A new study coauthored by professor Michael Luca of Harvard Business School and Dara Lee Luca of Mathematica Policy Research shows that an increase in minimum wage leads to an increase in the probability that a restaurant with a lower star rating will go out of business. While prior research in this area mainly used government data sets and costly surveys, this study shows that Yelp data can further the public understanding of the economy – both through more up-to-date reporting and a unique perspective on business attributes.
Disproportionate impact on lower-rated restaurants
The Harvard study looked at changes to the minimum wage in the San Francisco Bay area over nine years (from 2008-2016, there have been 21 such changes). It found that higher minimum wages led to an increased exit rate for restaurants, which traditionally have the highest percentage of employees at the minimum wage level. However, restaurants rated 3.5 stars or below out of a maximum of five are disproportionately impacted by increases to the minimum wage. For example, a one-dollar increase in the minimum wage leads to a 14% increase in the likelihood of exit for the median restaurant (which has 3.5 stars) but no detectable effect for five-star restaurants.
One factor explaining this difference is that lower-rated businesses are already more likely to exit, so higher minimum wage is more of a burden for this subset of businesses. Higher-rated businesses, on the other hand, may be able to more easily pay higher wages. For more details on why the researchers may be seeing this pattern, read the study abstract and full paper.
Implications for nationwide minimum wage increase
The current federal minimum wage has been stagnant at $7.25 per hour. This study suggests that an increase in the minimum wage could have a measurable impact on local businesses.
To think through the implications of this study, it’s helpful to have a sense of what policies are on the table at the local and federal level. Many policy makers are pushing for higher federal minimums, as states – and now cities – continue to set their own binding minimums. This would be a large change, leading to an average of a $3.50 per hour increase relative to current state minimums. The estimates in the paper suggest that if a federal minimum wage of $12 per hour were implemented (with no credit for tipped employees), an additional 2% of restaurants nationwide might close, relative to keeping the current state minimums for each state. With more than 600,000 nationwide restaurants (according to the Bureau of Labor Statistics), this amounts to more than 12,000 additional closings. However, the vast majority of these closings would be among restaurants in the bottom half of the quality distribution, and 4.5 and 5 star restaurants would see virtually no closings.
However, while some businesses may close, other research has shown that increases to the minimum wage have had modest or no impact on total employment – potentially because higher-rated or larger businesses are insulated from this shock, and potentially because business shifts to the remaining businesses which could lead to increased hiring.
Benefits of using Yelp data for economic research
Government data sets collected by agencies such as the Bureau of Labor Statistics and Census Bureau can be complemented by data from sources like Yelp. In another recent paper, “Big Data and Big Cities: The Promise and Limitations of Improved Measures for Urban Life,” Luca and his coauthors at Harvard and MIT describe some of the ways in which new data sources can improve our understanding of the economy. Expanding on these insights, we see Yelp as having three distinct benefits for researchers, relative to relying only on traditional government data sets.
Bureau of Labor Statistics and economic census data only become available after a lag that can range from months to years. Yelp data is available in near real-time, allowing researchers and policymakers to more quickly understand the impacts of different economic policies and act faster.
Whereas the public versions of the economic census and BLS data are aggregated to the zip code or county level, Yelp data can get as specific as city, neighborhood, or street.
The growth of online review platforms like Yelp allow for unique insights into the economy provided by consumers and businesses themselves, including content such as star ratings, photos, and reviews.
More generally, there is a vast opportunity available in using Yelp data to understand the economy. As part of Yelp’s ongoing economic research initiative, we will be working closely with Professor Luca (who is leading the academic part of this initiative) and other researchers to provide Yelp data and support for economic analyses. We also have an internal effort, led by Carl Bialik, who recently joined us from FiveThirtyEight, to explore the various ways that Yelp data can improve our understanding of the world around us. Taken together, we hope to surface insights that add value to policymakers, businesses, and consumers alike.
Local business takeaway
It’s tempting to think that local businesses would fear minimum wage increases. But multiple surveys over time have suggested that the majority of businesses actually favor increases to the minimum wage. There are many possible reasons for this, including a genuine desire of business owners to do right by their employees. Higher minimum wages can also allow a business to pay generous wages without fear of a nearby competitor gaining an edge through low wages. For local businesses concerned about the impact of the minimum wage, this study suggests that the effect is smaller for higher rated businesses. Yelp recommends businesses get engaged by claiming their free Yelp business page, uploading information and responding to reviews for free, and focusing on providing great customer service as a starting point for improving their online reputation. Businesses can find more tips and get started at biz.yelp.com.
For policymakers interested in what Yelp data can reveal about your local economy, we’d love to work with you. Reach us at firstname.lastname@example.org to connect with our data team.