Screening in Contract Design: Evidence from the ACA Health Insurance Exchanges

Abstract

We study insurers' use of prescription drug formularies to screen consumers in the ACA Health Insurance exchanges. We begin by showing that exchange risk adjustment and reinsurance succeed in neutralizing selection incentives for most, but not all, consumer types. A minority of consumers, identifiable by demand for particular classes of prescription drugs, are predictably unprofitable. We then show that contract features relating to these drugs are distorted in a manner consistent with multidimensional screening. The empirical findings support a long theoretical literature examining how insurance contracts offered in equilibrium can fail to optimally trade off risk protection and moral hazard.

FIGURE 1.. Formulary Data: Tiering in Empoyer and Exchange Plans

Notes: Histograms indicate the fraction of drugs contained in restrictive tiers in employer and exchange plans. Observations are plans. In panel A, restrictive tiers are defined as the specialty tier or higher. See Table 1 for a complete ranked listing of the tiers. Panel B repeats the histogram for the fraction of drugs requiring prior authorization or step therapy (PA/ST) or explicitly listed in the formulary as not covered.

FIGURE 2.. Actionable Selection Incentives Remain of Risk Adjustment

Notes: This figure shows the relationship between health care spending and simulated revenue for each therapeutic class of drugs for which spending and revenue are less than $70,000. Means are for total spending, revenue, or profit, calculated over the set of consumers who generate at least one drug claim in the class. Simulated revenue is calculated according to the HHS risk adjustment and reinsurance algorithms as described in the text. Each circle plots the spending and revenue means for a therapeutic class with marker sizes proportional to the number of consumers generating claims in the class. In panel A, the line at 45 degrees indicates the break-even point. In panel B, a horizontal solid line at zero indicates breakeven, and a dashed line plots the linear regression coefficient, weighted by the number of consumers represented in each class.

FIGURE 3.. Changing Incentives Due to Removal of Reinsurance

Notes: This figure plots how the relationship between health care spending and simulated revenue changes with the removal of reinsurance. Vertical lines describe the loss in net revenue associated with each class that is implied by the removal of reinsurance. The bottom point of each line plots revenue without reinsurance (dropping both the reinsurance premiums and payouts). The top point of each line plots revenue with reinsurance included. For a small set of therapeutic classes, each containing a large number of lower utilization enrollees, removing reinsurance generates a net gain in revenue. These are indicated with triangle markers. Simulated revenue is calculated according to the HHS risk adjustment and reinsurance algorithms as described in the text. The line at 45 degrees indicates the break-even point. See Figure 2 for additional notes.

FIGURE 4.. Determinants of Exrolee Costs by Selection-Incentive Stregth

Notes: This figure decomposes total enrollee costs into inpatient, outpatient, and drug costs. Drug costs are divided according to whether the drug is inside or outside of the defining therapeutic class. Each of the 220 therapeutic classes are ranked according to the strength of the selection incentive and then binned into 20 ventiles of the incentive measure. Classes are associated with increasingly unprofitable patients moving from left to right.

FIGURE 5.. Residual Plots of Restictive Tiering versus Selection Incentives

Notes: This figure plots residuals from a regression of formulary restrictiveness on drug-class fixed effects and plan fixed effects: $Y_{cj}={\gamma }_c+{\alpha }_j+{ϵ}_{cj}.$ To generate the plots, therapeutic classes c are grouped into ventiles according to the strength of the selection-incentive measure S_c. We then take the means of the residuals within each ventile, separately for employer and exchange plans. The vertical axes plot these means. The horizontal axes in the left column correspond to the mean of the selection incentive, normalized as a z-score. The horizontal axes in the right column correspond to the ventile number with ventile 20 including the ninety-fifth to one-hundredth percentiles of classes by incentives to avoid. In each panel, an OLS regression line is plotted separately for exchange and employer plans.