Wednesday, February 5, 2014

What the new CBO Really says about Obamacare

Here is the full portion of the Congressional Budget Office's recent 182-page study as it pertains to Obamacare, part-time jobs and the labor force projections. In short, the CBO predicts that over the next 10 years, because of Obamacare, there will be 2.5 million less full-time jobs by 2024 (less than there otherwise might have been) because:
  1. employers will hire more part-time workers to avoid paying for healthcare;
  2. and because some workers will prefer to work part-time to get the healthcare subsidies (as if workers had all that much to say about how many hours they can work).

The actual text from the CBO report is as follows, pertaining just to the part about the labor force projections. You be the judge.

The ACA includes a range of provisions that will take full effect over the next several years and that will influence the supply of and demand for labor through various channels. For example, some provisions will raise effective tax rates on earnings from labor and thus will reduce the amount of labor that some workers choose to supply. In particular, the health insurance subsidies that the act provides to some people will be phased out as their income rises—creating an implicit tax on additional earnings—whereas for other people, the act imposes higher taxes on labor income directly. The ACA also will exert conflicting pressures on the quantity of labor that employers demand, primarily during the next few years.

The ACA’s largest impact on labor markets will probably occur after 2016, once its major provisions have taken full effect and overall economic output nears its maximum sustainable level. CBO estimates that the ACA will reduce the total number of hours worked, on net, by about 1.5 percent to 2.0 percent during the period from 2017 to 2024, almost entirely because workers will choose to supply less labor—given the new taxes and other incentives they will face and the financial benefits some will receive. Because the largest declines in labor supply will probably occur among lower-wage workers, the reduction in aggregate compensation (wages, salaries, and fringe benefits) and the impact on the overall economy will be proportionally smaller than the reduction in hours worked. Specifically, CBO estimates that the ACA will cause a reduction of roughly 1 percent in aggregate labor compensation over the 2017–2024 period, compared with what it would have been otherwise. Although such effects are likely to continue after 2024 (the end of the current 10-year budget window), CBO has not estimated their magnitude or duration over a longer period.

The reduction in CBO’s projections of hours worked represents a decline in the number of full-time-equivalent workers of about 2.0 million in 2017, rising to about 2.5 million in 2024. Although CBO projects that total employment (and compensation) will increase over the coming decade, that increase will be smaller than it would have been in the absence of the ACA. The decline in full-time equivalent employment stemming from the ACA will consist of some people not being employed at all and other people working fewer hours; however, CBO has not tried to quantify those two components of the overall effect. The estimated reduction stems almost entirely from a net decline in the amount of labor that workers choose to supply, rather than from a net drop in businesses’ demand for labor, so it will appear almost entirely as a reduction in labor force participation and in hours worked relative to what would have occurred otherwise rather than as an increase in unemployment (that is, more workers seeking but not finding jobs) or underemployment (such as part-time workers who would prefer to work more hours per week).

CBO’s estimate that the ACA will reduce employment reflects some of the inherent trade-offs involved in designing such legislation. Subsidies that help lower-income people purchase an expensive product like health insurance must be relatively large to encourage a significant proportion of eligible people to enroll. If those subsidies are phased out with rising income in order to limit their total costs, the phaseout effectively raises people’s marginal tax rates (the tax rates applying to their last dollar of income), thus discouraging work. In addition, if the subsidies are financed at least in part by higher taxes, those taxes will further discourage work or create other economic distortions, depending on how the taxes are designed. Alternatively, if subsidies are not phased out or eliminated with rising income, then the increase in taxes required to finance the subsidies would be much larger.

CBO’s estimate of the ACA’s impact on labor markets is subject to substantial uncertainty, which arises in part because many of the ACA’s provisions have never been implemented on such a broad scale and in part because available estimates of many key responses vary considerably. CBO seeks to provide estimates that lie in the middle of the distribution of potential outcomes, but the actual effects could differ notably from those estimates. For example, if fewer people obtain subsidized insurance coverage through exchanges than CBO expects, then the effects of the ACA on employment would be smaller than CBO
estimates in this report. Alternatively, if more people obtain subsidized coverage through exchanges, then the impact on the labor market would be larger.

CBO estimates that the ACA will cause smaller declines in employment over the 2014–2016 period than in later years, for three reasons. First, fewer people will receive subsidies through health insurance exchanges in that period, so fewer people will face the implicit tax that results when higher earnings reduce those subsidies. Second, CBO expects the unemployment rate to remain higher than normal over the next few years, so more people will be applying for each available job—meaning that if some people seek to work less, other applicants will be readily available to fill those positions and the overall effect on employment will be muted. Third, the ACA’s subsidies for health insurance will both stimulate demand for health care services and allow low-income households to redirect some of the funds that they would have spent on that care toward the purchase of other goods and services—thereby increasing overall demand. That increase in overall demand while the economy remains somewhat weak will induce some employers to hire more workers or to increase the hours of current employees during that period.

In 2010, CBO estimated that the ACA, on net, would reduce the amount of labor used in the economy by roughly half a percent—primarily by reducing the amount of labor that workers choose to supply. That measure of labor use was calculated in dollar terms, representing the approximate change in aggregate labor compensation that would result. Hence, that estimate can be compared with the roughly 1 percent reduction in aggregate compensation that CBO now estimates to result from the act. There are several reasons for that difference: CBO has now incorporated into its analysis additional channels through which the ACA will affect labor supply, reviewed new research about those effects, and revised upward its estimates of the responsiveness of labor supply to changes in tax rates.

CBO anticipates that the ACA will lead to a net reduction in the supply of labor. In the agency’s judgment, the effects will be most evident in some segments of the workforce and will be small or negligible for most categories of workers. (The ACA also will slightly affect employers’ demand for labor, as discussed below, and the total effect on labor use will consist of the combined effects on supply and on demand.) In CBO’s view, the ACA’s effects on labor supply will stem mainly from the following provisions, roughly in order of importance:

  • The subsidies for health insurance purchased through exchanges;
  • The expansion of eligibility for Medicaid;
  • The penalties on employers that decline to offer insurance; and
  • The new taxes imposed on labor income

Some of those provisions will reduce the amount of labor supplied by some workers; other provisions will increase the amount of labor supplied by other workers. Several provisions also will combine to affect retirement decisions.

The ACA also could alter labor productivity—the amount of output generated per hour of work—which in turn would influence employment (for example, by affecting workers’ health or firms’ investments in training of workers). The effects on productivity could be positive or negative, however, and their net impact is uncertain, so they are not reflected in CBO’s estimates of labor supply or demand. Because the ACA could affect labor markets through many channels, with substantial uncertainty surrounding the magnitude of the effects and their inter-actions, CBO has chosen not to report specific estimates for each of the channels encompassed by its analysis.

Beginning in 2014, many people who purchase insurance through exchanges will be eligible for federal tax credits to defray the cost of their premiums, and some also will be eligible for cost-sharing subsidies to reduce out-of-pocket expenditures for health care. Those subsidies are largest for people whose income is near the federal poverty guideline (also known as the federal poverty level, or FPL), and they decline with rising income.

In 2014, for example, a single person or a family whose income is 150 percent of the FPL and is eligible for subsidies will pay 4 percent of their income for a certain “silver” health care plan purchased through an exchange; if their income is 200 percent of the FPL, they will pay 6.3 percent of their income for that plan. An increase in income thus raises the enrollee premium (and reduces the subsidy) both because the percentage-of-income formula applies to a larger dollar amount and because that percentage itself increases. People whose income exceeds 400 percent of the FPL are ineligible for premium subsidies, and for some people those subsidies will drop abruptly to zero when income crosses that threshold. Cost-sharing subsidies also phase out in steps with rising income, declining sharply at 150 percent, 200 percent, and 250 percent of the FPL.

CBO’s estimate of the impact that the subsidies will have on labor supply has three components: the magnitude of the incentive, the number and types of people affected, and the degree of responsiveness to the incentive among those who are affected.

For some people, the availability of exchange subsidies under the ACA will reduce incentives to work both through a substitution effect and through an income effect. The former arises because subsidies decline with rising income (and increase as income falls), thus making work less attractive. As a result, some people will choose not to work or will work less — thus substituting other activities for work. The income effect arises because subsidies increase available resources—similar to giving people greater income—thereby allowing some people to maintain the same standard of living while working less. The magnitude of the incentive to reduce labor supply thus depends on the size of the subsidies and the rate at which they are phased out.

Subsidies clearly alter recipients’ incentives to work and can certainly influence the labor supply of those who would gain eligibility by working and earning slightly less. But most full-time workers do not confront that particular choice—either their income is well above 400 percent of the FPL or they are offered employment-based health insurance and thus are generally ineligible for subsidies regardless of their income. Even so, one line of research indicates that the subsidies will affect the labor supply of many full-time workers with health insurance from their employer—precisely because they effectively forgo exchange subsidies when they take or keep a job with health insurance. If instead a worker switched to a part-time job, which typically does not offer health insurance, that worker could become eligible for exchange subsidies. In that view, exchange subsidies effectively constitute a tax on labor supply for a broad range of workers.

In CBO’s judgment, however, the cost of forgoing exchange subsidies operates primarily as an implicit tax on employment-based insurance, which does not imply a change in hours worked. Instead, the tax can be avoided if a worker switches to a different full-time job without health insurance (or possibly two part-time jobs) or if the employer decides to stop offering that benefit. The consequences of that implicit tax are incorporated into CBO’s estimate of the ACA’s effect on employment-based coverage—which is projected to decline, on net, by about 4 percent because of the ACA. Correspondingly, the negative effects of exchange subsidies on incentives to work will be relevant primarily for a limited segment of the
population—mostly people who have no offer of employment-based coverage and whose income is either below or near 400 percent of the FPL.

Nonetheless, another subgroup that has employment-based insurance does seem likely to reduce their labor supply somewhat. Specifically, those people whose income would make them eligible for subsidies through exchanges (or for Medicaid), and who work less than a full year (roughly 10 to 15 percent of workers in that income range in a typical year), would tend to work somewhat less because of the ACA’s subsidies. For those workers, the loss of subsidies upon returning to a job with health insurance is an implicit tax on working (and is equivalent to an average tax rate of roughly 15 percent, CBO estimates). That implicit tax will cause some of those workers to lengthen the time they are out of work—similar to the effect of unemployment benefits.

The implicit taxes that arise from the phaseout of the subsidies have effects on net income that are similar to the effects of direct taxes. With tax changes, however, the income and substitution effects typically work in opposite directions, whereas with the insurance subsidies the income and substitution effects work in the same direction to decrease labor supply. CBO’s estimate of the response of labor supply to the subsidies is based on research concerning the way changes in marginal tax rates affect labor supply and on studies analyzing how labor supply responds to changes in after-tax income.

The ACA significantly increases eligibility for Medicaid for residents of states that choose to expand their programs. In states that adopt the expansion, Medicaid eligibility is extended to most non-elderly residents whose income is below 138 percent of the FPL—including childless adults who previously were ineligible for Medicaid in most states regardless of their income. In states that have not expanded Medicaid, people whose income is between 100 percent and 138 percent of the FPL become eligible for subsidies through the exchanges; in those states, subsidies could decline abruptly if an enrollee’s income fell from just above the FPL to just below it (and vice versa). By 2018, CBO expects that around 80 percent of the potentially eligible population will live in states that have expanded Medicaid.

For some people, the ACA’s expansion of Medicaid will reduce the incentive to work—but among other people it will increase that incentive. As with exchange subsidies, access to Medicaid confers financial benefits that are phased out with rising income or (more commonly) eliminated when income exceeds a threshold; some people will thus work fewer hours or withdraw from the labor force to become or remain eligible (the substitution effect). Moreover, those financial benefits will lead some people to work less because the increase in their available resources enables them to reduce work without a decline
in their standard of living (the income effect).

At the same time, some people who would have been eligible for Medicaid under prior law—in particular, working parents with very low income—will work more as a result of the ACA’s provisions. In 2013, the median income threshold for that group’s Medicaid eligibility was 64 percent of the FPL (albeit with substantial state-to-state variation). The incentives and groups affected depend on whether a state has adopted the Medicaid expansion (and, in both cases, those incentives are intertwined with the effects of the exchange subsidies):

* In states that have chosen to expand Medicaid, the ACA now allows parents to qualify for Medicaid with income up to 138 percent of the FPL. And if their income rises above that threshold, those parents would generally be eligible for premium tax credits and cost-sharing subsidies for insurance purchased through the exchanges unless they are offered qualified employment-based health insurance. The subsidies will cover a smaller share of enrollees’ medical costs than Medicaid would, but under prior law those participants ultimately would have become ineligible for Medicaid and lost all benefits. As a result, some people who would have curtailed their hours of work in order to maintain access to Medicaid under prior law will now be able to increase their hours and income while remaining eligible for subsidized insurance.

* In states that have chosen to expand Medicaid, the ACA now allows parents to qualify for Medicaid with income up to 138 percent of the FPL. And if their income rises above that threshold, those parents would generally be eligible for premium tax credits and cost-sharing subsidies for insurance purchased through the exchanges unless they are offered qualified employment-based health insurance. The subsidies will cover a smaller share of enrollees’ medical costs than Medicaid would, but under prior law those participants ultimately would have become ineligible for Medicaid and lost all benefits. As a result, some people who would have curtailed their hours of work in order to maintain access to Medicaid under prior law will now be able to increase their hours and income while remaining eligible for subsidized insurance.

* In states that choose not to expand Medicaid, the availability of exchange subsidies also will lead some people to work more. Specifically, some people who would otherwise have income below the FPL will work more so that they can qualify for the substantial exchange subsidies that become available when income is equal to or just above the FPL significantly exceed the cutoff for Medicaid eligibility and because some low-income people live in states that are not expected to expand Medicaid.

A number of studies examining the impact of changes in Medicaid eligibility for parents and children have shown either no effects or small effects on the labor supply of single mothers; effects on two-parent households appear to be somewhat larger, in part because health insurance has stronger effects on the labor supply of secondary earners.

More recently, several studies have examined changes in state policies that affect childless adults—who constitute the majority of those gaining coverage through the Medicaid expansion—and larger effects have been reported. Some reductions in employment are reported among people who have gained Medicaid eligibility, although the findings differ regarding the magnitude and statistical significance of that effect.

Similarly, other research shows a rise in employment rates with the withdrawal of Medicaid coverage from childless adults who had previously been turned down for private insurance.

Because those studies examined state-level policy initiatives affecting program eligibility—instead of changes in eligibility attributable to income changes, which could merely reflect changes in employment—the results provide some useful insights into the potential effects of the ACA (even though other aspects of the studies raise questions about their applicability to an analysis of the ACA).

Taking that research into account, CBO estimates that expanded Medicaid eligibility under the ACA will, on balance, reduce incentives to work. That effect has a relatively modest influence on total labor supply, however, because the expansion of eligibility for Medicaid primarily affects a relatively small segment of the total population—both because most people’s income will significantly exceed the cutoff for Medicaid eligibility and because some low-income people live in states that are not expected to expand Medicaid.

No comments:

Post a Comment