This paper examines the effects of minimum wages on earnings and employment in restaurants and other low-wage sectors using policy discontinuities at state borders. By analyzing all contiguous county-pairs in the United States that straddle a state border, the study finds no adverse employment effects of minimum wage increases. Traditional methods that do not account for local economic conditions tend to produce spurious negative effects due to spatial heterogeneities in employment trends unrelated to minimum wage policies. The findings are robust to allowing for long-term effects of minimum wage changes. The study compares two approaches: local case studies and traditional national-level studies. Local case studies, which use policy discontinuities at state borders, show strong and similar positive effects of minimum wages on restaurant earnings, but employment effects are indistinguishable from zero. Traditional fixed-effects specifications, on the other hand, produce negative employment elasticities. The difference between these findings has important welfare implications, as traditional estimates suggest minimum wage increases do not raise the aggregate earnings of affected workers much, while local estimates suggest they substantially raise total earnings. The study reconciles conflicting results by showing that negative employment effects in national-level studies reflect spatial heterogeneity and improper control groups. It finds that in traditional fixed-effects specifications, employment levels and trends are negative prior to the minimum wage increase, while in local specifications, they are close to zero, indicating valid controls. The study also rejects other explanations for divergent findings, such as anticipation or lagged effects of minimum wage increases. The study finds that the lack of an employment effect is not restricted to restaurants but is also absent in other low-wage industries. The results suggest that national-level estimates suffer from omitted variables bias due to spatial heterogeneity, which is effectively controlled for by local estimates. The study also shows that standard errors from individual case studies often overstate their precision, emphasizing the importance of pooling across case studies for reliable inference. The study uses data from the Quarterly Census of Employment and Wages (QCEW) and finds that minimum wage effects on earnings are positive, while employment effects are not significant. The study also finds that the dynamic response to minimum wage increases is positive in the short term, with cumulative effects ranging from 0.215 to 0.316. The study concludes that minimum wage increases have positive effects on earnings but no significant adverse effects on employment, and that traditional fixed-effects specifications are biased due to spatial heterogeneity. The study provides robust evidence that minimum wage increases do not lead to significant job losses and that local estimates are more reliable for estimating minimum wage effects.This paper examines the effects of minimum wages on earnings and employment in restaurants and other low-wage sectors using policy discontinuities at state borders. By analyzing all contiguous county-pairs in the United States that straddle a state border, the study finds no adverse employment effects of minimum wage increases. Traditional methods that do not account for local economic conditions tend to produce spurious negative effects due to spatial heterogeneities in employment trends unrelated to minimum wage policies. The findings are robust to allowing for long-term effects of minimum wage changes. The study compares two approaches: local case studies and traditional national-level studies. Local case studies, which use policy discontinuities at state borders, show strong and similar positive effects of minimum wages on restaurant earnings, but employment effects are indistinguishable from zero. Traditional fixed-effects specifications, on the other hand, produce negative employment elasticities. The difference between these findings has important welfare implications, as traditional estimates suggest minimum wage increases do not raise the aggregate earnings of affected workers much, while local estimates suggest they substantially raise total earnings. The study reconciles conflicting results by showing that negative employment effects in national-level studies reflect spatial heterogeneity and improper control groups. It finds that in traditional fixed-effects specifications, employment levels and trends are negative prior to the minimum wage increase, while in local specifications, they are close to zero, indicating valid controls. The study also rejects other explanations for divergent findings, such as anticipation or lagged effects of minimum wage increases. The study finds that the lack of an employment effect is not restricted to restaurants but is also absent in other low-wage industries. The results suggest that national-level estimates suffer from omitted variables bias due to spatial heterogeneity, which is effectively controlled for by local estimates. The study also shows that standard errors from individual case studies often overstate their precision, emphasizing the importance of pooling across case studies for reliable inference. The study uses data from the Quarterly Census of Employment and Wages (QCEW) and finds that minimum wage effects on earnings are positive, while employment effects are not significant. The study also finds that the dynamic response to minimum wage increases is positive in the short term, with cumulative effects ranging from 0.215 to 0.316. The study concludes that minimum wage increases have positive effects on earnings but no significant adverse effects on employment, and that traditional fixed-effects specifications are biased due to spatial heterogeneity. The study provides robust evidence that minimum wage increases do not lead to significant job losses and that local estimates are more reliable for estimating minimum wage effects.