The heterogeneity of the seven programs considered in this study explains the wide range of estimates for administrative costs and savings. Findings indicate that implementation costs are highest when states aggressively pursue automated strategies, but if these strategies reach large numbers of applicants and reduce per capita eligibility determination costs, states can expect to recoup their investment, in some cases quite quickly. Alternatively, states that primarily pursue ELE as an outreach strategy, to engage historically difficult-to-enroll populations, may incur upfront and ongoing costs that are difficult to recapture, particularly if administrative processing efficiencies do not result or are absorbed by contractors. However, even in these states, program administrators uniformly viewed ELE as a worthwhile strategy because it facilitated the goal of coverage expansion.
For example, although Louisiana’s ELE process was the most expensive to implement, it is delivering notable savings in terms of staff time. More than 90 percent of these savings are attributable to renewals; Louisiana’s automated ELE process negates the need for staff to spend any time on approximately 15,000 Medicaid renewals each month. These represent 30 percent of all children’s Medicaid renewals. Louisiana also avoids most new ongoing costs, because its strategy does not involve mailings to potentially eligible applicants who may not enroll. Given the magnitude of these savings, Louisiana offset its initial implementation costs and even realized net administrative savings within the first year of ELE program operations, although its up-front costs were the highest reported among states included in the first year of this evaluation.
Apart from Louisiana, Alabama has processed far more children than any other state, resulting in the second-largest net administrative savings from ELE. These savings accrue even though the current ELE process—with eligibility workers manually checking SNAP and TANF databases— saves staff fewer than five minutes per ELE application or renewal. Again, savings are primarily attributable to renewals—approximately 70 percent of Alabama’s ELE cases in a given year are renewals rather than new applications.
Given the results in Louisiana and Alabama, using ELE for renewals has clear potential; however, other states did not uniformly perceive ELE renewals as an advantage over existing renewal facilitation techniques. For example, New Jersey and Maryland already have processes for administrative and ex parte renewals, respectively, allowing them to automatically renew coverage or send pre-populated forms to parents to sign and return, and they did not believe ELE renewals would produce additional efficiencies.
In Oregon and Iowa Medicaid—states whose ELE programs reach relatively few families—ELE processes are essentially cost neutral from an administrative perspective. Substantial up-front investments made by Iowa Medicaid may or may not be recovered over the long term, depending on the trend in ELE-facilitated enrollments over the coming years.
Maryland and New Jersey, which have partnered with their state tax agencies, spend more on doing tax-based ELE mailings than they save in administrative costs. In New Jersey, only about 1 percent of these mailings result directly in enrollments, and savings from processing an application accrue to a contractor. Maryland does not track ELE enrollments so we cannot estimate a conversion rate. However, even if Maryland’s conversion rate were significantly higher than New Jersey’s, the fact that its ELE process saves eligibility staff no time compared to the traditional application process means that, as it currently operates, it has no potential to reduce administrative costs.
Although the estimates in this chapter reflect the best available information, we recognize that, in each state, some of the costs of ELE have been omitted due to challenges in recall or missing data. Accordingly, we interpret the final calculations as generally indicative of the direction and magnitude of the likely savings or costs that states considering ELE might expect to accrue. In the second year of the evaluation, we will revisit these estimates to assess whether states’ early cost experiences with ELE have persisted, as well as the impact of anticipated changes in ELE programs, such as Alabama’s plans for enhanced automation and Maryland’s improved application targeting.
Finally, we note that, in this study, administrative costs were narrowly defined to focus on expenses associated with eligibility processing. Our estimates, therefore, do not include the costs and savings of covering children in public health insurance programs who otherwise would not be insured. Capitated premiums or payments for services used by children who are newly enrolled in public insurance could easily outweigh any administrative savings from their expedited enrollment or renewal; in 2006, nationwide average Medicaid costs for non-disabled children were $249 per member per month (Lipson et al. 2010).