http://www.smallbusinessinstitute.biz Should family firms internationalize? Evidence from the Survey of Business Owners Mark D. Heileman1, Timothy L. Pett2 1Rollins College, mheileman@rollins.edu 2Rollins College, tpett@rollins.edu A B S T R A C T The purpose of this study was to explore the characteristics and performance of family-owned firms with internationalization. We were motivated to determine if there were significant differences between family-owned firms with internationalization and other firm types, specifically, family-owned firms without internationalization and non-family-owned firms. The study draws on the Census Bureau’s Survey of Business Owners (SBO) Public Use Microdata Sample (PUMS). SBO response variables regarding owner demo- graphics, business acquisition, business context, and a number of business performance outcome measures were the outcome variables of interest in this study. A comparison of means was applied to test whether or not there were differences in response vari- ables across the family-owned and non-family-owned firm types. The results indicated that family-owned firms with international- ization, on average, had lower business closures and higher sales than the other firm types, and that firms with internationalization were more efficient in terms of sales per employee and sales per payroll. This study contributes to understanding the characteristics and performance between family-owned and non-family-owned firms in conjunction with those that internationalize and those that did not internationalize. A novel feature of the study experimental design was the incorporation of primary owner characteristics and whether there were any business acquisition, attribute, and performance correlations. The findings suggest practical implica- tions for business growth strategy with regards to exporting, establishing international operations, or outsourcing business functions out of the USA. The study concludes with a discussion of the findings and offers potential future research directions. Keywords: Journal of Small Business Strategy 2018, Vol. 28, No. 01, 1-13 ISSN: 1081-8510 (Print) 2380-1751 (Online) ©Copyright 2018 Small Business Institute® APA Citation Information: Heileman, M. D., & Pett, T. L. (2018). Should family firms internationalize?: Evidence from the Survey of Business Owners. Journal of Small Business Strategy, 28(1), 1-13. w w w. j s b s . o rg Family-owned firm, Internationalization, Performance Introduction Family-owned firms make important contributions to today’s businesses in the form of job creation and growth in the economy. There is widespread agreement that fam- ily-owned firms will eventually consider internationaliza- tion as one strategy for business growth (Kontinen & Ojala, 2010; Patel, Pieper, & Hair, 2012). Understanding the char- acteristics of these firms and their business owners may shed light on the importance in measuring the economy’s overall well-being (Lichtenstein, 2014). A number of stud- ies have examined the relationships between small- and medium-sized enterprises (SMEs), which are often family businesses (Patel et al., 2012), and owner demographics (Blackburn, Hart, & Wainwright, 2013), business acquisi- tion (Fairlie & Robb, 2009), and performance (Mittelstaedt, Harben, & Ward, 2003; Wincent, 2005; Wolff & Pett, 2000). However, few studies have explored the relationship be- tween owner demographics, business acquisition, and business context using efficiency measures of performance as demonstrated by sales per employee and sales per pay- roll (Blackburn, Hart, & Wainwright, 2013; Heileman, Pett, & Mayer, 2016; Pett & Wolff, 2016). Furthermore, few stud- ies have examined these performance outcomes using a holistic measure of internationalization as it relates to fam- ily firms (Kontinen & Ojala, 2010; Pukall & Calabrò, 2014). There is an increasing interest in ascertaining whether family-owned firms achieve superior performance when compared with their non-family counterparts. Further re- search is needed to understand the determinants of family firm performance (Graves & Shan, 2014). The motivation for this research was to determine if there were significant differences between family-owned firms with internation- alization and other firm types, specifically, family-owned firms without internationalization and non-family-owned firms. Theoretical Framework and Hypotheses Today’s global markets and environment play an im- portant role in the growth and success of family-owned firms. Growth is a major strategic decision for all business enterprises, and global expansion is one available option for firms to consider. However, internationalization is a http://www.smallbusinessinstitute.biz http://www.jsbs.org 2 M. D. Heileman, & T. L. Pett Journal of Small Business Strategy / Vol. 28, No. 1 (2018) / 1-13 complex pathway not often chosen by family firms. With- out timely and productive responses to global shifts, many family businesses potentially face decline or failure (Patel et al., 2012). The intention of this study was to explore family-owned firms with internationalization in conjunction with the business characteristics and performance. The central re- search question we sought to investigate was: Do different firm types—that is family-owned firms with international- ization, family-owned firms without internationalization, not family-owned firms with internationalization, and not family-owned firms without internationalization—have dif- ferent characteristics such as owner demographics, busi- ness acquisition, and business context, as well as different business performance? Numerous studies have examined SMEs relative to size and characteristics such as owner demographics and business acquisition methods (Blackburn et al., 2013; Fair- lie & Robb, 2009; Mittelstaedt et al., 2003; Wincent, 2005; Wolff & Pett, 2000). On the other hand, few studies have explored the relationship between owner demographics, business acquisition, and business context with measures of performance as demonstrated by sales and employee efficiency with regards to sales (Blackburn et al., 2013; Hei- leman et al., 2016; Pett & Wolff, 2016). Family Firms Family firms represent a significant economic force, yet no clear consensus exists regarding the definition of what constitutes a family business (Abdellatif, Amann, & Jauss- aud, 2010; Kontinen & Ojala, 2010). Chua, Chrisman, and Sharma (1999) provided a comprehensive review of the lit- erature on the definition of family business. Several obser- vations were made about these definitions. First, with few exceptions, the definitions did not differentiate between governance and management. Second, some required controlling ownership or family management alone while others required both ownership and management. Thus, the generally accepted definitions were placed into three different groups: (i) family owned and family managed, (ii) family owned but not family managed, and (iii) family man- aged but not family owned firm. The Census Bureau (2012) approach could follow Chua et al.’s (1999) first two forms of a family business, defin- ing family business firms as having two or more members of the same family owning the majority of the business (where family refers to spouses, parents/guardians, chil- dren, siblings, or close relatives). This research uses this definition as presented in the report provided by the Cen- sus Bureau’s characterization of a family firm. Overall, previous research found that family business- es generally perform better and enjoy a sounder finan- cial structure than do non-family businesses (Abdellatif, Amann, & Jaussaud, 2010). On the other hand, research findings indicate that while the extent of innovation pos- itively influences the long-term sales growth of entrepre- neurial firms, it is negatively moderated by family involve- ment (Chang, Memili, Chrisman, & Welsh, 2011). Internationalization Internationalization is an ambiguous term in the family business literature. Internationalization is the process of in- creasing involvement of firms in international markets, al- though there is no agreed definition of internationalization (Kontinen & Ojala, 2010). Bose (2016) identified four crit- ical success factors for SME internationalization—current scenario, future prospects, competencies, and strategies— along with a framework for presenting the relationship among these factors. However, Cieślik, Kaciak, and Welsh (2010) found that early involvement in international activ- ities negatively affected the survival and regularity of sales of small exporters but positively impacted larger exporters. The internationalization of large multinational enterpris- es has numerous theories of how firms internationalize including the eclectic theory, Uppsala model, and inter- nationalization theory (Reuber, 2016). Nonetheless, these theories have provided little support for understanding or explaining the internationalization of either SMEs or family firms. Many studies implicitly or explicitly assume that sales generated outside the home country (or exports) is re- ferred to as internationalization (Pukall & Calabrò, 2014). However, there are other dimensions to international- ization involving the firm’s exploitation of opportunities across national borders in order to create goods and ser- vices (Kontinen & Ojala, 2010; Patel et al., 2012; Pukall & Calabrò, 2014). Which component of internationalization should be analyzed depends on the research question asked. To achieve a more comprehensive grasp of this mul- tidimensionality, in this study internationalization was de- fined as firms that (i) have exports or (ii) have established operations outside the USA or (iii) have outsourced busi- ness functions out of the USA. Business Acquisition The business owner may have acquired the business by founding it, purchasing it, inheriting it, or receiving it as a transfer or gift. Family firms tend to avoid external own- ership to stay independent, and may be willing to accept negative outcomes such as poorer performance than that of non-family firms (Kraus, Mensching, Calabrò, Cheng, & Filser, 2016). Family firm owners accumulate economic and noneconomic wealth they may wish to transfer to the next generation (Carr, Chrisman, Chua, & Steier, 2016). In general, owners of inherited businesses have less involve- ment with decision-making than owners of founded or pur- chased businesses (Becherer, Finch, & Helms, 2006). Fairlie and Robb, (2009) found inherited businesses are more suc- cessful and larger than non-inherited businesses. Howev- er, because inheritances only made up 1.6 percent of all SME businesses in Fairlie and Robb’s (2009) dataset, the importance in determining broad business outcomes was slight. The findings suggest that management experience 3 M. D. Heileman, & T. L. Pett Journal of Small Business Strategy / Vol. 28, No. 1 (2018) / 1-13 prior to starting or acquiring a business generally improves business outcomes. How the business was acquired by the owner led to our first set of hypotheses: Hypothesis 1A. Family firms that internationalize are more likely to inherit (or receive by transfer) the business com- pared to either family firms with no internationalization or non-family owned firms. Hypothesis 1B. Non-family firms are more likely to buy or found a business compared to family firms. Business Context The business context seeks to provide a description of the environment in which the business operates (Por- ter, 2004). Research has shown that larger businesses tend to survive for a longer time (are older) than smaller com- panies. According to Bercovitz and Mitchell (2007), this is because these firms have a greater business scale and business scope that enhances long-term survival, indepen- dent of baseline profitability, owing to greater availability of financial resources, organizational routines, and external ties. While new ventures tend to quickly exhaust the limit- ed financial resources of founders, family, and friends, they often pursue external startup capital. Plummer, Allison, and Connelly’s (2016) findings imply that a startup firm's char- acteristics and actions are signals that remain relatively un- noticed unless the firm combines them with a third-party affiliation which enhances the signal, consequently increas- ing the likelihood of receiving external startup capital. The context of the franchise industry differs from other indus- tries, for example, a franchisor/franchisee are legally sep- arate entities, professionally and economically dependent (symbiotic), and operationally indistinguishable from each other from the viewpoint of the consumer (Parsa, 1996). Contextual constructs in this study included the year the business was established, amount of startup capital, and whether or not the firm was a franchise. The above discus- sions led to our second hypothesis: Hypothesis 2. Family firms that internationalize tend to be more established (older) and better (higher) funded com- pared to either family firms with no internationalization or non-family-owned firms. Business Performance The relationship between business activities and per- formance of SMEs has long been investigated (Lu & Beam- ish, 2001). Pett and Wolff (2007) developed the theoret- ical arguments for a contingent path relationship among variables representing the environment, capabilities, stra- tegic orientation, and firm performance. The proposition underpinning their study is that internal consistency (or fit) among these contingent relationships yields higher perfor- mance levels. Their findings support the belief that inter- nally consistent resources lead to higher levels of perfor- mance. Mittelstaedt et al. (2003) examined whether there is a minimum size that firms must achieve in order to take advantage of the benefits of exporting. Their argument was built on the contributions of previous research in the areas of SMEs and export success and SMEs in the export devel- opment process (Wolff & Pett, 2000). Analysis of manu- facturing exports indicated that firm size was a necessary and sufficient condition for export success among small manufacturing firms. Wincent (2005) developed and em- pirically tested a framework on how firm size can matter for firm behavior and performance in strategic networks of SMEs. The author’s research considered statistical anal- ysis of standardized questionnaires as well as analysis from face-to-face interviews with managers in SME networks. The findings suggest that firm size can be an important determinant for firm performance. In addition, the study found that networking inside and outside the SME network plays a role. The study suggests that different networking behaviors can have different roles for pursuing corporate entrepreneurship. Blackburn et al. (2013) contributed to the understand- ing of the factors that influence SME performance, in par- ticular, growth. Their study utilized SMEs employing less than 250 employees, which may or may not be similar to family firms. The results suggest that size and age of an en- terprise significantly explain more of performance and they are more important than strategy and the entrepreneurial characteristics of the owner. Heileman et al. (2016) explored the relationship of SME characteristics and performance taking into account the firm size based on number of employees. The results suggest that smaller SMEs have different characteristics and performance regarding owner demographics, business acquisition methods, and business performance metrics compared to larger SMEs. How a business performed in terms of closures, sales, and sales efficiency metrics led to the final set of hypotheses: Hypothesis 3A. Family firms that internationalize tend to have higher performance outcomes than family firms with no internationalization. Hypothesis 3B. Firms that internationalize tend to have higher performance outcomes than firms with no interna- tionalization. Method This current study draws on the US Census Bureau’s Survey of Business Owners (SBO) Public Use Microdata Sample (PUMS). The Census Bureau’s SBO provides a com- prehensive source of information, which is regularly collect- ed, on selected economic and demographic characteristics for businesses and business owners by gender, ethnicity, race, and veteran status. The SBO PUMS file includes na- tional- and state-level data as well as detailed character- istics of businesses and their owners, although protecting 4 M. D. Heileman, & T. L. Pett Journal of Small Business Strategy / Vol. 28, No. 1 (2018) / 1-13 the confidentiality of survey respondents. The SBO is part of the economic census which is collected every five years in years ending in “2” and “7” (US Census Bureau, 2012). The SBO PUMS for this study contains 2,165,680 data ob- servations from the most recent publicly available 2007 survey. The study experimental design involved classifying firms into one of four different categories based on fami- ly-owned firm internationalization. Data The Census Bureau’s PUMS file was created for the 2007 SBO. The SBO collected information on a number of variables related to family and non-family firms regarding gender, ethnicity, race, and veteran status of business own- ers, to name a few. The SBO furthermore collected char- acteristics of businesses and business owners, such as the sources of capital used to start the business and the age of the business owner. The SBO produced estimates for a variety of industry classifications and geographic levels (US Census Bureau, 2012). The SBO included all nonfarm businesses filing Internal Revenue Service tax return forms for individual proprietor- ships, partnerships, or any type of corporation, having re- ceipts of $1,000 or more. The SBO included firms with paid employees and firms with no paid employees. The SBO was conducted on a company or firm basis, rather than on an establishment basis, where a company or firm is a business consisting of one or more domestic establishments that the reporting firm specified under its ownership or control (US Census Bureau, 2012). Design For the purposes of this study, internationalization was defined as firms that have exports or have established operations outside the USA or have outsourced business functions out of the USA, as reported to the SBO. This ap- proach provides a broader understanding of internation- alization as suggested by Patel et al. (2012). Furthermore, family-owned businesses were indicated in the SBO by the respondents. Each firm in the dataset was positioned into one of four categories: (i) family-owned firms with internationalization, (ii) family-owned firms without in- ternationalization, (iii) not family-owned firms with inter- nationalization, or (iv) not family-owned firms without in- ternationalization. These four categories were used as the predictor variables for this study. The sample used for our analysis included firms that meet a minimum business activity expectation to rule out business activities that were deemed casual or side-busi- nesses owned by wage/salary workers, as suggested by other researchers (Fairlie & Robb, 2007; Fairlie & Robb, 2009). Specifically, the business must have operated for at least twelve months and not be a seasonal or occasional business. These criteria resulted in the exclusion of 42.3 percent of the firms in the original dataset. The final data- set used in this study excluded the SBO responses for firms deemed to be casual or side-businesses. The study dataset consisted of 1,096,923 survey observations. Descriptive characteristics and performance measures of the four firm types in our sample are presented in Table 1. The sample size is the number of firms identified in each of the four categories. The average employment, payroll, and sales for each firm category are provided. The SBO pro- vides responses for up to four owners of a business, but for consistency purposes, only characteristics for the first own- er, that is Owner 1, were used in this study. Demographic information about the representative Owner 1 (or prima- ry owner of the business) is provided along with how that owner acquired the business. Business context including the year the business was established, the amount of start- up capital used, and operated as a franchise for each firm type group is presented. The number of businesses which closed (that is, no longer operating at the time of the sur- vey) is reported in each firm category. Two efficiency met- rics in regards to sales per employee and sales per payroll were calculated and are provided in the table. The sales per employee metric was calculated as the firm’s receipts divided by the number of employees in 2007 and the sales per payroll metric was calculated as the firm’s receipts di- vided by the firm’s payroll in 2007. The final dataset used in this study included all SBO responses for firms reporting as either family-owned or not family-owned, excluding casual or side-businesses. The study dataset contained a total of 1,096,923 survey obser- vations. Each firm in the study was placed into one of four categories based on whether or not it was family-owned and whether or not it had internationalized. There were 59,995 family-owned firms with internationalization, 318,885 family-owned firms with no internationalization, 89,995 not family-owned firms with internationalization, and 628,088 not family-owned firms with no internation- alization. The demographic and business acquisition data is analyzed as reported for the first (or primary) owner in the survey. The gender, education, and age averages by firm type are also presented in Table 1. The data indicate that the primary owner tended to be most often male, and that the proportion of male ownership tended to increase as firms internationalize. The data illustrates that the edu- cation level for the primary owner of a firm averaged be- tween some college and an associate’s degree. The data also indicates that the primary owner age for firms aver- aged between 45 and 64 years old. Table 1 contains measures relating to business acqui- sition statistics showing whether the owner founded, pur- chased, inherited, or received the business via a transfer or gift, respectively, by firm type. The data indicates all types of firms were primarily founded by the owner. However, non-family-owned firms were more often founded by the primary owner than family-owned firms. Contextual constructs are presented, which include the year the business was established, amount of startup cap- ital, and whether or not the firm was a franchise. The year 5 M. D. Heileman, & T. L. Pett Journal of Small Business Strategy / Vol. 28, No. 1 (2018) / 1-13 the business was established used a categorical approach which indicates an average between 1980 and 1999 for family-owned firms with internationalization, while fami- ly-owned firms with no internationalization averaged be- tween 1990 and 2002. As for non-family-owned firms with internationalization it was from 1990 to 2002 on average, and for not family-owned firms with no internationalization it averaged from 2000 to 2003. The amount of startup cap- ital averaged between $50,000 and $249,999 for firms with internationalization, while firms with no internationaliza- tion averaged between $25,000 and $99,999 respectively. Very few firms operated as a franchise, but family-owned firms with no internationalization were more likely to op- erate as a franchise than the other firm types as presented in Table 1. The SBO PUMS data contain information on four major business performance outcomes: operating, employment, payroll, and sales. Although none of these measures alone represents a universally agreed upon measure of success, taken together they provide a better view of what it means to be successful in business. The key performance indica- tors for this study included currently operating business and the firm’s sales, as well as the relative efficiency of the firm types in terms of sales per employee and sales per payroll. Results for these four performance measurements across the groups are provided in Table 1. The performance results indicate that family-owned firms with international- ization are less likely to have closed and have higher aver- age sales than any other firm type. The results also suggest that firms with internationalization are generally more effi- cient in regards to sales per employee and sales per payroll than firms without internationalization. Results The SBO response variables regarding owner demo- graphics (that is, gender, education level, and age), busi- ness acquisition (that is, founded, purchased, inherited, or transferred/gifted), and business context (that is, year established, amount of startup capital, and operated as a franchise) are outcome variables in this study. The coding for the SBO response variables is provided in Appendix I. The key performance indicators for this study were business currently operating and the firm’s receipts in 2007, or fundamentally “sales.” In order to consider the rel- ative efficiency of the firm, sales per payroll and sales per Table 1 Firm type group characteristics and performance Family-owned Firms Non Family-owned Firms Internationalization No Internationalization Internationalization No Internationalization Sample Size 59,995 318,885 89,955 628,088 Mean Employment 38.51 21.59 29.58 15.42 Mean Payroll $1,698,960 $712,820 $1,476,070 $574,240 Mean Receipts $11,692,610 $4,067,960 $9,000,790 $2,694,080 Female Owner 10,676 67,199 18,075 177,033 Male Owner 46,993 251,570 70,703 451,016 Owner Founded the Business 34,217 218,993 60,182 484,465 Owner Purchased the Business 13,607 68,809 20,032 105,258 Owner Inherited the Business 6,494 13,900 3,130 10,331 Owner Received Transfer/Gift of Business 5,787 16,541 2,651 12,012 Mean Owner Education Level Some College to Associate's Degree Some College to Associate's Degree Some College to Associate's Degree Some College to Associate's Degree Mean Owner Age 45 to 64 45 to 64 45 to 64 45 to 64 Mean Year Business Established 1980 to 1999 1990 to 2002 1990 to 2002 2000 to 2003 Mean Amount of Startup Capital $50,000 to $249,999 $25,000 to $99,999 $50,000 to $249,999 $25,000 to $99,999 Operated as a Franchise 2,000 16,203 2,247 18,616 Business Closed (Not Operating) 2,812 (5%) 20,314 (6%) 7,742 (9%) 65,856 (10%) Sales per Employee $388,612 $232,609 $440,199 $227,746 Sales per Payroll $8.75 $7.29 $8.99 $6.45 The samples are based on the 2007 Survey of Business Owners (SBO) as described by the Public Use Microdata Sample (PUMS) provid- ed by the US Census Bureau. The data used in this study are the first owner responses only, the business operated at least 12 months, and was not a seasonal or occasional business. 6 M. D. Heileman, & T. L. Pett Journal of Small Business Strategy / Vol. 28, No. 1 (2018) / 1-13 employee were also calculated for performance measures. A data transformation using the natural logarithm func- tion was applied to ratio-level performance measurements (that is, sales, sales per payroll, and sales per employee) in order to improve the interpretability of the results. These four performance measurements provide the principal outcome variables for the study. The outcome variables in- cluding the SBO response variables and the calculated per- formance measures are provided in Table 2. The outcome variable descriptive statistics and correla- tions for all firms included in the study dataset are present- ed in Table 2. For example, the mode of the first variable, gender, was 1 (male), the mean was 0.73, and the standard deviation was 0.44. This result can be interpreted as more male verses female respondents or about 3 to 1 male to fe- male. The noticeable gap between the mode and the mean for the outcome variables startup capital and sales indicate a positively skewed probability distribution for those out- come variables. Comparison of Means In this study we gathered characteristic and perfor- mance information about four firm types, or populations, in order to compare if any differences existed across the four groups. We compared the means across each business characteristic or performance measure for the four firm types listed in Table 3. We calculated a 95 percent confi- dence interval for the population mean for each business characteristic and performance measure for each firm type shown in Table 3. The confidence intervals were computed as follows: Lower Confidence Limit (LCL) = Upper Confidence Limit (UCL) = n s tx + This approach allowed us to determine if there was a statistically significant difference in population means for any two confidence intervals across a particular business characteristic or performance measure when the confi- dence intervals had no overlapping ranges. If the two con- fidence intervals being compared did not have overlapping ranges, then we had evidence there was a statistically sig- nificant difference in the population means at p < 0.05. (An independent samples t-test was used to confirm the statis- tically significant differences in means.) Owner demographics, that is gender, education, and age, were examined to determine if any differences existed across the four groups (see Table 3). Differences between the groups were found on most of the demographic char- acteristics. However, there were some instances when no Table 2 SBO response variables, descriptive statistics, and correlations Variable Mode Mean Std. Dev. 1 2 3 4 5 6 7 8 9 10 11 12 13 Gender 1 0.73 0.44 Education 6 4.08 2.41 0.09 Age 4 3.68 1.77 0.13 0.51 Founded 1 1.09 0.59 0.11 0.43 0.54 Purchased 2 1.57 0.71 0.07 0.53 0.62 0.32 Inherited 2 1.71 0.68 0.11 0.59 0.70 0.65 0.82 Transferred/ Gifted 2 1.71 0.68 0.11 0.59 0.71 0.63 0.82 0.94 Established 3 3.87 2.74 -0.16 -0.12 -0.34 -0.24 -0.08 -0.14 -0.15 Startup$ 1 5.09 3.48 0.02 -0.02 -0.01 0.08 -0.12 -0.10 -0.10 0.00 Franchise 2 1.95 0.28 -0.01 0.00 -0.02 -0.05 0.04 0.00 0.00* 0.01 0.00* Closure 2 1.89 0.31 0.10 0.15 0.15 0.09 0.11 0.12 0.11 -0.23 -0.01 0.02 LN(Sales) 2.3 8.27 10.69 0.03 0.02 0.03 0.04 0.00 0.00 0.00 -0.06 0.05 -0.01 0.03 LN(Sales/ Employee) 4.6 5.56 7.03 0.03 0.02 0.01 0.01 0.00* 0.00 0.00* -0.03 0.03 0.01 0.02 0.25 LN(Sales/ Payroll) 1.1 1.96 3.11 0.02 -0.02 0.00 0.01 -0.01 -0.01 0.00 0.00* 0.02 -0.01 0.00 0.13 0.58 *Not significant at p < 0.05 (2-tailed). 7 M. D. Heileman, & T. L. Pett Journal of Small Business Strategy / Vol. 28, No. 1 (2018) / 1-13 Ta bl e 3 Pr od uc tio n of 9 5% c on fid en ce in te rv al s fo r t he m ea ns o f b us in es s ch ar ac te ri sti cs a nd p er fo rm an ce Fa m ily -o w ne d Fi rm s N ot F am ily -o w ne d Fi rm s In te rn ati on al iz ati on N o In te rn ati on al iz ati on In te rn ati on al iz ati on N o In te rn ati on al iz ati on M ea n St d. D ev . Va lid O bs . LC L U CL M ea n St d. D ev . Va lid O bs . LC L U CL M ea n St d. D ev . Va lid O bs . LC L U CL M ea n St d. D ev . Va lid O bs . LC L U CL O w ne r D em og ra ph ic s: G en de r 0. 81 0. 39 57 ,6 69 0. 81 0. 82 0. 79 0. 41 31 8, 76 9 0. 79 0. 79 0. 80 0. 40 88 ,7 78 0. 79 0. 80 0. 72 0. 45 62 8, 04 9 0. 72 0. 72 Ed uc ati on 4. 67 2. 02 59 ,9 95 4. 66 4. 69 4. 43 1. 97 31 8, 88 5 4. 43 4. 44 4. 72 2. 14 89 ,9 55 4. 70 4. 73 4. 66 2. 07 62 8, 08 8 4. 65 4. 66 A ge 4. 35 1. 32 59 ,9 95 4. 34 4. 36 4. 31 1. 21 31 8, 88 5 4. 30 4. 31 4. 05 1. 40 89 ,9 55 4. 04 4. 05 4. 03 1. 30 62 8, 08 8 4. 03 4. 04 Bu si ne ss A cq ui si tio n: Fo un de d 1. 34 0. 56 59 ,9 93 1. 34 1. 34 1. 30 0. 47 31 8, 88 5 1. 29 1. 30 1. 21 0. 53 89 ,9 19 1. 21 1. 22 1. 17 0. 45 62 8, 08 8 1. 17 1. 18 Pu rc ha se d 1. 70 0. 53 59 ,9 93 1. 70 1. 71 1. 77 0. 44 31 8, 88 5 1. 77 1. 77 1. 67 0. 57 89 ,9 19 1. 67 1. 68 1. 78 0. 48 62 8, 08 8 1. 78 1. 78 In he ri te d 1. 86 0. 38 59 ,9 93 1. 86 1. 87 1. 94 0. 27 31 8, 88 5 1. 94 1. 94 1. 87 0. 46 89 ,9 35 1. 86 1. 87 1. 93 0. 35 62 8, 08 8 1. 93 1. 93 Tr ns fr d/ G ift ed 1. 88 0. 37 59 ,9 93 1. 87 1. 88 1. 93 0. 28 31 8, 88 5 1. 93 1. 93 1. 87 0. 46 89 ,9 36 1. 87 1. 88 1. 93 0. 35 62 8, 08 8 1. 93 1. 93 Bu si ne ss Co nt ex t: Es ta bl is he d 2. 56 2. 19 59 ,9 95 2. 54 2. 58 3. 37 2. 44 31 8, 88 5 3. 36 3. 38 3. 63 2. 78 89 ,9 55 3. 62 3. 65 4. 03 2. 68 62 8, 08 8 4. 02 4. 03 St ar tu p$ 5. 63 3. 21 59 ,9 95 5. 60 5. 65 4. 98 3. 19 31 8, 88 5 4. 97 4. 99 5. 66 3. 32 89 ,9 55 5. 64 5. 69 4. 90 3. 52 62 8, 08 8 4. 89 4. 91 Fr an ch is e 2. 01 0. 56 59 ,7 28 2. 00 2. 01 1. 94 0. 28 31 8, 87 6 1. 94 1. 94 1. 97 0. 28 89 ,7 92 1. 97 1. 97 1. 96 0. 24 62 8, 08 0 1. 96 1. 96 Bu si ne ss Pe rf or m an ce : Cl os ur e 1. 95 0. 21 59 ,8 44 1. 95 1. 95 1. 94 0. 24 31 8, 10 0 1. 94 1. 94 1. 91 0. 28 89 ,5 81 1. 91 1. 92 1. 89 0. 31 62 6, 21 0 1. 89 1. 90 LN (S al es ) 16 .2 7 18 .0 6 59 ,9 95 16 .2 3 16 .3 2 15 .2 2 17 .1 0 31 8, 88 5 15 .2 0 15 .2 4 16 .0 1 18 .0 2 89 ,9 55 15 .9 6 16 .0 6 14 .8 1 17 .2 2 62 8, 08 8 14 .7 8 14 .8 3 LN (S al es /E m pl y) 12 .8 7 14 .2 1 44 ,4 80 12 .8 3 12 .9 1 12 .3 6 13 .6 4 19 5, 00 5 12 .3 4 12 .3 7 12 .9 9 14 .4 4 56 ,9 76 12 .9 6 13 .0 3 12 .3 4 13 .8 4 31 8, 87 8 12 .3 2 12 .3 5 LN (S al es /P ay ro ll) 2. 17 2. 99 46 ,2 94 2. 15 2. 19 1. 99 3. 01 20 7, 79 4 1. 98 2. 00 2. 20 3. 31 60 ,2 42 2. 17 2. 22 1. 86 3. 11 34 7, 21 2 1. 85 1. 88 8 M. D. Heileman, & T. L. Pett Journal of Small Business Strategy / Vol. 28, No. 1 (2018) / 1-13 difference between the groups was found. For example, by examining the confidence intervals for gender between family firms with no internationalization and non-fami- ly firms with internationalization, the results suggest no significant differences as indicated by the overlap of the confidence limits between the two groups. Overall, the re- sults suggest owners of family firms that internationalized tended to be more male, better educated, and older; while owners of non-family firms with no internationalization had the lowest values for gender (less male), education, and age (youngest) across the four groups. Business acquisition was next examined by examining the differences for the variables founded, purchased, inher- ited, and transferred/gifted across the four groups—those results are presented in Table 3. A mean lower score indi- cates agreement. As an example, for the variable, founded the business, non-family firms with no internationalization had the lowest mean score (1.17). Likewise, non-family firms with internationalization had the lowest mean score (1.67) for purchased. Family firms with internationalization had the lowest score for inherited (1.86) and similar low score for transfer/gifted (1.88) with non-family firms with internationalization when examining the confidence inter- vals for the two groups. These results support both Hypothesis 1A and 1B. Hy- pothesis 1A suggests that family firms that international- ize would be more likely to inherit or receive by transfer/ gift the business compared to either family firms with no internationalization or non-family owned firms. Because internationalization often takes resources, organizational commitment, and time to carryout, these businesses we suggest would more likely be transferred from generation to generation. The results support this hypothesis. Hypoth- esis 1B suggests that non-family firms are more likely to buy or start a business compared to family firms. These results also support this hypothesis. Business context was examined in terms of when es- tablished (year), amount of startup capital, and whether or not the business was a franchise. These contextual vari- ables help to understand how these firms operate. Family firms with internationalization scored significantly lower on established (2.56) than the other three groups. This sug- gests that family firms that internationalize are the oldest firms. Firms that had internationalized, either family (5.63) or not family (5.66), had significantly more startup capital than firms with no internationalization presence. Finally, family firms that internationalized tended to have fewer franchisees (2.01) compared to all groups. These results partially support Hypothesis 2 which states family firms that internationalize are more established (older) and bet- ter (higher) funded compared to either family firms with no internationalization or non-family firms. The results sug- gest that all firms that internationalized had more startup capital compared to those that did not internationalize. Finally, we wanted to examine the difference in per- formance across the four groups; with performance being characterized beyond just financial performance to include business survival. Survival of the firm, which is the inverse of closure, is significantly healthier for family firms (1.95) compared to other groups. A line chart of the means for the business closure performance measure is given in Fig- ure 1. The data point markers indicate the mean business closures reported (that is, yes or no) for each firm group. The business closure means are significantly different for each firm group. The data suggests that family-owned firms with internationalization go out of business less fre- quently than any other firm group, and that family-owned businesses generally go out of business less often than not family-owned businesses. Family firms with internationalization sales (trans- formed by the natural logarithm) were significantly higher (16.27) than any other group. A line chart for the means for the natural logarithm of sales performance measure is provided in Figure 2. The data point markers illustrate the mean receipts for each firm group. The means for the nat- ural logarithm of sales are significantly different for each firm group. The data suggests that family-owned firms with internationalization have the highest sales on average, and that firms with internationalization generally have higher sales than firms without internationalization. Sales per employee was measured for each of the four groups; here non-family firms that internationalized scored the highest (12.99) followed by family firms that interna- tionalized (12.87). A line chart illustrates the means for the natural logarithm of sales per employee performance measure in Figure 3. The data point markers indicate the mean sales per employee for each firm group. The means for the sales per employee are not significantly different for family-owned firms with no internationalization and not family-owned firms with no internationalization (which is indicated in Figure 3 with the circle around the data point markers). While the data suggests that not family-owned firms with internationalization have the most efficient sales per employee, and that firms with internationalization gen- erally have more efficient sales per employee than firms without internationalization. The results illustrate the sig- nificant differences between those firms that international- ize compared to firms that do not internationalize. Finally, the means for the natural logarithm of sales per payroll performance measure suggest that again, interna- tionalization matters. Non-family firms with internation- alization scored the highest (2.20) and family firms with internationalization scored (2.17) with the confidence in- terval suggesting no difference between these two groups. These results are also illustrated in Figure 4. The data point markers indicate the mean sales per payroll for each firm group. Again, the means for the sales per payroll are not significantly different for family-owned firms with inter- nationalization and not family-owned firms with interna- tionalization (which is indicated in Figure 4 with the circle around the data point markers). This suggests that firms with internationalization have the most efficient sales per payroll, and that not family-owned firms with no interna- tionalization generally have the least efficient sales per 9 M. D. Heileman, & T. L. Pett Journal of Small Business Strategy / Vol. 28, No. 1 (2018) / 1-13 1.89 1.90 1.91 1.92 1.93 1.94 1.95 1.96 Means Plot for Business Closure Family-owned Firm w/ Internationalization Family-owned Firm w/o Internationalization Not Family-owned Firm w/ Internationalization Not Family-owned Firm w/o Internationalization Figure 1. Means plot for the business closure performance measure. payroll. These results taken together suggest that overall fami- ly firms that internationalize have better performance out- comes than other firms. These results support Hypothesis 3A which suggests when family firms internationalize they will experience higher performance outcomes than fami- ly firms with no internationalization. The results also sug- gest those firms that internationalize have better perfor- mance outcomes than firms that do not internationalize, regardless of whether or not the firms are family-owned or non-family-owned. These results support Hypothesis 3B which states that any firms that internationalize will have higher performance than those firms that do not interna- tionalize. Implications and Conclusion The limitations of this study are the constraints on the generalizability, the applications to practice, and the utility of findings that result from the ways in which we principally chose to design the study. The survey data for this study was obtained from the Census Bureau’s 2007 SBO PUMS. We excluded data from firms that did not meet a minimum business activity expectation. To be included, the business must have operated for at least twelve months and must not be a seasonal business or an occasional business. The SBO collected demographic information for at least one owner and up to four owners of each business firm. We designated Owner 1 as the primary owner of interest for this study, thus excluding data and analyses for Owner 2, Owner 3, and Owner 4 if that data were reported. The SBO covers both firms with paid employees and firms with no paid employees. For business performance evaluation pur- poses, we excluded data from firms with no employees or with no payroll in order to determine finite values for the business efficiency metrics (that is, sales per employee and sales per payroll). This study contributes to understanding the character- istics and performance between family-owned firms and non-family-owned firms, both with internationalization and without internationalization. A novel feature of the study experimental design was the incorporation of prima- ry owner characteristics and whether there were any busi- ness acquisition and business performance correlations. The findings suggest that internationalization (as defined in this study) of family businesses improves performance on a number of fronts, including reduced closures and increased 10 M. D. Heileman, & T. L. Pett Journal of Small Business Strategy / Vol. 28, No. 1 (2018) / 1-13 14.5 15.0 15.5 16.0 16.5 Means Plot for LN(Sales) Family-owned Firm w/ Internationalization Family-owned Firm w/o Internationalization Not Family-owned Firm w/ Internationalization Not Family-owned Firm w/o Internationalization Figure 2. Means plot for the natural logarithm of sales performance measure. sales. The research implications indicate opportunity for future research using the SBO PUMS data. For instance, we could extend the research to include all reported owners of a firm, as well as to estimate the effect of each business characteristic on the business performance differences be- tween family-owned businesses with internationalization, family-owned firms without internationalization, non-fam- ily-owned firms with internationalization, and non-fami- ly-owned firms without internationalization. Furthermore, we could examine the industries of the different firms and test for significant differences among the industries. The SBO is part of the economic census program, which the Census Bureau is legally required to conduct every five years (US Congress, 2009), in the years ending in “2” and “7.” The Census Bureau combines data from the SBO with data from other sources, including economic surveys, economic censuses, and administrative records. The SBO is the primary source of statistics about the demographic characteristics of the owners of approximately 28 million American businesses, together with how the business is organized and its activities. Many economic and social re- searchers routinely use the SBO statistics (US Census Bu- reau, 2012). Understanding the characteristics of family-owned firms and internationalization is essential to better under- stand the competitive nature of today’s businesses as well as for policymakers. This study is policy relevant given the importance of successful family business ownership for job creation, economic growth, income generation, and wealth accumulation. The findings suggest practical implications for business growth strategy with regards to exporting, es- tablishing operations outside the USA, or outsourcing busi- ness functions out of the USA. Although these results are based on what could be de- scribed as a unique measure of a family business, that is, two family members owning more than fifty percent of the business, other measures along this line may provide addi- tional insight on defining a family firm. Likewise, we used 11 M. D. Heileman, & T. L. Pett Journal of Small Business Strategy / Vol. 28, No. 1 (2018) / 1-13 12.3 12.4 12.5 12.6 12.7 12.8 12.9 13.0 Means Plot for LN(Sales/Employee) Family-owned Firm w/ Internationalization Family-owned Firm w/o Internationalization Not Family-owned Firm w/ Internationalization Not Family-owned Firm w/o Internationalization Figure 3. Means plot for the natural logarithm of sales per employee performance measure. 1.8 1.9 2.0 2.1 2.2 Means Plot for LN(Sales/Payroll) Family-owned Firm w/ Internationalization Family-owned Firm w/o Internationalization Not Family-owned Firm w/ Internationalization Not Family-owned Firm w/o Internationalization Figure 4. Means plot for the natural logarithm of sales per payroll performance measure. 12 M. D. Heileman, & T. L. Pett Journal of Small Business Strategy / Vol. 28, No. 1 (2018) / 1-13 a more holistic view of internationalization of the family firm including have exports or have established operations outside the USA or have outsourced business functions to define internationalization. Future research could examine the generalizability of these findings in other settings and countries. References Abdellatif, M., Amann, B., & Jaussaud, J. (2010). Family ver- sus nonfamily business: A comparison of interna- tional strategies. Journal of Family Business Strate- gy, 1(2), 108–116. Becherer, R. C., Finch, J. H., & Helms, M. M. (2006). The influences of entrepreneurial motivation and new business acquisition on strategic decision making. Journal of Small Business Strategy, 16(2), 1–13. Bercovitz, J., & Mitchell, W. (2007). When is more better? The impact of business scale and scope on long- term business survival, while controlling for prof- itability. Strategic Management Journal, 28(1), 61–79. Blackburn, R. A., Hart, M., and Wainwright, T. (2013). Small business performance: Business, strategy and owner-manager characteristics. Journal of Small Business and Enterprise Development, 20(1), 8–27. Bose, T. K. (2016). Critical success factors of SME interna- tionalization. Journal of Small Business Strategy, 26(2), 87–109. Carr, J. C., Chrisman, J. J., Chua, J. H., & Steier, L. P. (2016). Family firm challenges in intergenerational wealth transfer. Entrepreneurship Theory and Practice, 40(6), 1197–1208. Chang, E. P. C., Memili, E., Chrisman, J. J., & Welsh, D. H. B. (2011). What can drive successful entrepreneurial firms? An analysis of Inc. 500 companies. Journal of Small Business Strategy, 22(2), 27–49. Chua, J. H., Chrisman, J. J., & Sharma, P. (1999). Defining the family business by behavior. Entrepreneurship Theory and Practice, 23(4), 19–39. Cieślik, J., Kaciak, E., & Welsh, D. H. B. (2010). The effect of early internationalization on survival, consistency, and growth of export sales. Journal of Small Busi- ness Strategy, 21(1), 39–64. Fairlie, R. W., & Robb, A. M. (2007). Why are black-owned businesses less successful than white-owned busi- nesses: The role of families, inheritances, and busi- ness human capital. Journal of Labor Economics, 25(2), 289–323. Fairlie, R. W., & Robb, A. M. (2009). Gender differences in business performance: Evidence from the Charac- teristics of Business Owners survey. Small Business Economics, 33, 375–395. Graves, C., & Shan, Y. G. (2014). An empirical analysis of the effect of internationalization on the performance of unlisted family and nonfamily firms in Australia. Family Business Review, 27(2), 142–160. Heileman, M. D., Pett, T. L., & Mayer, A. J. (2016). Small- to medium-sized enterprise characteristics and per- formance: An exploratory examination of the Cen- sus Bureau’s Survey of Business Owners. Journal of Business and Entrepreneurship, 28(1), 145–164. Kontinen, T., & Ojala, A. (2010). The internationalization of family businesses: A review of extant research. Journal of Family Business Strategy, 1(2), 97–107. Kraus, S., Mensching, H., Calabrò, A., Cheng, C. F., & Fils- er, M. (2016). Family firm internationalization: A configurational approach. Journal of Business Re- search, 69(11), 5473–5478. Lichtenstein, J. (2014). Demographic characteristics of busi- ness owners. Small Business Administration Office of Advocacy, Issue Brief, (2). Lu, J. W., & Beamish, P. W. (2001). The internationalization and performance of SMEs. Strategic Management Journal, 22(6–7), 565–586. Mittelstaedt, J., Harben, G., & Ward, W. (2003). How small is too small? Firm size as a barrier to exporting from the United States. Journal of Small Business Management, 14(1), 68–84. Parsa, H. G. (1996). Franchisor-franchisee relationships in quick-service-restaurant systems. Cornell Hotel and Restaurant Administration Quarterly, 37(3), 42–49. Patel, V. K., Pieper, T. M., & Hair, J. F. (2012). The global fam- ily business: Challenges and drivers for cross-bor- der growth. Business Horizons, 55(3), 231–239. Pett, T. L., & Wolff, J. A. (2007). SME performance: A case for internal consistency. Journal of Small Business Strategy, 18(1), 1–16. Pett, T. L., & Wolff, J. A. (2016). Entrepreneurial orientation and learning in high and low-performing SMEs. Journal of Small Business Strategy, 26(2), 71–86. Plummer, L., Allison, T., & Connelly, B. (2016). Better togeth- er? Signaling interactions in new venture pursuit of initial external capital. Academy of Management Journal, 59(5), 1585–1604. Porter, M. E. (2004). Competitive strategy: Techniques for analyzing industries and competitors. New York, NY: Free Press. Pukall, T. J., & Calabrò, A. (2014). The internationalization of family firms: A critical review and integrative model. Family Business Review, 27(2), 103–125. Reuber, A. R. (2016). An assemblage-theoretic perspective on the internationalization processes of family 13 M. D. Heileman, & T. L. Pett Journal of Small Business Strategy / Vol. 28, No. 1 (2018) / 1-13 firms. Entrepreneurship Theory and Practice, 40(6), 1269–1286. US Census Bureau. (2012). Survey of Business Owners (SBO). Retrieved from www.census.gov/pro- grams-surveys/sbo.html. US Congress. (2009). Title 13, United States code—cen- sus. Retrieved from www.gpo.gov/fdsys/pkg/US- CODE-2009-title13/html/USCODE-2009-title13. htm. Wincent, J. (2005). Does size matter?: A study of firm be- havior and outcomes in strategic SME networks. Journal of Small Business and Enterprise Develop- ment, 12(3), 437–453. Wolff, J. A., & Pett, T. L. (2000). Internationalization of small firms: An examination of export competitive pat- terns, firm size, and export performance. Journal of Small Business Management, 38(2), 34–47. APPENDIX I SBO response variable coding. Name Code Gender 0 = Female 1 = Male Education 0 = Not Reported 1 = Less than High School 2 = High School 3 = Technical School 4 = Some College 5 = Associate's 6 = Bachelor's 7 = Master's + Age 0 = Not Reported 1 = Under 25 2 = 25 to 34 3 = 35 to 44 4 = 45 to 54 5 = 55 to 64 6 = 65 or over Founded 0 = Not Reported 1 = Yes 2 = No Purchased 0 = Not Reported 1 = Yes 2 = No Inherited 0 = Not Reported 1 = Yes 2 = No Transferred/Gifted 0 = Not Reported 1 = Yes 2 = No Established 0 = Not Reported 1 = Before 1980 2 = 1980 to 1989 3 = 1990 to 1999 4 = 2000 to 2002 5 = 2003 6 = 2004 7 = 2005 8 = 2006 9 = 2007 10 = Don't Know Startup$ 0 = Not Reported 1 = Less than $5,000 2 = $5,000 to $9,999 3 = $10,000 to $24,999 4 = $25,000 to $49,999 5 = $50,000 to $99,999 6 = $100,000 to $249,999 7 = $250,000 to $999,999 8 = $1,000,000 or more 9 = Don't Know 10 = Not Applicable Franchise 0 = Not Reported 1 = Yes 2 = No 3 = Franchiser owned portion Closure 0 = Not Reported 1 = Yes 2 = No Name Code