نقش تست دارایی در اصلاحات بیمه سلامت عمومی

Starting slowly with the 1996 Welfare Reform Act and culminating in the 2010 Affordable Care Act, means-tested public health insurance eligibility expanded to include adults in low-income families regardless of their asset holdings. This paper quantifies the effects of these eligibility expansions within the context of the 2010 Affordable Care Act. I construct a dynamic stochastic general equilibrium model with indivisible labor supply expanded to include an endogenous household choice of health insurance coverage and calibrate it to U.S. data. I establish that changes in the distribution of labor and welfare associated with removal of asset testing are driven by exit of high productivity and high wealth households from the labor market. I then expand my analysis to the 2010 Affordable Care Act to demonstrate that removal of asset testing is critical to the obtained results even when combined with other provisions of the Act. Finally, I find that a simple asset test for eligibility of health insurance transfers undoes the distortion to the household labor supply decision among high productivity types. These results are robust to the introduction of employer premium contributions, an independent health insurance market, and idiosyncratic shocks to eligibility for employment-based health insurance.

The dramatic expansion of Medicaid program in the 2010 Affordable Care Act (ACA) is the culmination of an expansion process that started with the 1996 Welfare Reform Act that expanded state choice to eliminate the asset test for parents in low-income families.1 By January 1997, four states had eliminated the Medicaid asset test for adults in families. Twenty-four states had eliminated the Medicaid asset test for adults in families by January 2012. Effective January 1, 2014, Medicaid eligibility expanded to include all low-income adults regardless of their asset holdings through the 2010 ACA with individual states retaining the right to opt-out. In addition, the ACA legislation expands federal financing of public health insurance programs, requires state implementation of health insurance “exchange” markets with means-tested transfers such as premium payments and limits on household out-of-pocket expenses, and provides health insurance coverage requirements for individuals and firms. The purpose of this paper is to evaluate the consequences of public health insurance expansion, and the role of the asset test in particular, within the context of the ACA legislation. I answer the following questions: To what extent does removal of asset testing of Medicaid benefits alone alter household decisions to participate in the labor force and purchase employment-based health insurance? Do the introduction of the insurance exchange and the health insurance mandate of the 2010 ACA alter the results? Ultimately, as in any analysis of public policy reform, a key question is the effect on household welfare. To that end, I conduct a welfare analysis to find what households win and lose in the reform and the role of the asset test in generating our results. To answer these questions, I conduct a quantitative analysis using a dynamic stochastic general equilibrium model with heterogeneous households and incomplete markets in the spirit of Aiyagari (1994) and Huggett (1993). Each period, agents make a choice of labor force participation and employment-based health insurance coverage. This decision is made given the individual agent ability level, wealth, expected medical expense shocks, and public health insurance eligibility. The agent can also self-insure by saving income across periods. The government provides a means-tested consumption floor (Hubbard et al., 1995) and administers a health insurance program that represents Medicaid and Medicare. Medicaid eligibility is determined by income and asset thresholds. The model is calibrated to match the distribution of medical expenditures and health insurance participation rates at the family level that are estimated using data from the Current Population Survey (CPS) and the Medical Panel Expenditure Survey (MEPS). My key results can be summarized as follows. First, removal of the Medicaid asset test plays a key role in altering household work incentives. Elimination of the asset test increases the reservation wage of workers approaching retirement by providing access to government health insurance outside the workplace. Households accumulate more assets on average and exit the labor force earlier than in the baseline model. I find a decline in average employment of agents over 50 years old and a 4.1 percent decline in aggregate productivity. Aggregate household participation in the labor market also decreases by 4.1 percentage points. However, the introduction of the insurance exchange undoes some of these labor disincentives by introducing transfers that are subject to a minimum income threshold. This minimum income threshold reduces the incentive for early retirement with the exception of wealth-rich households that meet the income threshold through asset income alone. The introduction of the complete reform is characterized by a 3.6 percent decline in aggregate labor productivity and a modest decline in labor force participation of 1.8 percentage points. Second, removal of the Medicaid asset test is associated with an ex-ante welfare loss of −0.6 percent in consumption equivalents, while the complete reform policy undoes the welfare loses from Medicaid expansion and yields a modest welfare loss of −0.2 percent. The magnitude of ex-ante welfare gains depend on the presence of trading frictions that limit access to employment-based health insurance with empirically reasonable assumptions yielding welfare gains from the complete reform of up to 1.1 percent. The distribution of welfare gains is also sensitive to the inclusion of an asset test for government health insurance transfers. In particular, the addition of an asset test for Medicaid or insurance exchange eligibility has a dramatic effect on the distribution of welfare by targeting transfer payments to wealth-poor agents and those with low labor productivity. To illustrate the findings most clearly, the baseline model abstracts from several features of the health insurance system. The purpose of these abstractions is to illustrate, in a minimalistic fashion, the potential role of asset test of government health insurance benefits. In particular, the model abstracts from employer contributions to employee payment of health insurance premiums, an independent or direct-purchase health insurance market, as well as frictions that limit the ability of agents to purchase health insurance through their employer. While the model fails to capture some of the more “realistic” features of the U.S. health insurance system, I conduct a sensitivity analysis to show that changes in allocations and prices due to the reform are not driven by omitting these features. Nevertheless, exogenous restrictions to the availability of employment-based insurance, to the extent that they exist, have implications for the welfare effects of reform. There is a growing literature that explicitly addresses the effects of various aspects of the 2010 health care reform. Brügemann and Manovskii (2010), Feng (2009), Hansen et al. (2012), Huang and Huffman (2010), Jung and Tran (2011), Pashchenko and Porapakkarm (2010), and Tsujiyama (2013) are recent examples of papers that conduct analysis of reform within a dynamic stochastic general equilibrium framework. The key contribution of this paper to the literature is the focus on removal of the Medicaid asset test that is omitted from the previous studies. In particular, I show that removal of the asset test as part of the ACA reform results in a strikingly different labor supply response than that found in Pashchenko and Porapakkarm (2010) among others. Critically, these differences in results stem largely from the decision to model a Medicaid asset test in the baseline economy. The paper proceeds as follows. Section 2 presents the model. The details of the calibration procedure and an evaluation of the benchmark model are found in Section 3. Section 4 details the results and Section 5 presents a sensitivity analysis. Section 6 concludes.

In this paper, I develop a dynamic stochastic general equilibrium model with idiosyncratic productivity and medical expenditure risk to quantify the effects of removal of the Medicaid asset test within the broader 2010 ACA legislation on the allocations of insurance coverage, employment, and welfare. The Medicaid asset test plays a key role in altering household work incentives. Elimination of the asset test increases the reservation wage of workers approaching retirement by providing access to government health insurance outside the workplace. Households accumulate more assets on average and exit the labor force earlier than in the baseline model. Aggregate effect of the change in work incentives is a decline in average employment of agents over 50 years old and a 4.1 percent decline in aggregate productivity. Aggregate household participation in the labor market also decreases by 4.1 percentage points. However, the introduction of the insurance exchange undoes some of these labor disincentives by introducing transfers that are subject to a minimum income threshold. This minimum income threshold reduces the incentive for early retirement with the exception of wealth-rich households that meet the income threshold through asset income alone. The introduction of the complete reform is characterized by a 3.6 percent decline in aggregate labor productivity and a modest decline in labor force participation of 1.8 percentage points. Removal of the Medicaid asset test is associated with an ex-ante welfare loss of −0.6 percent in consumption equivalents, while the complete reform policy undoes the welfare loses from Medicaid expansion and yields a modest welfare loss of −0.2 percent. The magnitude of ex-ante welfare gains depend on the presence of trading frictions that limit access to employment-based health insurance with empirically reasonable assumptions yielding welfare gains from the complete reform of up to 1.1 percent. The distribution of welfare gains is also sensitive to the inclusion of an asset test for government health insurance transfers. In particular, the addition of an asset test for Medicaid or insurance exchange eligibility has a dramatic effect on the distribution of welfare by targeting transfer payments to wealth-poor agents and those with low labor productivity.