in response to premium increases in Florida’s SCHIP program, and found a price elasticity for
the disenrollment hazard rate of 2.2; that is, a 10 percent increase in the monthly premium would
produce a 22 percent increase in the probability of disenrollment. The authors concluded that, at
least in the short-term, SCHIP families are very price-sensitive and premium increases may
differentially affect children in families with the lowest income and residents in rural areas.
Marton (2005) examined the impact of introducing a premium for families between 151
percent and 200 percent FPL on the duration of premium-paying SCHIP enrollment spells in
Kentucky, drawing a natural control group from families not subject to the premium (those with
income between 101 percent and 150 percent FPL). He found strong price responsiveness in the
short run—the probability of disenrollment increased from 5 percent to 21 percent in the first
two months after the premium was introduced—but weak responsiveness in the long run and no
significant disenrollment among the control group. He concluded that the introduction of a
premium for families with modest incomes reduced the length of enrollment.
Crowd-Out of Private Coverage
Crowd-out is a term that refers to the propensity of people to substitute public for (higher-
cost) private coverage when both are available to them. We include crowd-out in the discussion
of price elasticity of health insurance because it reflects a change in demand for private coverage
as a result of its substitute (public program) becoming available at zero price; in this sense,
crowd-out is a measure of the cross-price elasticity between public and private coverage.
The literature on crowd-out is extensive, but there is no consensus about whether crowd-out
is significant (Davidson et al. 2004). Some studies have linked a reduction in private coverage to
a gain in public enrollment following an expansion of eligibility for the public program.
However, there are many reasons that people might lose private coverage other than crowd out: