Effects of Implementing State Insurance Market Reform, 2011-2012. Methods


The assessment of premium rate increases follows on a prior study where NORC analyzed the same outcomes for the years 2008 to 2011. The findings from this prior study were presented in the report entitled “Trends in Premiums in the Small Group and Individual Insurance Markets, 2008-2011,” or the Trends study. The Trends study addressed similar research questions; however it involved a different sample of states and different methods for collecting information on rate increases. While this report includes data from both studies, the Trends study included an incomplete picture of 2011, because data collection occurred during the summer and fall of 2011 before all 2011 rate filings were approved and publicly available – all Trends estimates for 2011 were given a cautionary footnote as a reminder.

This study builds upon the research and methods of the Trends study. For this study, NORC extracted data from the 24 states included in the Trends study that have public websites, through which state insurance departments make insurance rate filings available, and 5 additional states that have publicly available filings. The added states represent a mixture of population size, regulatory disposition, and geography. In total, 1,654 filings were collected for the study, 690 of which took effect in 2011 and 964 in 2012.

To increase replicability and transparency of data collection, the project team built a relational database to track queries submitted to state insurance department portals, filings downloaded from state portals, and data captured from filings in the scope of the study. NORC saved electronic copies of all these valid filings. Quality control review re-examined statistical outliers (premium increases of more than 20 percent and less than -10 percent) and observations for which the approved rate was higher than the proposed rate. NORC also conducted an audit of 100 random observations and found an error rate of approximately one percent.

To account for differences in the impact of each observation on the population insured in the individual/conversion and small group markets, weights were derived for each carrier, product and state. Carrier weights were created separately for the individual/conversion markets and small group markets using the National Association of Insurance Commissioners (NAIC) April 2011 reporting of member months in both of these markets. NORC based the weights assigned to each product on the enrollment data in the filings. In the small group market, NORC used data from the Medical Expenditure Panel Survey – Insurance Component (MEPS-IC) on the prevalence of various insurance products in that market to adjust filing weights. State weights reflect the share of national enrollment for persons enrolled in the small group and individual markets based on data from NAIC.

In calculating standard errors a finite multiplier (the percentage of enrollment in the state in the sample) was used, which greatly reduces the size of standard errors. This corrective factor helps to account for the relatively large proportion of the total affected population that is included in the sample and the collected data. Using t-tests at the p=.05 significance level, NORC tested for differences between years and covariates such as product type, state regulatory review, and market concentration.

View full report


"20130607InsMktReformReportFnl.pdf" (pdf, 1.46Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®