To understand the effects of insurance regulation on the labor market and insurance coverage, this' study uses a difference-in-difference-in-differences analysis to compare five states that passed minimum maternity length-of-stay laws with states that waited until after a federal law was passed. On average, we do not find statistically significant effects on labor market outcomes such as hours of work and wages. However, we find that employees of small firms in states with maternity length-of-stay mandates experienced a 6.2-percentage-point decline in the likelihood of having employer-sponsored insurance. Implementation of federal health reform that requires minimum benefit standards should consider the implications for firms of differing sizes.
Mandates that require insurers to cover specific health care benefits have been a popular legislative tool for state regulation of health insurance over recent decades, largely in response to concerns about adverse selection or stinting, particularly under certain managed care arrangements (Laugesen et al. 2006). Under the Patient Protection and Affordable Care Act (ACA) of 2010, most individuals will be required to carry health insurance that meets certain minimum benefit standards, and the majority of Americans will likely continue to receive insurance through their employers. Understanding how state-mandated minimum health benefits have affected the labor market and employer-sponsored health insurance (ESI) coverage in the past--particularly among small firms, which are less likely to offer ESI and more sensitive to changes in the cost of insurance--can shed light on issues that could arise as health reform is implemented with required standardized benefits. Analysis of past adverse impacts of regulatory strategies on small firms offers important lessons. New regulation in one area of health insurance--such as the standardization of benefits--also requires a supportive set of regulatory strategies to compensate or address risk selection. Those compensating mechanisms were largely absent in the 1990s, but are a key feature of the new legislation. (1)
Mandated benefits have been an attractive tool because they do not require collecting new taxes or shifting government revenues to provide a benefit that policymakers believe should be required--possibly for reasons of paternalism, because of externalities, or due to problems of adverse selection (Summers 1989). There may also be an efficiency argument for using mandates within a largely employer-sponsored insurance system as opposed to other methods of providing a benefit: according to standard economic theory, when an employee fully values a mandated benefit wages will shift by the full cost of the benefit without any distortionary effects on labor supply.
This efficiency argument in favor of mandates breaks down, however, when wages cannot adjust by the full value of the benefit to the employee. A particular case may arise when the mandated benefit is of value to a specific identifiable group, such as women, but there are barriers to relative wage adjustment. There is a potential efficiency loss from a "group-specific mandate"; if wages are unable to freely adjust there may be considerable deadweight loss from the mandate (Gruber 1994a). In addition to wages, mandates could affect level of employment, including hours worked. The simple model discussed previously does not predict an effect on hours, since health insurance is a fixed cost, and the theoretical effects are somewhat more ambiguous. If employers seek to replace full-time workers with part-time workers who are not covered by ESI, average hours would fall. On the other hand, it has been suggested that hours may be expected to increase in response to a mandate for at least two reasons. First, there may be an income effect as workers adjust the balance of benefits versus other consumption goods. Second, employers may seek to offset the increased cost of the mandate by inducing employees to work more hours or by adjusting the mix of labor input, reducing the number of employees while increasing the number of hours per worker (Gruber 1994a; Kaestner and Simon 2002). …