. In addition, we conduct our analysis using the full sample and the results are very similar. 11

12 The number of lessee firms varies from year to year from 324 firms in 2004 to 234 firms in The data show that the percentage of lessee firms in total firms decreases over the sample period from a maximum of 23.3% in 2006 to 14.4% in The higher leasing activity in the first few years of operations suggests that startups take advantage of the flexibility feature of leasing contracts especially when they are the most financially constrained Descriptive statistics Table 3 reports descriptive statistics for the full sample as well as for subsamples of lessee and non-lessee firms. Overall, the percentage of firms that use leasing contracts over the period is 22.9%, but if we look at firms that use leasing contracts in at least one year during the sample period, this percentage is much higher (42.1%). A higher proportion of lessee firms tend to use more personal debt and/or business debt and have a higher debt ratio compared to that of non-lessee firms. In terms of size, lessee firms are larger than non-lessee firms, regardless of the measure we use: assets, sales, or number of employees. When we classify firms in four asset categories, we find that only very small firms with assets under $100,000 have a lower incidence of leasing activity relative to other asset size categories. The lower proportion of lessee firms (56.3%) in this category suggests that very small startups may finance their assets with inside equity or informal debt (i.e. loans from friends and family). The average credit risk score measures the ability of startups to secure debt financing. For the full sample the credit score is 2.96 and the value is very close for both lessee firms (2.91) and non-lessee firms (3.02). Therefore, based on this measure, all firms have the same ability to raise debt. The percentage of firms with research and development activity is 16% for the full sample but is higher for lessee firms (17.6%) and lower for non-lessee firms (14.2%). Small businesses with R&D activity could be viewed as businesses with growth opportunities. Typically, higher 12

13 growth options businesses should use less debt financing to mitigate the underinvestment problem (Myers, 1977) and use more leasing instead. Our data confirm that firms with higher R&D activity use more leasing financing. Asset specificity is another variable of interest in our analysis because unique assets are not easily deployable. The percentage of lessee firms with specific assets (5.5%) is lower relative to non-lessee firms (7.8%). This shows that asset specificity is an important characteristic that differentiate leasing activity among startups. About 19.8% of firms possess intellectual property rights and both lessee and non-lessee firms share about the same percentage of IP rights. In addition, startups that offer services rather than products have a higher propensity of leasing activity (91.2% versus 84.9%). Although 48.7% of firms in the sample are home-based, only about 37.2% of them report leasing activity. The majority of lessee firms utilize a rented space. With respect to legal organizational form, startups organized as limited liability corporations (33.6%) or S-corporations (29.9%) have a higher propensity of leasing activity compared to sole proprietorship (25.7%). Owner s characteristics such as gender, experience and education are different for lessee and nonlessee firms. Data show that only 25.4% of lessee firms are led by female entrepreneurs (compared to 37.9% for non-lessee firms) and lessee firms owners have longer experience (13.1 years) and are less educated compared to non-lessee firms. Data also show a higher proportion of firms in Agriculture, Construction and Transportation/Communications have, on average, more leasing activity. Insert Table 3 here 13

14 4. Empirical Results 4.1. Univariate analysis Table 4 reports the difference in business, owners and industry characteristics between lessee firms and non-lessee firms, year by year. When comparing the two groups in terms of intellectual property rights, R&D activity and asset specificity we find interesting results. First, there is no significant difference between lessee and non-lessee firms with respect to intellectual property rights, which suggests that IP alone does not explain why some firms use leasing and some don t. However, with respect to research and development activity, a slightly higher proportion of lessee firms report R&D compared to non-lessee firms and this difference is statistically significant in five out of eight years. In addition, we find that a lower proportion of lessee firms have assets that are specific, not easily deployable compared to non-lessee firms and the difference is statistically significant in seven out of eight years. This result suggests that both growth options and asset specificity could explain the leasing decisions of startups. A significantly higher proportion of lessee firms use personal debt, business debt or a combination of the two in every single year. In addition, the debt ratio of lessee firms is significantly higher than that of non-lessee firms which suggests that lessee firms are more leveraged. This result leads us to conclude that leasing and debt financing are complements rather than substitutes. The result is consistent with the findings of Ang and Peterson (1984), Krishnan and Moyer (1994) for large corporations and Bathala and Mukherjee (1995) for small corporations. When comparing the average size of firms in the two different groups, we find that lessee firms are significantly larger in size in any single year, regardless of the measure used for size (assets, sales or number of employees). This result is counterintuitive as one might expect smaller, financially constrained startups to use leasing more and debt financing less. Even more interesting 14

15 is the fact that a significantly lower proportion of sole proprietors use leasing. Sole proprietors have unlimited liability which will make debt financing riskier, so one might expect that leasing is a more flexible, less risky choice for this category of entrepreneurs. We find that lessee firms have lower credit risk scores relative to non-lessee firms and the difference is statistically significant in five out of eight years. This finding suggests that startups with leasing activity are less risky and have the ability to secure debt financing, yet they choose to lease some of their assets. In terms of information asymmetry, we find very weak evidence that lessee and non-lessee firms differ; however, with respect to business location, a significantly lower proportion of lessee firms are home-based relative to non-lessee firms. The univariate analysis also reveals that a significantly higher proportion of lessee firms provide services rather than products and operate in a rented space rather than from owners home. In addition, data show that a significantly higher proportion of lessee firms are organized as S- corporations, and a lower proportion of non-lessee firms are organized as sole proprietorships. Owners characteristics such as experience, education, and age are important factors that may explain the decisions to lease. We find significant differences between lessee and non-lessee firms, as a higher proportion of lessee firms tend to have younger, more experienced, less educated owners and are led by male entrepreneurs relative to non-lessee firms. It is worth noting that owner s race shows no significant difference between lessee and non-lessee firms. With respect to industry classification, we find that a significantly higher proportion of lessee firms are in the construction and transportation/communication/utilities industries, whereas a significantly lower proportion of lessee firms are in the services industry. Insert Table 4 here 15

16 4.2. Multivariate analysis We use Logit models to estimate the probability that leasing activity will occur. The independent variables are factors such as firm, owner and industry characteristics that may explain why some firms lease more than others. Let y represent the propensity of firm i to use leasing financing. The relationship between the observed outcome y and response propensity can be written as: y * 0 0 if y * 1 0 if y (1) i.e. y 0 for firms with leasing financing y X u, y 0 if y * 1 0 Otherwise (2) where is a vector of coefficients, X is a vector of independent variables and u are disturbance terms. The dependent variable is a binary variable that takes the value of 1 if the startup has reported leasing activity in a given year and 0 otherwise. The independent variables are firm, owner and industry characteristics such as: a) Firm characteristics: R&D activity, intellectual property rights, asset uniqueness, firm size, debt ratio, debt injections, information asymmetry, credit risk, business location and legal status, and whether the business provides a service or a product. b) Owner s characteristics: owner's experience, education, gender, race and age. c) Industry: we use SIC to classify businesses in eight industry groups. 16

17 Our main focus is to test if asset uniqueness and growth opportunities are significant determinants of leasing activity in U.S. startups. Table 5 reports the results from Logit regressions. We use alternative control variables for leverage and size, as follows: in Models 1-4 we use lagged debt ratio, in Model 5 we use a dummy variable that takes the value of 1 if the startup has used any debt, and in Model 6 we use percentage debt injections to total capital injections as a proxy for leverage. For firm size we use natural logarithm of total assets (Model 1, 5 and 6), natural logarithm of sales (Model 3), total number of employees (Model 4) and asset categories (Model 2). Insert Table 5 here Table 5 shows that regardless of the alternative definitions for leverage and size, high technology startups, one of our measures for asset uniqueness, are less likely to lease. This confirms our first hypothesis and the results are significant at the 1% level. The results are similar to those found for large corporations as Smith and Wakeman (1985) report that specialized assets are less likely to be leased due to the fact that those assets have almost no value to others. The alternative variable for asset uniqueness, intellectual property, has the right sign but is not significant. Therefore, we can conclude that IP alone does not explain the propensity of leasing, but controlling for the IP rights, startups in the high-technology sector have a lower propensity of leasing. Our second hypothesis that startups with growth opportunities should lease more finds strong support in our analysis. The coefficient for R&D variable is positive and significant in all models. This result lead us to conclude that after controlling for leverage and size, startups characterized by high growth opportunities have the propensity to lease more. There hasn t been any study that examined the relationship between growth opportunities and leasing in small firms. 17

18 However, one study (Bathala and Mukherjee, 1995) document a positive relationship between current growth (measured as sales growth rate) and leasing. We complement their earlier result by reporting that startups with high growth opportunities lease more compared to those that have low growth opportunities. Our hypothesis that owner s characteristics may impact their risk-taking behavior and therefore influence the likelihood of leasing finds some support in our analysis. The coefficient for education is negative and significant which suggests that highly educated owners are less likely to lease. Experience and race variables are insignificant, whereas owner s age is negative and statistically significant in three models only. Prior research indicates that older individuals are more risk averse, yet our findings reveal that older owners have a lower propensity of leasing. We also find that gender is a significant factor in the leasing decisions of startup firms, as female entrepreneurs have a lower propensity to lease compared to their male counterparts. Consistent with prior studies on large firms, our results show that the higher the leverage the higher the propensity of leasing activity. This result confirms that leasing and debt financing are complement not substitute to each other. Our result is consistent with Bathala and Mukherjee (1995) who examined 104 small corporations and reported that 48% of small firm owners viewed the debt-leasing relationship as complement not substitute. We find that size is an important characteristic in explaining the leasing propensity of startups. Consistent with some studies for large firms (Lasfer and Levis, 1998), we find a positive and significant relationship between firm size and leasing activity. Our results hold regardless of the proxy used for firm size (total assets, sales or number of employees) and the coefficients are significant at 1% level. The result is counterintuitive as smaller firms, especially startups, bear higher costs for getting external financing due to information asymmetry and, therefore, should 18

19 lease more. We explain this positive relationship between firm size and the likelihood of leasing among privately held firm based on the investment opportunity set of each firm. Larger startups are more diversified and more likely to have larger investments in fixed assets. Since leasing contributes to financing of fixed assets, larger startups are more likely to use leasing relative to smaller ones. This result is consistent with Bathala and Mukherjee (1995) who document that 46% of surveyed businesses in their sample do not lease. Our measure for the firm s ability to secure debt financing, credit risk score, is negative and statistically significant in all models. The credit risk score variable takes values from 1 (the best score) to 5 (the worst score). Therefore, startups with higher credit risk scores are viewed as the riskiest by potential lenders. The negative coefficient for credit risk score in our Logit regressions suggests that riskier startups (less able to access debt financing) have a lower the propensity of leasing. We find that the information asymmetry metric is not statistically significant in our analysis, which suggests that regardless of the level of information asymmetry, startups have the same propensity to lease. Contrary to our prediction, the legal form of organization does not impact the decision to lease or not to lease. Our results show that the coefficients for different legal forms of organization are not statistically significant. This suggests that owners liability (limited or unlimited) has no significant impact on the leasing decisions in U.S. startups. Our results also suggest that leasing is more/less prevalent in some industries than in others. We find that startups in agriculture and construction industries have a higher propensity of leasing, while those in finance/insurance/real estate have a lower propensity of leasing (coefficients are not reported in Table 5). While these results are different from those reported in studies on large firms (Gosman and Hanson, 2000) we claim that small businesses differ from large corporations 19

20 operating in the same industry in terms of debt capacity, asset structure and marginal tax rates. Therefore, the higher propensity of leasing in certain industries may be driven by the positive effects of leasing on small firms relative to that on larger corporations. In addition, we find that home-based startups have a lower probability of leasing activity and those providing a service rather than a product have a higher propensity to lease some of their assets. 5. Conclusions In this study we use a unique dataset to explore the impact of asset uniqueness and growth opportunities on the leasing decisions of U.S. startups during the period. Although prior research has explored the buy versus leasing decision and the determinants of leasing for large corporations, our research makes a unique contribution by examining factors that explain these decisions for U.S. startup firms. We find that asset uniqueness is a significant determinant of the propensity to lease. Startups that operate in the high-technology sector are less likely to use lease financing because their assets are unique and not easily deployable. However, having intellectual property rights such as patents or copyrights does not play a significant role in the leasing decisions of startups. Our results also indicate that startups with high growth opportunities have the propensity to lease more. While Bathala and Mukherjee (1995) report a positive relationship between current growth and leasing for small corporations, we contribute to the leasing literature by reporting a positive relationship between future growth opportunities and the likelihood of leasing for U.S. startups. We are the first to examine the impact of owners socio-economic and demographic characteristics that impact the leasing decisions of U.S startups. The KFS survey allows us to test 20

21 whether characteristics that define the owners risk taking behavior have an impact on the financial-decision making, including the decision to lease. Our findings reveal that female entrepreneurs, and those that are highly educated are less likely to lease. In addition, we find weak evidence that older entrepreneurs have a lower propensity to lease. Our results confirm prior findings on large corporations as larger and more leveraged startups are more likely to lease. In addition, there is a higher propensity to lease in industries such as agriculture and construction and a lower propensity to lease in finance/insurance/real estate. While the legal form of organization does not play a significant role in the leasing decisions, homebased startups have a lower likelihood to lease assets, whereas those that provide a service rather than a product are more likely to lease. Overall our findings provide new evidence on the leasing decisions of U.S. startups that help us gain a better understanding of the ways in which small firms make their leasing decisions. 21

22 References Adams, M. and Hardwick, P. (1998). Determinants of the leasing decisions in the United Kingdom listed companies, Applied Financial Economics 8, Ang, J., and Peterson, P. (1984). The leasing puzzle, The Journal of Finance 39, Ang, J., Cole, R. and Lawson, D. (2010) The role of owner in capital structure decisions: An analysis of single owner corporations. Journal of Entrepreneurial Finance 4, Bathala, C. and Mukherjee, T. (1995). A survey of leasing in small firms. Journal of Small Business Finance 4, Bertrand, M.and Schoar, A. (2003). Managing with style: The effect of managers on firm policies, Quantitative Journal of Economics, Blanchflower, N., Levine, P.B., and Zimmerman, D.j. (2003). Discrimination in the small business credit market, The Review of Economic and Statistics 85, Bruder, J., Neuberger, D. and Rathke-Doppner, S. (2011). Financial constraints of ethnic entrepreneurship: Evidence from Germany, International Journal of Entrepreneurial Behavior and Research 17, Cavalluzzo, K.S., Wolken, J.D. and Cavalluzzo, L.C. (2002). Competition, small business financing and discrimination: Evidence from a new survey, The Journal of Business 75, Cavalluzzo, K.S. and Wolken, J.D. (2005). Small business loan turndowns, personal wealth and discrimination, The Journal of Business 78, Cole, R. and Sokolyk, T. (2013). How do startup firms finance their assets? Evidence from the Kauffman Firm Survey, Working paper El-Gazzar, S., Lilien, S. and Pastena, V. (1986). Accounting for leases by lessees, Journal of Accounting and Economics 8, Einsfeldt, A. and Rampini, A. (2009) Leasing, ability to repossess, and debt capacity. Review of Financial Studies 22, Equipment Leasing and Finance Association, July 21, Farhat and Robb (2014) Applied Survey Data Analysis using STATA: The Kauffman Firm Survey Data, Ewing Marion Kauffman Foundation Finucane, T (1988). Some empirical evidence on the use of financial leases, Journal of Financial Research 11,

23 Graham, J., Lemmon M., and Schallheim, J. (1998). Debt, leases, taxes, and the endogeneity of corporate tax status, Journal of Finance 53, Gosman, M. and Hanson, E. (2000). The impact of leasing on lenders evaluation of firms debt levels, Commercial Lending Review 15, Hall, B. (2002). The Financing of Research and Development, Oxford Review of Economic Policy, 18 (1), Herranz, N., Krasa, S. and Villamil, A. (2009).Small firms in the SSBF. Annals of Finance 5, Krishnan, V.S. and Moyer, R.C. (1994). Bankruptcy costs and the financial leasing decision, Financial Management 23, Lin, J. R, Wang, C-J, Chou, D.W, Cheh, F. C. (2013). Financial constraint and the choice between leasing and debt, International Review of Economics and Finance 27, Lasfer, M.A. and Levis, M. (1998). The determinants of the leasing decisions of small and large companies, European Financial Management 4, Mehran, H., Taggart, R., and Yermack, D. (1999). CEO ownership, leasing and debt financingstatistical data included, Financial Management 28, McInish, T.H. (1982). Individual investors and risk-taking, Journal of Econ Psyhol 2, Myers, S. C., (1977). Determinants of corporate borrowing, Journal of Financial Economics 5, Myers, S., Dill, D. and Bautista, A. (1976). Valuation of financial lease contracts, Journal of Finance 31, Myers, S. C. and Majluf, N. S. (1984). Corporate financing and investment decisions when firms have information that investors do not have, Journal of Financial Economics 13, Mukherjee, T. K. (1991). A survey of corporate leasing activity, Financial Management 29, Refiner, U. (2005). Expertise for the Chapter on Wirtschaftliche Potenziale des Alters on the Fifth Report in Elderly People, Federal Government of Germany. Robb, A. M. and Robinson, D.T. (2012). The capital structure decisions of new firms, Review of Financial Studies, forthcoming Sharpe, S.A. and Nguyen, H.H. (1995). Capital market imperfections and the incentive to lease, Journal of Financial Economics 39,

24 Smith, C. and Wakeman, L. (1985). Determinants of corporate leasing policy, Journal of Finance 40, Yan, A. (2006). Leasing and debt financing: Substitutes or complements? Journal of Financial and Quantitative Analysis 41,

25 Table 1 Variable Description Variable Description Used Leasing Equals 1 if the firm used leasing, =0 otherwise Used Personal Debt Equals 1 if the firm used personal debt, =0 otherwise Used Business Debt Equals 1 if the firm used business debt, =0 otherwise Used Any Debt Equals 1 if the firm used debt, =0 otherwise Debt Ratio Ratio of debt to total capital % Debt Injection Percent of debt injections to total capital injections Ln (Debt+1) Natural logarithm of total debt Ln (Assets+1) Natural logarithm of total assets Ln (Sales+1) Natural logarithm of total sales Asset Categories $0-$100,000 Equals 1 if the firm assets are between 0 and $100,000 and =0 otherwise $100,000-$500,000 Equals 1 if the firm assets are between $100,000 and 500,000 and =0 otherwise $500,000-$1,000,000 Equals 1 if the firm assets are between $500,000 and $1,000,000 and =0 otherwise $1,000,000+ Equals 1 if the firm assets are above $1,000,000 and =0 otherwise No. employees Number of full time and part time employees Credit Risk Score Commercial Credit Risk Score ( 5 is very high risk, 1 very low risk) Information asymmetry Equals 1 if D&B did not report a credit score for the business, =0 otherwise Intellectual Property (IP) Equals 1 if business has patent or copyright or trademark, =0 otherwise R&D activity Equals 1 if business has at least one employee responsible for R&D, =0 otherwise Assets Specific Equals 1 if the firm operates in a Hi-Tech sector, =0 otherwise Provides Service Equals 1 if business provides service, =0 otherwise Provides Product Equals 1 if business provides product, =0 otherwise Location Home Based Equals 1 if business is home-based, =0 otherwise A rented or leased space Equals 1 if business is located in a rented or leased space, =0 otherwise Legal Status Sole Proprietorship Equals 1 if business is sole proprietorship, =0 otherwise Limited Liability Company Equals 1 if business is limited liability company, =0 otherwise Subchapter S-Corporation Equals 1 if business is subchapter S-Corporation, =0 otherwise Owner s Characteristics Female owner Percentage of female owners Owner s work experience Average work experience for all owners (in years) Owner s age Average age for all owners (in years) Owner s education Equals 1 if majority of owners have a college degree or above, =0 otherwise White Owner Percentage of White (Non-Hispanic) owners 25

26 Table 2 Sample Distribution Year Non-Lessee Firms Lessee Firms Total N % N % , , , , , , , , , , , , , , , ,630 26

27 Table 3 Descriptive statistics Firm characteristics Full Non-Lessee Lessee firms* sample firms Used Leasing % Used Personal Debt % Used Business Debt % Use Any Debt % Debt ratio % Ln (Debt+1) $ Ln (Assets+1) $ Ln (Sales+1) $ Asset Categories $0-$100,000 % $100,000-$500,000 % $500,000-$1,000,000 % $1,000,000+ % No. employees # Credit Risk Score # Information asymmetry % Have Intellectual Property % R&D activity % Assets Specific % Provide Service % Provide Product % Firm Location Home Based % A rented or leased space % Other % Legal Status Sole Proprietorship % Limited Liability Company % Subchapter S-Corporation % Other % Owner s Characteristics Female owner % Owner s work experience Years Owner s age Years Owner s education % White Owner % Industry 27

28 Agriculture % Construction % Manufacturing % Transportation, Communications, Utilities % Wholesale Trade % Retail Trade % Finance, Insurance, Real Estate % Services % N (firm-year observations) N 13,040 6,448 6,592 * Firms that used leasing in at least one year during the sample period. 28

29 Table 4 Difference in Business, Owners and Industry Characteristics between Lessee Firms and Non-Lessee Firms Lessee Firms Firm characteristics Used Personal Debt % 0.723** 0.717*** 0.729*** 0.719*** 0.73*** 0.706*** 0.651** 0.603** Used Business Debt % 0.460*** 0.505*** 0.510*** 0.504*** 0.496*** 0.482*** 0.468*** 0.442*** Used Any Debt % 0.814*** 0.810*** 0.830*** 0.815*** 0.822*** 0.798*** 0.764*** 0.714*** Debt ratio $ 0.396*** 0.284*** 0.299*** 0.299*** 0.301*** 0.260*** 0.230*** 0.203*** Ln (Debt+1) $ 6.499*** 6.213*** 6.390*** 6.338*** 6.536*** 6.023*** 5.443*** 5.054*** Ln (Assets+1) $ *** *** *** *** *** *** *** *** Ln (Sales+1) $ 7.502** 9.209*** 9.688*** 9.835*** *** *** *** *** Asset Categories $0-$100,000 % 0.677*** 0.596*** 0.534*** 0.525*** 0.515*** 0.547*** 0.55*** 0.56*** $100,000-$500,000 % 0.239*** 0.264*** 0.29*** 0.296*** 0.293*** 0.25*** 0.253** 0.249*** $500,000-$1,000,000 % 0.050** 0.068*** 0.087*** 0.087*** 0.090*** 0.101*** 0.085*** 0.092** $1,000,000+ % 0.034** 0.072*** 0.089*** 0.093*** 0.103*** 0.102*** 0.112*** 0.099*** No. employees # 2.507*** 4.194*** 4.802*** 5.465*** 5.207*** 5.204*** 5.476*** 5.793*** Credit Risk Score # 3.268*** 3.094*** 2.849*** 2.751*** 2.658*** Information asymmetry % ** Have Intellectual Property % R&D activity % ** 0.252*** 0.185** ** 0.137** Assets Specific % ** 0.055** 0.053** 0.055** 0.055** 0.057** 0.055** Provide Service % 0.914*** 0.920*** 0.917*** 0.907** 0.911*** 0.907*** 0.908*** 0.909*** Provide Product % Firm Location Home Based % 0.394*** 0.372*** 0.363*** 0.372*** 0.371*** 0.371*** 0.365*** 0.373*** A rented or leased space % 0.499*** 0.499*** 0.504*** 0.496*** 0.497*** 0.494*** 0.497*** 0.486*** Other % ** 0.133** 0.132** 0.135** 0.138** 0.141** 29

30 Legal Status Sole Proprietorship % 0.299*** 0.281*** 0.273*** 0.260*** 0.250*** 0.239*** 0.232*** 0.225*** Limited Liability Company % Subchapter S-Corporation % 0.264*** 0.286*** 0.291*** 0.299*** 0.309*** 0.317*** 0.315*** 0.315*** Other % Owner s Characteristics Female owner % 0.241*** 0.254*** 0.257*** 0.255*** 0.26*** 0.257*** 0.253*** 0.252*** Owner s work experience Years ** ** ** ** ** ** ** ** Owner s age Years *** *** *** 46.97*** 48.07*** *** *** ** Owner s education % 0.457*** 0.447*** 0.447*** 0.443*** 0.443*** 0.444*** 0.449*** 0.449*** White Owner % ** Industry Agriculture % ** 0.036** 0.038** 0.038** Construction % 0.068*** 0.077*** 0.069*** 0.066*** 0.073*** 0.073*** 0.073*** 0.074*** Manufacturing % Transportation, Communications, Utilities % 0.052** ** 0.053** 0.053** 0.053** 0.053** 0.054** Wholesale Trade % Retail Trade % Finance, Insurance, Real Estate % Services % 0.534** 0.546** 0.531** 0.530** 0.527** 0.522*** 0.530*** 0.526*** Number of firms Non-Lessee firms Firm Characteristics Used Personal Debt % Used Business Debt % Used Any Debt % Debt ratio $ Ln (Debt+1) $