Young startup firm exports and productive efficiency

ABSTRACT

We employ the United States Census Bureau’s 2007 Survey of Business Owners to examine the relationship between the productive efficiency of startup firms and their level of exports. We find, consistent with previous work in the literature, that a small firm’s exports are positively related to their level of productive efficiency. This result holds after controlling for various individual owner, as well as firm, characteristics. Our results are also robust to the inclusion of a two-stage estimation strategy to control for the effects of endogeneity in the efficiency–exports relationship.

KEYWORDS:

• Production Efficiency
• startup firm
• exports

JEL CLASSIFICATION:

• F23
• F29
• L26

Acknowledgements

We would like to thank Richard Moore at the Census Bureau for his help with the Survey of Business Owners. We would additionally like to thank David Cho for comments. All mistakes are our own.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Recent studies utilizing the SBO include Mora and Davila (2014) and Wang and Liu (2015).

² Race is coded as first Hispanic, then any of White, Asian, Black, American Indian, Native Hawaiian or Pacific Islander, or Other race. Only firms with a primary owner who identified themselves as belonging to one race – as opposed to multiple races – are retained for the purpose of the present analysis. This restriction is made to simplify the intuition regarding the effect of owner demographics. Work experience includes either self-employed or owning a business. Age and Education were coded in a continuous fashion, although binary indicators had little effect in changing the results. The target customer reports address whether the firm sells more than 10% of their product to (a) the Federal government, (b) State and Local governments, (c) other businesses and (d) individuals.

³ We would have preferred to employ total current firm capital rather than solely employing the firm’s startup capital. Unfortunately, the SBO limits information on this front to either (a) startup capital or (b) expansion capital. Expansion capital is only provided in a binary fashion – existence rather than the specific amount. When we attempted to condition on whether or not the firm had employed expansion capital, we encountered issues of regression convergence for firms that did not have any expansion in place. It is also true that, even with such a condition in place, selection into firms using startup or additionally expansion capital would be a relevant concern. We have noted these points as a caveat.

⁴ A firm is defined as an ‘employer firm’ if, at the time of the survey, it had a payroll and a nonzero number of employees. It is also true that older firms and firms with no establishment age were not used in the comparison since their characteristics were, in principle, simply too different from the characteristics of interest here.

⁵ The increase in the coefficient magnitude may occur because we are catching a failure to conform to the exclusion restriction (discussed below), however, we believe we are seeing an indication that firms that export may also have a certain degree of inefficiency in selecting into exporting as a ‘last resort’. By using instruments that control for underlying abilities, we should be teasing out a firm’s inherent propensity to export, rather than its decision to do so as an involuntary decision. In the process, we will clean the relationship from the negative part of selection.