Poster Paper:
Measuring the dollar value of employer-provided health insurance contributions: a synthetic approach
*Names in bold indicate Presenter
Study Design/Method/Data: The current CPS ASEC estimate of ECHIP is modeled using the 1977 National Medical Care Expenditure Survey. A few years ago, researchers at the Census Bureau developed a Bayesian model that combined micro-data from 2010 MEPS-IC and 2011 CPS ASEC that offered more accurate estimates than did the old ECHIP model. However, this new approach for estimating ECHIP did not replace the old method due to concerns about the availability of MEPS-IC microdata before the release of the CPS ASEC public-use file.
The goal of this paper is to update the proposed approach and develop a Bayesian model that uses MEPS-IC aggregated data and CPS ASEC micro-data to produce more accurate ECHIP information than is on the current public-use file. We will assess our estimates by examining the ratio of ECHIP to total health insurance costs (the sum of ECHIP and the person’s out-of-pocket premiums), and by describing ECHIP costs by socioeconomic characteristics and by policyholder status.
Results: Preliminary results suggest that the MEPS-IC-based synthetic measure offered more accurate ECHIP values than those in the current public-use CPS ASEC data, regardless of whether individual or aggregate inputs were used. Both inputs offered similar predicted mean and median ECHIP amounts.
Conclusion: Using an aggregate model for producing a CPS ASEC estimate of ECHIP produces improved, realistic estimates employers’ contributions. An aggregate model is preferred over a microdata model because it is easier to do in real time production environment.
