تجزیه و تحلیل کمی اصلاح بیمه سلامت: جداسازی مقررات از توزیع مجدد

Two key components of the recent U.S. health reform are a new regulation of the individual health insurance market and an increase in income redistribution in the economy. Which component contributes more to the welfare outcome of the reform? We address this question by constructing a general equilibrium life-cycle model that incorporates both medical expenses and labor income risks. We replicate the key features of the current health insurance system in the U.S. and calibrate the model using the Medical Expenditures Panel Survey dataset. We find that the reform decreases the number of uninsured more than twice and generates substantial welfare gains. These welfare gains mostly come from the redistributive measures embedded in the reform, rather than from the regulatory changes.

The Patient Protection and Affordable Care Act, which culminated a long and vigorous health reform debate, was finally signed by the President of the U.S. in Spring 2010. This bill introduces a wide range of measures aiming primarily to increase health insurance coverage. In particular, the bill substantially changes the rules under which the individual insurance market operates and introduces penalties for those without insurance. At the same time it contains a set of measures that increase income redistribution in the economy. The goal of this paper is to provide a quantitative analysis of the upcoming reform in order to isolate the welfare effects of the new regulation of the individual market from the effects of the increased income redistribution. To do this, we construct a general equilibrium life-cycle model where agents face two types of risks: uninsurable labor income risk and persistent medical expenses risk that can be partially insured. People with high medical expenses have higher disutility from work and suffer a loss in productivity. We allow agents to be heterogeneous by educational level (exogenously fixed), which affects their ability to generate income and to access employer-based health insurance. We replicate the key features of the current health insurance system. First, in our model the insurance system consists of three components: individual market, employer-based market, and public insurance. Second, public insurance is available only to the lowest-income individuals, while people with high income are more likely to get employer-based coverage. Third, the majority of the uninsured can obtain insurance only from the individual market because they do not have access to the employer-based market and are not eligible for public insurance. At the same time this group of people tends to have low income. Fourth, public insurance is free and employer-based premiums are community rated. Those who purchase insurance in the individual market face risk-rated premiums that depend on their current medical shock. After calibrating the model to the key facts of the U.S. insurance system using the Medical Expenditures Panel Survey, we introduce the changes specified in the Patient Protection and Affordable Care Act (hereafter called the Bill). These changes can be broadly divided into two groups. First, there is a new regulation of the individual health insurance market that aims to create a risk-pooling mechanism outside the employer-sponsored market. In particular, insurance companies will be banned from conditioning premiums on individuals’ health status or history of claims. The price of an insurance policy can only vary by age. This restriction is known as age-adjusted community rating. To prevent creamskimming by insurers, another provision in the Bill is guaranteed issue which prevents insurance companies from denying coverage to individuals based on their health status. A possible outcome of a combination of community rating with guaranteed issue is an adverse selection spiral and to prevent this, the Bill requires all individuals without health insurance to pay a penalty unless the insurance premium constitutes too high a proportion of their income. Second, the Bill includes a set of redistributional measures. In particular, the Bill includes provisions to expand Medicaid. Currently, Medicaid covers several categories of population (for instance, adults with dependent children, pregnant women) with income below a threshold that varies significantly from state to state.1 After the reform all people under 65 years old with income below 133% of the Federal Poverty Line (FPL) will become eligible for Medicaid. Also low-income people will be able to get subsidies when buying insurance in the individual market. The goal of the subsidy is to keep premiums people pay for a standard insurance policy below a prespecified percentage of their income. When evaluating the welfare effects of the reform, as a welfare criteria we use the average utility of people who are alive at the beginning of the reform and live through the transition period. This welfare function favors redistribution across people with different income net of medical expenses. The reform introduces two additional channels of redistribution in the economy: first, from the healthy to the sick (through community rating in the individual market); second, from the high income to the low income (through subsidies and Medicaid expansion). Since neither of these new redistributional mechanisms is conditioned on income net of medical expenses, the resulting welfare effect of each mechanism is unclear: any redistribution from the healthy to the sick involves some redistribution from the healthy who are poor to the sick who are rich. Similarly every redistribution from the rich to the poor will involve some redistribution from the rich who are sick to the poor who are healthy. To adequately gauge the welfare effects of these redistributive channels we need to carefully capture the correlation between labor income and medical expenses. We do this by explicitly accounting for the fact that people with high medical expenses have lower productivity and lower labor supply. We find that the reform has a large effect on the fraction of the uninsured in the economy: this number decreases from 19.7% to 8.9%. The reform has the largest effect on young people in the lowest educational group, with the fraction of uninsured among high-school dropouts aged between 25 to 29 years old decreasing from 61.2% to 7.5%. Also the reform induces more participation in the individual market with the fraction of individually insured increasing from 7.3% to 18.5%. In terms of welfare, we find that the reform brings substantial gains equivalent to 0.64% of the annual consumption. However, these welfare gains mostly come from the redistributive measures embedded in the reform. If the reform is implemented without subsidies and Medicaid expansion, its welfare effects are significantly smaller. The intuition behind this result is as follows. Welfare gains from the reform are largely driven by the change in the welfare of low-income people. For the majority of this group, insurance premiums constitute a high fraction of income and they gain a lot from having subsidized coverage. On the other hand, the new regulation of the individual market by itself has a limited effect on health insurance affordability for low-income people who often prefer to stay uninsured if not subsidized.

The health reform bill recently signed by the President includes a wide range of measures which aim to increase the health insurance coverage in the U.S. The new law significantly changes the rules under which the individual insurance market operates. At the same time, it includes a set of redistributive measures that decrease the price of insurance for low-income people. This paper measures the welfare effects of the reform and decomposes them into two parts – one that is due to the new regulation of the individual market, and other due to the increased income redistribution in the economy. We construct a general equilibrium heterogeneous model with a rich representation of the current U.S. health insurance system. We calibrate the model using Medical Expenses Panel Survey to match the key insurance statistics of the U.S. economy. We find that the reform brings significant welfare gains, however these gains are mostly achieved by the redistributive part of the reform – Medicaid expansion and premium subsidies. If the reform only changes the regulation of the individual market and introduces penalties for the uninsured, the welfare gains almost disappear. Most of the currently uninsured have low income and they gain a lot from having subsidized health insurance. Reorganizing the individual insurance market alone has a limited effect on these people because non-subsidized insurance premiums, whether community rated or not, constitute such a significant portion of their income that they often prefer to stay uninsured if not subsidized.