Questions and Answers about Side Effects
Q1. What is the best way to read illustrated ebooks?
There are many options. Here's my view.
Q2. What is the primary reason that you have different estimates (as compared to previous studies) of the Economic Consequences of the Health Reform?
I find that the ACA has some very large taxes on employment and incomes. The existence of some of these taxes has been ignored in previous studies. My study is not at all unusual in terms of the effects of each unit (say, percentage point) of taxation. That’s why so much of Side Effects is devoted to identifying and measuring the various taxes.
Q3. Are you saying that the ACA’s eligibility rules amount to an unemployment insurance program?
Unemployment insurance (UI) pays benefits only to people who left jobs by layoffs, but not to anyone who left a job for another reason. In this regard, the ACA is more generous than UI. The ACA pays benefits to people who left jobs (that offered health coverage) for any reason: layoff, quit, early retirement, sabbatical, starting a new business, etc. Thus, the ACA is more like non-employment insurance. (Recall that there are two kinds of non-employment: “unemployment” and “out of the labor force”). Also note that UI benefits are subject to income tax, whereas the ACA's assistance is not.
Whereas UI increases the number of people unemployed, the ACA’s non-employment insurance will increase the number of people who are not employed (equivalently, it will reduce the number of people who are employed). It will do so by: encouraging employers to lay off employees, encouraging employees to quit or retire, discouraging non-employed persons from accepting jobs, and by discouraging employers from creating new jobs.
Q4. Aren’t UI benefits different because they pay cash for the unemployed to spend any way they want, whereas ACA beneficiaries have to spend their assistance on health insurance?
When it comes to the subsidies, this is an important difference. (The ACA’s penalties are owed in cash, though). According to the analysis in Side Effects, ACA subsidies are worth less than they cost the U.S. Treasury and have a smaller effect on employment than they would if the benefits were paid in unrestricted cash. To learn more about this issue, the book studies prior health insurance assistance programs and accounts for the fact that there will be people who are eligible for ACA assistance but do not take it.
The book also explains how the income tax treatment of ACA subsidies is different from unemployment benefits and other cash subsidies, and how this matters for the law’s economic consequences.
Q5. Aren’t people leaving their jobs voluntarily? Shouldn’t we celebrate that people can escape the drudgery of work, instead of calling it “jobs lost”?
Basic economics is clear that society as a whole has lost something valuable when either an implicit or an explicit tax results in less employment. (The only theoretical exception to this result, not applicable to the average American worker, is when market distortions would have resulted in excessive employment. In this situation, recessions should be celebrated too.)
An individual offered a payment to quit his job may celebrate that payment because he can either take advantage of it or, at worst, turn it down and continue to work. But how does the same individual feel about paying extra taxes to provide payments to other people who quit working? The UI features of the ACA are redistribution, plain and simple. Only half of redistribution is voluntary, and that voluntary half tells us little about the full consequences of redistribution.
Also note that the ACA’s taxes come on top of other taxes that had, on average, already caused Americans to work too little. As one of the ACA’s implicit tax causes a person to quit their job, that person will not only receive the ACA assistance, but also stop paying payroll taxes, pay less income tax, and perhaps receive benefits from other longstanding assistance programs like food stamps. In this way, a little bit of new assistance can put a lot of new burdens on taxpayers.
Separately, but most important, labeling the ACA’s employment effects as “voluntary” (deliberately?) obscures one of the most basic principles of public finance: that in the long run it doesn’t matter whether employees are discouraged from working or employers are discouraged from hiring. A job is a relationship between employer and employee and basic economics has nothing to say about “fault” when that relationship ceases or fails to exist.
When a new UI payment prevents a job from happening, journalists say that the potential employee chose not to work. But, from an economic point of view, the same result could be described as one in which the employer failed to raise the offered salary in order to better compete with what UI offers. Maybe the employee ought to be disappointed that his employer is unwilling or unable to compete with the benefit system for his time and talents?!
When a new employer tax prevents a job from happening, journalists say that the employer chooses to hire less. But it’s also true that the potential employee failed to reduce his salary demands (or failed to increase the value he creates as an employee) so that the employer would still find the job to be profitable even with the penalty. For the purposes of long-run analysis, basic economics has nothing to say about these distinctions; it just says that the tax prevented a job from happening.
Q6. Isn't the ACA irrelevant for the vast majority of Americans who live above the poverty line?
Quite the opposite. The employer penalty does not discriminate among employees on the basis of their family income: a worker from a high-income family counts just the same as a worker from a low-income family. The ACA's new health insurance subsidies are for families between 100 percent and 400 percent of the poverty line (on the basis of calendar year income) and thus technically unavailable to poor people. This 100-400 range is often called "low income" but the fact is that half of Americans live in a family with that income. A majority of people live in a family that could be in that range with a small adjustment in their work schedules or income reporting. An even larger majority of people live in a family that occasionally (but not necessarily perennially) has income in that range. Even the Medicaid expansion primarily applies to people in families above the poverty line. Also, families with older (but not elderly) adults get larger subsidies, and such families tend to be at the higher end of the 100-400% FPL range.
Of course, there are other eligibility criteria. Nevertheless, I estimate that the law directly changes work incentives for almost 50 percent of workers. During a typical week after the ACA is fully implemented, almost sixty million people (workers and dependents) will be experiencing significant labor market disruption, not to mention those experiencing lesser changes in their wages or work schedules.
This doesn't even begin to count the taxes or Medicare benefit cuts that will directly affect nonpoor families and are needed to pay for the ACA's new subsidies.
Q7. What are some of the other economic surprises in the book?
Q8. How could regular people possibly adjust their work habits in response to disincentives (e.g., implicit taxes) that they don't understand?
I don't assume that everyone, or even most people, have an intimate understanding of tax incentives. I use the historical experience from ACTUAL tax changes experienced by ACTUAL people, however knowledgeable or ignorant they were. The real questions are whether "this time is different" and whether taxpayers will be less knowledgeable than they have been in the past.
We also must recognize that people can get assistance and advice from others who are knowledgeable. Personal finance columnists began offering this advice as soon as the exchanges opened in 2013. I also expect large employers to help their employees navigate the new Obamacare rules, kind of like McDonalds help(ed?) their employees navigate the food stamp program.
In fact, there are many unemployed people who can "do the math." Take a look at what this recruiter learned as he interviewed job applicants:
Q9. How is it possible that the ACA causes workers to "work less productively" yet its GDP impact is less than its labor-hours impact (in percentage terms)?
The taxes created by the ACA are not uniform across demographic groups, employers, industries, and regions. Workers and firms will rearrange themselves -- in some cases changes places with each other -- in order to reduce their tax burden and enhance their subsidies. Most of these rearrangements are counterproductive -- that's why they were not done before the ACA -- but workers and firms do them because the tax savings is enough to justify the lost productivity.
If all workers continue to work the same number of hours, and just did so less productively, then the result would be a GDP impact that exceeded the labor-hours impact in percentage terms (both impacts are in the direction of less activity). But an additional effect of the ACA's taxes is to push less-productive workers out of working all together, in which case they are no longer counted in the productivity statistics.
Technically, my results say that the ACA reduces total factor productivity (by definition, this holds constant the composition or "quality" of the workforce) and increases GDP per hour by altering the composition of the workforce.
Q10. More than three hundred economists, including a few Nobel laureates, told Congress that the ACA was creating jobs. What do they have to say about Side Effects?
Side Effects is a book about the economY and documents the large labor taxes created by the ACA. These taxes – especially the implicit full-time employment tax and the high marginal penalty rates created by the small-firm exemption from the employer penalty – are conspicuously absent from the writings (at least those since 2007) by economists who support the ACA.
Even when neutral, nonpartisan institutions write about the state of the labor market now and in the near future, they rarely mention the ACA. Take the economic growth forecasts by the Federal Reserve Open Market Committee or the International Monetary Fund (but see below on the CBO). As recently as September 2013, the President of the United States himself insisted “you’re not going to meet an economist who says that [repealing the ACA is] the number one priority in terms of boosting growth and jobs … at least not a serious economist.”
I am not an expert on economistS, serious or unserious. My sources are limited to written methods and conclusions about the economY (including the economists’ letter), which I evaluate based on their merits rather than the number of economists who may agree. I did not and do not attempt to systematically survey economistS’ opinions or to put economic hypothesis up for election with the American Economic Association.
At the same time, I am repeatedly asked about the state of economistS’ opinions. For anyone who cares to obtain an amateur opinion on such matters, continue to read below. This material is not in Side Effects.
I see only three possible explanations for ACA-supporters’ silence on the ACA’s taxes:
(1) The economists who support ACA were, at the time of writing, not aware of the large taxes. Indeed, I wrote the book because many people – economists and noneconomists alike – were not fully aware of the ACA’s taxes.
I’ve met at least one signatory to the economists’ letter who was not aware of the taxes until I pointed them out. I recommended this person (and others) to send a letter of clarification to Congress that walks back some of the “strong conclusions” in the original letter, but to my knowledge no clarification letter has been written or will ever be written. Perhaps that’s because lack of awareness is not the real issue, or because it’s difficult to change one’s mind and therefore the facts documented in Side Effects are coming too late.
(2) Bad economics for a good cause: the economists who appear to support the ACA (e.g., they write “we support the ACA”) are really supporting fundamental change in the health care industry, and not the ACA specifically. I’ve met at least one signatory to the economists’ letter (different from the signatory cited above) who said that he/she was aware that the ACA itself does a lot of economic damage, but that it was important that something be done about healthcare, and that a bad law was better than no law because the bad law could be adjusted and revised to make it good.
There are other versions of “bad economics for a good cause”. For example, economists supporting the ACA may understand that the law creates economic costs but, based on their own value judgments, believe that the costs are overwhelmed by the law’s benefits. They are concerned that other political actors may put more/too much weight on the costs, so they pretend that the costs do not exist. Their objective here is not to get the economics right, but rather to pass a law that has more benefits than costs. Here's how M.I.T. Professor Jonathan Gruber explains it,
To be clear, I totally disagree with the approach of doing bad economics for a good cause; I’m just reporting what I was told. IMO, Americans deserve to be told the whole truth about their laws, and any economist engaged in bad economics for a good cause is practicing amateur political science. Also remember that I do not know how many ACA supporters are in this category (except that the number is at least two).
(3) Undisclosed research: Economists supporting the ACA have already researched the ACA’s tax provisions and concluded that they were small or inconsequential. They have been too busy with other things to write up their findings for other economists to read.
Logically, undisclosed research is always a possibility, so I ask advocates about this when I get a chance. So far I have not found any undisclosed research.
None of this proves that Side Effects is “right” or that nobody should support the ACA. It just attempts to explain why ACA supporters did not address the law’s major taxes, especially the implicit employment tax and the anti-competitive effects of small-firm exemptions.
Q11. Isn't the ACA going to "bend the cost curve" and stimulate entreprenuership? Have you accounted for that?
If we had a law that only eliminated health insurance waste, that would be a good thing and might also increase employment. Health insurance waste has both wealth and substitution effects on employment, but there are studies suggesting that health insurance waste is a lot like a labor tax and that the substitution effect might dominate. In other words, a law that only cut health insurance waste might increase employment in much the same way that a tax cut would.
However, even if the ACA does reduce health insurance waste, and did so without offsetting that cut with added administrative costs for employers and employees, the law includes a lot more tax effects. Side Effects counts all of them, and finds that the other taxes in the ACA overwhelm the effect of cutting health insurance waste.
Although the total effect of the ACA is to reduce employment, I do find that the law will increase employment in a small subset of the economy. These are especially employers who, before the ACA, found health insurance too costly to offer and suffered a competitive disadvantage in the labor market because competiting employers were offering it. Perhaps one might label such employers as "entrepreneurs," especially because the employers tend to be small. However, the size of their incentive to expand employment is bounded by the gap between their costs of offering coverage and their competitors' cost of offering coverage.
Q12. Does the Congressional Budget Office (CBO) agree with Side Effects?
Based on their writings and one meeting I had with CBO economists, I can guess that they agree with parts of Side Effects. They wrote in 2010 that the ACA creates large implicit marginal income tax rates (CBO 2010, p. 48). Side Effects’ chapter 5 agrees with them, although I do not know enough about the underlying details of CBO’s work to say whether I agree with their exact marginal income tax rate estimates (or whether they would agree with mine).
For the first time in 2014, the CBO briefly explained that the ACA has implicit employment taxes (CBO 2014). I also met with them in 2013 to show them my methodology for estimating implicit marginal tax rates that include implicit employment taxes. But the 2014 CBO publication is too cryptic, and our 2013 meeting too brief, for me to know whether we agree on the details of implicit employment taxation. As far as I understand it, I agree with their conclusion that “… the loss of subsidies upon returning to a job with health insurance is an implicit tax on working.” However, that same sentence is surrounded by two numerical estimates that I find puzzling, perhaps because I do not understand what the estimates are supposed to represent.
The 2014 CBO report also contains these three consecutive sentences: “In that view, exchange subsidies effectively constitute a tax on labor supply for a broad range of workers. [break for a new paragraph] 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.” I don’t understand what “operates primarily as an implicit tax on employment-based insurance” means. One reading of these sentences would be that CBO disagrees with Adam Smith’s theory of equalizing differences (Side Effects Chapter 7). But I expect that CBO would take more than three sentences to discard Adam Smith, so for now I must conclude that I am not getting their true meaning.
It seems clear that, as of February 2014, the CBO expected the ACA to have a lesser long-run effect on aggregate work hours (at most 2 percent) than I do (3 percent or more). One might reasonably deduce that CBO disagrees with some component of my estimates. On the other hand, unlike so many agencies in Washington, the CBO has a track record of staying abreast of the latest research and using that research to update their estimates and approaches. They had seen not Side Effects when they wrote in February 2014, and it is possible that reading it and improving their approach will be part of their ongoing work on public policy and the economy.
I don’t know what the CBO thinks about the anticompetitive effects (in the labor market) of the ACA’s employer penalty because they have not written about them.
Unlike ACA advocates, the CBO never fell into the trap of saying that Romneycare = Obamacare, therefore Obamacare has no noticeable effect on the labor market (CBO 2012). This makes me believe that they agree with the high-level conclusion of Side Effects chapter 10: that Romneycare is too different from Obamacare to use as a benchmark for the economic consequences of the ACA.
Q13. Who is Maeve?
A strong woman