Put aside the political and religious fights for a moment. There’s a simple fact about contraception that gets lost in much of the coverage: it saves money. Lots of money….
Contraception costs $100-$600 annually and cuts the risk of unplanned pregnancies to nearly nothing. Liebman finds various estimates of the savings: One paper put it at $6,000-$10,000 per person for every two years on contraception. Another put it at $13,000 over five years of contraceptive coverage. A study of Medicaid found that contraceptive services saved $4.26 for every dollar invested.
So contraception coverage is a massive net winner for insurance companies and policies should default to contraceptive coverage with no barriers to usage.
Right?
Austin Frakt at the Incidental Economist:
A fuller review of the literature on the cost and cost offsets of contraceptive coverage by Daniel Liebman, a colleague, finds that the evidence is thin that, from an insurer’s perspective, contraceptive coverage pays for itself in the long term. Moreover, it almost certainly does not in the short. The cost of contraceptive coverage is immediate, and the possible offsets (reduced pregnancies) are downstream, often years in the future….
In a 1995 study in the American Journal of Public Health and in a 2000 Milliman Studyassessing Texas’ contraceptive mandate, the authors argue that the eventual savings of contraceptive coverage may not necessarily accrue to an insurer.
Both sets of statement are true from a certain point of view. These posts are answering two slightly different questions. Ezra Klein is summarizing whether or not contraception availability reduces total system wide medical costs. Austin Frakt’s point is that system costs may be reduced but the health insurance company that is paying for the contraception for people who receive employer sponsored coverage is not gaining the cost savings.
The argument that contraception reduces total system costs is fairly straightforward. Pregnancies in general are expensive. Unplanned pregnancies tend to be more likely to occur with individuals who are in a more stressed life situation so the probability of stressed and complicated pregnancies (controlling for age etc) is higher. Reducing the number of unplanned pregnancies reduces costs even if the unplanned pregnancy in Year 1 is “replaced” with a planned pregnancy in Year 3. Voila, significant costs savings.
Employer sponsored health care has a risk pool that is significantly different than the general population, and very different than the pre-Expansion Medicaid population. Individuals who have employer sponsored health insurance tend to be more likely to be employed, more likely to not be living in poverty, more likely to be reasonably healthy, more likely to have a high school or greater educational attainment and more likely to be out of their teens than the general population. All of those “more likely” qualifiers reduce the probability of both unplanned pregnancies and reduce the probablity of complicated pregnancies. Furthermore, individuals with employer sponsored health insurance have significantly more resources than average to substitute pharmaceutrical/prescription birth control for other methods (condoms for instance). From an insurance company point of view, contraception coverage is “belts and suspenders” coverage for most women with employer sponsored health insurance.
That view can change if we change the population being considered, or if we broaden the scope of the cost-benefit assignment system. A broader scope that includes social service costs or a targeted population of individuals with low substitutability of prescription and non-prescription birth control usage will show massive costs savings to society as a whole but a narrow scope that is restricted to only immediate claims impact will show a wash.
In situations like that, the correct public policy move is to either have a federal mandate of universal coverage via private insurance so that no company has a competitive advantage to deny OR a direct federal subsidy so that system costs can be reduced even if the payers don’t see a positive return on investment. And then the second phase of good public policy for either option is to sacrifice a goat to placate the asshole five on the Supreme Court and hope that works.