Has the Affordable Care Act increased part-time employment?

Abstract

We examine the impact of the Affordable Care Act (ACA) on part-time employment. Because the ACA’s employer health insurance mandate applies to individuals who work 30 or more hours per week, employers may try to avoid the mandate by cutting workers’ hours below the 30-hour threshold in order to avoid having to provide them with health insurance. Although the employer mandate only went into effect in 2015, many observers have argued that forward-looking employers began to shift towards a part-time workforce well in advance of the mandate. To test this hypothesis, we examine relative shifts across two categories of part-time workers (25–29 hours and 31–35 hours). We find some evidence of a shift from the 31–35-hour category into the 25–29-hour category after the passage of ACA in March 2010. However, that shift is not more pronounced among low-wage workers or among workers in industries and occupations most likely to be affected by the mandate. Thus, there is little evidence that the ACA has caused the shift across hours categories, or led to an increase in part-time employment. However, the ACA could cause a shift towards part-time work in the future as the mandate takes effect.

Keywords:

• Health insurance
• Affordable Care Act
• employment
• part-time employment

JEL Classification:

• I18
• I11
• J32

Acknowledgements

We thank Regan Kuchan and Brittany Pineros for excellent research assistance. The views contained in this article reflect those of the authors only and not necessarily of their affiliated institutions.

Notes

¹ The mandate was initially scheduled to go into effect in 2014, but was postponed until 2015. It went into effect in 2015 for employers with 100 or more full-time equivalent employees during the previous year, and it will be extended in 2016 to employers with 50 or more full-time equivalent employees in the previous year (KFF, 2014). For employers who are affected in 2015, lookback periods for determining full-time status may begin as early as 2014.

² For example, Caldeira (2013) and Bauman (2014b).

³ We also exclude September 2009 due to the fact that Labor Day occurred during the reference week for the survey, resulting in an unusually high fraction of workers working 32 hours (a 4-day week).

⁴ Unweighted regressions yield similar results.