Fraud in the Workplace? Evidence from a Dependent Verification Program

In recent years many employers, both in the private and public sectors, have implemented dependent verification (DV) programs, which aim to reduce employee benefits costs by ensuring that ineligible persons are not enrolled in their health plan as dependents. However, little is known about their efficacy. In this paper, we evaluate a DV program using a panel of health plan enrollment data from a large, single-site employer who implemented it several years ago. We find that relative to all other years, dependents were 2.7 percentage points less likely to be reenrolled in the year that DV was introduced, indicating that this fraction of dependents was ineligibly enrolled prior to the program’s introduction. These disenrollment effects were especially large for same-sex partners and older children. We show that the program did not induce employees to leave the employer’s plan and (say) put themselves and their dependents on the spouse’s plan. We also show that disenrollment occurred because dependents were actually ineligible, not because of compliance costs that might be associated with providing documentation. The DV program saved about $46 per enrolled employee. A considerable fraction of these cost savings came from removing older children who didn’t meet additional criteria. Therefore, the dependent coverage provision of the Affordable Care Act of 2010, which essentially renders all children up to age 26 eligible in all employer health plans, will substantially limit the future cost saving potential of such programs. Hence, as the state governments and private employers that have implemented DV programs adapt to the new regulatory environment, the popularity of dependent verification programs may well diminish.