When Do States Erode Protections? The Case of Contemporary Child Labor in the United States

• In Event: Policymaking Across the American States

Abstract

Conventional wisdom holds that child labor in the United States is something from a bygone era. However, in the past decade child labor violations have risen by 400%, just as numerous U.S. states have moved to weaken their child labor laws. We argue that these erosions of child labor protections represent a pattern of policymaking driven by both group interests in national party coalitions and economic conditions. We create an original state-by-year dataset of 15 distinct types of child labor policies from 2008 to 2024 to test this argument in a modern differences-in-differences (DiD) framework.

We first describe the landscape of U.S. child labor policies across states. There is a high degree of fragmentation in child labor policy, leading to a patchwork of conflicting and highly variable policies. Federal child labor regulations leave open critical loopholes, especially in agriculture where, for example, 12-year-olds can legally work overnight on school nights, and 16-year-olds can work in hazardous occupations (an allowance that contradicts American international commitments). Some—but not all—states enact their own regulations to supplement federal regulations, and state child labor policies are far from settled. Roughly a third of states have changed their child labor policies since 2008, with about half of these weakening their protections.

We use a new fixed effects counterfactual (FEct) estimator for DiD designs to examine the drivers of eroding child labor protections (Liu, Wang, and Zu 2024). We find that shifting to Republican control of the governorship and both legislative houses leads to weaker child labor policies, on average, and this is driven entirely by unified Republican control in times of excess demand for labor. Specifically, we find that experiencing both Republican trifecta control and low unemployment leads states to decrease their child labor policy strength by 8.7% of a standard deviation, on average. Effects are concentrated in industries outside of agriculture and entertainment. In contrast, but consistent with the broader implications of our theory, unified Democratic control under high labor demand does not cause state policymakers to weaken their child labor laws but rather to increase state minimum wages. We find no evidence in support of competing hypotheses related to labor union strength and immigration. Results are consistent in a series of recommended validity checks, sensitivity analyses, and falsification checks with related measures in place of Republican trifecta control.

This study contributes to literatures on the group interests of national party coalitions, labor and employment policy, and human rights. The erosions of child labor protections that we document provoke cause for concern as they occur within a fragmented intergovernmental system that fails to clearly address hazardous and exploitative working conditions for minors.

Author

• Melissa Arnold Lyon, University at Albany
  Presenting Author