Appendix A: Methods

The following question guided this study: Is there a significant difference in service quality between providers in states with no or lighter license requirements and those in neighboring states with more burdensome requirements?

Data

To measure quality, we relied on Yelp business ratings, which others have found to accurately reflect service quality. ¹ These ratings span the period from October 2004 through October 2020 for locksmiths and from October 2004 through June/July 2019 for all other occupations, although not all businesses may have had ratings in every year of those time spans. All ratings are on a five-star scale, with one star being the worst rating a business can receive and five stars the best. Businesses on Yelp also have written reviews, but our analysis used only the numerical rating. For each business, the rating represents the average across the entire time span. Average ratings are not available by year.

Sample

The sampling unit (and the unit of analysis) was businesses within specific occupations. To determine the sample of occupations and states in our study, we used data from the second edition of the Institute for Justice’s License to Work report to identify occupations with stark licensing differences across neighboring states. ² Because IJ released the second edition of License to Work in 2017, we checked relevant licensing requirements to ensure nothing significant had changed in the intervening years. 

Licensing is not limited to state laws. Counties and cities can and often do enact their own licensing laws. ³ Failing to account for such laws, where they exist, would produce spurious analytical results. We therefore also examined a small sample of cities in border counties in our states of interest to determine whether local licensing was present (it was not). We also examined salon/shop licensing at the state and local levels for barbering- and cosmetology-related occupations as this, too, can sometimes affect licensing requirements for workers. However, such requirements had no impact on our analyses.

We limited the businesses in our sample to those within narrow bandwidths on either side of state borders. For each occupation-state comparison, bandwidths were determined independently and automatically for each comparison using MSE-optimal procedures in Stata’s rdrobust program. Consequently, bandwidths differ by occupation-state comparison (see Table A1). Before creation of the bandwidths, comparisons that used the CA-NV border (interior designer and tree trimmer) were first narrowed only to counties in the Lake Tahoe region. This is due to a lack of businesses anywhere else along the states’ shared border. ⁴ These counties provided a sufficient number of firms for our analyses, while other regions along the CA-NV border were too rural to do so. Final sample sizes for businesses by occupation-state comparisons are presented in Table A2. 

Table A1: Final Comparisons and Bandwidths

Occupation	States	Bandwidth
Barber	NJ to PA	21 miles
Cosmetologist	NY to CT	16 miles
Cosmetologist	NY to NJ	16 miles
Interior Designer∗	CA to NV	11 miles
Locksmith	PA to NJ	12 miles
Manicurist†	CT to MA	23 miles
Manicurist	CT to NY	17 miles
Tree Trimmer	NV to CA	19 miles
Tree Trimmer	VA to MD	19 miles

Table A2: Number of Businesses for Each Occupation-State Comparison

	Lower Burden/Unlicensed		Higher Burden/Licensed		Total
Occupation	State	# of Firms	State	# of Firms	# of Firms
Barber	NJ	237	PA	507	744
Cosmetologist	NY	1,241	CT	120	1,361
Cosmetologist	NY	1,234	NJ	610	1,844
Interior Designer	CA	16	NV	20	36
Locksmith	PA	169	NJ	157	326
Manicurist	CT	282	MA	106	388
Manicurist	CT	185	NY	226	411
Tree Trimmer	NV	31	CA	15	46
Tree Trimmer	VA	83	MD	81	164

Analysis

We analyzed these data using regression models, treating individual businesses as the unit of analysis. In the analyses, we employed a geographic regression discontinuity design to isolate the potential effect of high licensing burdens in counties bordering either unlicensed or less burdensomely licensed states. ⁵ The areas in which these businesses are located should be similar except for variation in state licensing requirements. In geographic discontinuity, if a license were an important determinant of service quality, we would expect to find a measurable difference in service quality between states when comparing businesses across borders.

We ran separate regressions for each occupation-state comparison. This allowed for a clear interpretation of a specific license’s effect rather than the general effect of all the licenses captured in our models, which vary in their burden. We also performed manipulation testing on density discontinuity. ⁶ This addressed the common concern in geographic regression discontinuity that firms are not randomly distributed around the border (in this case) and may, instead, show evidence of self-selection or nonrandom sorting into control and treatment status. Specifically, businesses may choose, for example, to locate more frequently on the unlicensed side of the border, and those that do may have some characteristics that may produce a systematic effect on service quality. The implication is that those characteristics—rather than licensing—may explain differences, or lack thereof, in service quality. Density checks revealed an inconsistent pattern, with a mixture of nonsignificance, greater density in licensed states and greater density in unlicensed states. If anything, the clearer trend showed greater density in the more populated state in each comparison (e.g., greater density in NY compared to CT for the manicurist comparison). Thus, results do not suggest a clear indication of selection bias.