Sunday, June 28, 2015

Thoughts on Obergefell v Hodges

On Friday morning the US Supreme Court declared that states cannot deny gay people the right to marry.

There's a common line of criticism I'm hearing from conservatives--including some of the justices who dissented in Friday's case--that goes like this: without necessarily making it clear whether they think gay marriage should be legal, they assert that "5 unelected judges" was not the right way to legalize gay marriage and that gay people should have sought validation through legislatures instead. I think there are a lot of things wrong with this critique.

For one, the critique sounds like a derailing tactic. Arguing for a more difficult path to legalization is a way of opposing gay marriage without having to stake your reputation on explicitly anti-gay arguments. I think by and large we should just call out these arguments for what they are: arguments against gay marriage from people who know, deep down, that they're wrong.

Even ignoring that, I think these critics are wrong that legalization through the legislatures would somehow be more legitimate or democratic. For one thing, it isn't true that only "5 unelected judges" made this decision--while far from unanimous, there was already a broad consensus among lower courts that the constitution protects the right of gay people to marry. But more importantly, the court's decision was better than a statute, because it recognized that equality in marriage is a fundamental right, not a privilege that legislatures can grant or revoke. The gay rights movement has experienced false starts before, and has experienced backlashes, sometimes violent, in the US and abroad. Marriage is supposed to be something permanent, something too fundamental to allow the next governor, the next state assembly, or the next ballot initiative to revoke. Friday's ruling gives gay people some degree of protection from future backlash, without which we can't truly say their marriages are equal.

One of the most common anti-gay talking points more generally has been that we shouldn't overturn "thousands of years" of traditional marriage. This, I think, has always been their worst line of argument, because all it really says is that gay people have always been a relatively small minority. That's just biology--homosexuality only seems to occur in about 3 to 4 percent of the population, and that has been more or less the case in every society throughout history. Pointing out that a group has been oppressed for millennia is not argument against protecting their rights but an argument in favor.

Overall, Friday's decision has been one of the most heartening events in recent US history. We pride ourselves on being a beacon of freedom, but spent the past couple decades mired in various anti-freedom enterprises, from NSA overreach to the torture of prisoners detained indefinitely without any hope of trial or due process. On Friday, in at least one small respect after a decade of darkness, the USA resumed the cause for freedom--not the first, but one of the first, and by far the largest country to affirm the freedom to marry for all LGBT people.

Thoughts on King v Burwell

If you haven't heard, Burwell won. A loss would have meant almost 9 million Americans would have lost their health insurance and tens of millions more would have seen at least a 47 percent increase in their insurance costs. 6 justices on the Supreme Court decided to avoid that outcome.

Some conservatives have tried to argue that the decision was one of judicial overreach, of an unelected court rewriting legislation. I think the opposite is true. The interpretation that the plaintiffs in King sought to impose was never more than a technicality based on a mere four ambiguous words--total--in an almost thousand-page bill, an interpretation that, if imposed by the court, would have fundamentally altered the whole law, imposing a particularly draconian regulation that Congress never explicitly put there.

The court's ruling on Thursday morning was a bit surprising and weird. To be sure, I expected a ruling against the plaintiffs, but it was surprising because like most court watchers, I expected it to be decided under a Chevron doctrine, or perhaps constitutional avoidance, but not a ruling on the plain text. The decision was weird because the court held both that the four words in question--"established by the State"--were ambiguous but that the context nevertheless compels us to interpret them to include both state-run and federally-run exchanges.

The way I interpret this is the court is saying that Congress cannot hide mountains in molehills. The technicality that the architects of the plaintiff's case in King sought to exploit was not merely an unintended effect of a law--the court does not protect Congress from the unintended consequences of legislation--but rather a whole new regulatory regime supposedly being promulgated in the space of just 4 words, that would have fundamentally altered the way in which the legislation as a whole works. The court ruled that such a regulation requires more than 4 words to create. The court ruled that if Congress wants a regime in which only states that run their own exchanges are eligible for subsidies, that fact must be explicitly articulated in the bill, not merely implied through the use of tricky language in one four-word mention in the bill. The court said that neither the IRS nor the courts have the power to creatively exploit typos and technicalities after the fact to fundamentally alter an act of Congress.

So, as an advocate for public health, I guess I'm glad that nearly 9 million Americans won't be losing health insurance--though I'm not sure that maintaining the status quo counts as a "win." But frankly, I'm much more pleased that we now have a precedent like this on the books. To see why, just think about one of the common--but most outrageous--objections that conservatives raised against the ACA in the first place: death panels.

The conservative paranoia over "death panels" comes from this passage from Section 1233 of HR 3200 that was briefly inserted into the ACA but ultimately deleted from the final bill for political reasons:
"For purposes of reporting data on quality measures for covered professional services furnished during 2011 and any subsequent year, to the extent that measures are available, the Secretary shall include quality measures on end of life care and advanced care planning that have been adopted or endorsed by a consensus-based organization, if appropriate. Such measures shall measure both the creation of and adherence to orders for life-sustaining treatment"
Basically, the bill would have authorized Medicare beneficiaries to see a doctor once every 5 years--if they want--to make plans for end-of-life care, and would have included doctor's willingness to offer these consultations and adherence to the patient's preferences expressed in them as part of their quality reporting criteria. There is absolutely nothing in the bill that said that government bureaucrats--a "death panel"--would decide who gets care and who doesn't. But critics charged that the language of the bill could be read in a way that authorized HHS to do exactly that.

Thursday's court ruling said otherwise. Applying the precedent from King, Congress cannot secretly promulgate such a massive regulatory change through language this vague. Shouldn't we all breath a sigh of relief knowing that Michael Cannon can never sue the government to force them to kill your grandmother?

Monday, June 22, 2015

Are liberals more biased in social science?

...the answer to headlines that ask a question is almost always "no."

On twitter, Avik Roy pointed out this post on a study that performed an experiment on how biased people are when reading social science research papers, and found that the bias was much higher in liberals than conservatives.

Here's the experiment. Two groups were each shown two papers. In the first group, they gave subjects a research paper that concluded that affirmative action was bad for black people and a paper that concluded that same-sex relationships were just as healthy as opposite-sex relationships. The second group was shown essentially identical papers, but with the conclusions (and presumably data, etc) reversed so that they showed that affirmative action was good for black people and that same-sex relationships were less healthy than opposite-sex relationships. The result:
"People were asked to indicate how true they considered the article to be, and how biased they considered the author to be.
    And the resounding answer was: Liberals were far more biased. Liberals viewed the articles reporting "liberal" results (affirmative action and same sex relationships are good) as truer and reflecting less author bias than the articles reporting "conservative" results. Conservatives, in contrast, viewed the truthfulness and bias in the articles as nearly identical, regardless of their results."
Liberals were far more likely than conservatives to label the papers whose results they agreed with as true and unbiased.

Ok, this is a bit weird. I don't think anyone would seriously suggest readers should accept a result as "true" based on one paper, even if the methodolgy looked sound. In science, replication is everything. There are one-off results, there is fake data, there are typos, there is p-hacking. I do think people are inadequately critical of papers whose results they agree with--weak methods should fail to revise our priors, not strengthen them--but this study asked respondents whether they thought the result was "true" not (as far as I can tell from the description) to list out criticisms of the methods. Criticisms of methods should be independent of our priors, but as for assessing "truth" of the result, a rational Bayesian updater would require more evidence to revise a strong prior belief than a weak prior belief because, as they say, "extraordinary claims require extraordinary evidence."

Hence, rational, critical evaluators should be somewhat "biased" towards their prior beliefs. At issue in this experiment was the fact that the "bias" was asymmetric across groups despite the fact that both groups saw evidence of identical strength. Under the assumption of rationality, this means Liberals had stronger prior beliefs. The authors claims that this means liberals are more "biased."

But take a look at the topics chosen. I've never heard a conservative argue that affirmative action was harmful to the black people it is designed to help. That's an extraordinary claim for any part of the political spectrum. And while there are far-right factions that do argue that gay relationships are less healthy than straight ones, this is not generally a strongly held belief among conservatives. On both questions, the authors chose topics about which conservatives have fairly weak priors that don't necessarily even correspond to the belief that the authors label as "conservative." But these are two questions about which most liberals do have fairly strong prior beliefs.

Thus, I'm left to wonder whether liberals really are more biased in general--that is, have stronger beliefs about most things--or whether the authors merely singled out two questions on which liberals have stronger priors.

I don't really think that liberals and conservatives have exactly equal amounts of "bias." But since I have no prior beliefs about which group is more biased, I take equivalence as the default until I come across good enough evidence to revise that prior. This article failed to revise my priors.

Friday, June 19, 2015

Consumption taxes and the "double taxation" of income

A short sentence in this piece about Rand Paul's tax plan set off a chain reaction in my brain that can only be resolved by blogging. Here it is:

"This is the traditional argument for flat taxes: they're a form of consumption tax, and economists think consumption taxes help growth by exempting savings and investment from taxation."
Clicking through to their link brings us to a common talking point about consumption taxes:
"Economists tend to find that consumption taxes are better for the economy than income taxes, because income taxes discriminate against savers.
     To see why, imagine you make $50,000 in wages and there's a flat 20 percent tax on all income. You'd pay $10,000 in taxes on your wages. That leaves you with $40,000.
     Now you've got a decision to make: do you want to take $5,000 of the $40,000 you have left and invest it, or do you want to take that $5,000 and spend it on a really awesome television? If you invest it and make money off the stocks, then the thing you bought with your money—the profits those stocks made for you—will get taxed again. If you just buy the TV, the government doesn't tax you a second time."
I'm afraid Dylan Matthews has engaged in an unintended slight-of-hand here: you don't generally pay income taxes on investment income.1 We have a capital gains tax that is separate from the income tax.

So the real difference between income and consumption taxes is that consumption taxes let you invest first and then pay taxes on both the income and the interest earned on it, while income taxes require you to pay taxes first, reducing the amount you can invest, but then doesn't tax the interest income. The question is, which is more efficient?

The answer is that consumption taxes are weakly more efficient, but not for the reasons commonly argued.2 In fact, income taxes don't distort the investment decision any more than consumption taxes do. The only difference between the two is that the income tax distorts the labor supply decision more. For illustration, consider the basic DSGE model of an infinitely lived representative household.The household problem is to maximize \begin{align*}\max_{C_t,L_t,K_{t+1}}U=\sum_t \beta^t u\left(C_t,1-L_t\right)&\\ subject~to~K_{t+1}-\left(1-\delta\right)K_t+\left(1+s\right)C_t\leq\left(1-\tau\right)w_tL_t+r_tK_t& \end{align*} where [$]C_t[$],[$]L_t[$], and [$]K_t[$] are consumption, labor, and capital, respecively, in period [$]t[$], [$]w_t[$] and [$]r_t[$] are the wage rate and interest rate, [$]\beta,\delta \gt 0[$] are constants, and [$]u\left(\centerdot \right)[$] is a period utility function satisfying standard convexity and Inada assumptions. We'll let [$]u_{c,t}[$] and [$]-u_{l,t}[$] denote the derivatives of utility respect to consumption and labor respectively. Firms seek to maximize profits according to a linearly homogenous production function given by [$]Y=AL_t^\alpha K_t^{1-\alpha}[$] by solving the problem:[$$]\max_{L_t,K_t} AL_t^\alpha K_t^{1-\alpha}-w_tL_t-r_tK_t.[$$] And finally, the government taxes consumption and labor at rates [$]s[$] and [$]\tau[$] respectively to finance government spending [$]G_t[$] according to it's budget constraint given by [$$]sC_t+\tau w_t L_t=G_t.[$$]

The first order conditions for the household problem are \begin{align} Labor~Supply:&~\frac{u_{c,t}}{u_{l,t}}=\frac{\left(1+s\right)}{\left(1-\tau\right) w_t} \\ Euler:&~u_{c,t}=\beta E_t\left[u_{c,t+1}\left(1-\delta+r_{t+1}\right)\right]\\ Budget~Constraint:&~K_{t+1}-\left(1-\delta\right)K_t+\left(1+s\right)C_t=\left(1-\tau\right)w_tL_t+r_tK_t \end{align} Notice right away we can see that the first order condition for investment [$]K_{t+1}[$] actually doesn't have any tax rates in it at all. These taxes don't actually distort the investment decision at all, but rather distort the consumption and labor decisions—this shouldn't be very surprising since the two taxes are a consumption tax and income tax, after all. From the firm problem we have: \begin{align} w_t&=\alpha A\left(\frac{K_t}{L_t}\right)^{1-\alpha} \\ r_t&=\left(1-\alpha\right) A\left(\frac{L_t}{K_t}\right)^{\alpha} \end{align} and of course, [$$]s=\frac{G_t-\tau w_tL_t}{C_t}[$$] follows from the government's budget.

It turns out this model has a steady-state solution, so stripping out all the time subscripts and substituting out [$]r[$]: \begin{align} \frac{u_c}{u_l}&=\frac{\left(1+s\right)}{\left(1-\tau\right) w} \\ \left(1+s\right)C&=\left(1-\tau\right)wL+\frac{1-\beta}{\beta}K\\ w&=\alpha A\left(\frac{K}{L}\right)^{1-\alpha} \\ \frac{1-\beta+\delta\beta}{\beta}&=\left(1-\alpha\right) A\left(\frac{L}{K}\right)^{\alpha} \label{fixedK}\\ s&=\frac{G_t-\tau wL}{C} \end{align} From here it should be readily apparent why it is incorrect to say the income tax discourages investment. To see this, consider the case where labor supply is fixed. Then from equation [$]~\eqref{fixedK}[$] we have that [$]K[$] is fixed as well—that is, if the income tax does not cause labor supply to change, then it follows that it does not have any effect on investment either. And from there it follows that if labor supply is fixed, then households are totally indifferent between income and consumption taxes, and that consumption is identical either way. In other words, the income tax doesn't discourage investment, it discourages labor, and affects investment only indirectly by reducing labor supply and thus the amount workers have to invest.

But in the real world, labor supply is not fixed. And if you do the algebra from the system above you'll see that it turns out that when labor is not perfectly inelastic, the income tax causes a relatively larger distortion of the labor supply decision than consumption taxes do, meaning that the consumption tax is relatively more efficient. That is, for a given amount of government revenue, financing it with consumption taxes will result in higher consumption, higher labor supply, and higher utility.

Of course, this is an extremely silly way to analyze tax policy for the simple reason that the vast majority of government spending is actually just transfers from rich to poor. That is, transfers based on income. Switching from income to consumption taxes can therefore exacerbate the labor disincentive through the spending side of the ledger in even more detrimental ways.

1 There are two types of capital gains taxes in the US: for short-run investments sold (I think) within a year of being purchased, the capital gains are taxed as personal income under the income tax. For long-term investments they are taxed according to a separate flat tax rate and not considered income for the income tax.

2 A standard practice in economics is to call a policy associated with a potential Pareto improvement more efficient, regardless of whether the potential improvement actually exists within the real-world policy space. A potential Pareto improvement is a policy that makes at least one person better off but where all of the people harmed by it could theoretically be fully compensated with lump-sum transfers financed with the surplus from those made better off, such that every individual would either prefer or be indifferent to the policy change plus transfers. In practice, however, these theoretical transfers are never actually feasible because the government doesn't possess that kind of information, and can't costlessly administer that kind of bureaucracy. Thus, the relevance of economic "efficiency" is pretty dubious—it would be more informative to look at the welfare effects of the real-world policy, including the distributional effects. When income distribution is taken into account, income taxes are generally preferable to consumption taxes because income is more highly correlated with household types (although this is not unambiguous; the correct answer depends on the empirical incentive effects, nature of the household heterogeneity, and criteria by which welfare is measured.)

Tuesday, June 16, 2015

Do the uninsured really pay full charges?

In my previous post I pointed out a study that suggested that providers actually bill less, total, to uninsured patients for a doctor's visit than they bill to insured patients (and their insurers). This flies against widespread claims that the uninsured have to pay more for healthcare than insured patients because they have to pay full charges instead of the negotiated discounts insurers get. It also challenges the conclusions of all research studies based on charge data—charges being more than double, on average, than what providers actually bill for—which had been justified on the grounds that uninsured patients typically have to pay full charge prices. Today the New York Times has a column out highlighting a study in Health Affairs by Ge Bai and Gerard F. Anderson that compares charges to costs at various types of hospitals. So if charges are meaningless, what good is a charge-to-cost comparison? The Times:

"The typical response from the hospitals is that their list prices are not relevant to most patients because they don't pay the full price. The two big government insurance programs, Medicare and Medicaid, pay far less than the list prices; private insurers typically bargain for discounts. Even so, high list prices raise the ceiling from which discounts are negotiated and thus drive up premiums for many privately insured patients.
    "High prices will hit millions of people who will remain uninsured in coming years for one reason or another despite passage of the Affordable Care Act, according to the Health Affairs study."
So the Times, and the Health Affairs study, repeat the common refrain that charges matter because uninsured patients have to pay full charges. Is that really true?

I've not done a complete review of the literature on this question. However, I did look up each citation offered for this claim and attempted to trace these citations back to the original data. It has been a lesson in citogenesis. Bai and Anderson cited three papers in defense of their assertion that uninsured patients typically have to pay full charges:

  1. Anderson (2007)
  2. Thompkins, Altman, and Eilat (2006)
  3. Reinhart (2006)
None of these three papers directly presents data on the prevalence of uninsured patients being billed for full charges, but instead each outsource this claim to other papers, or to unsourced information. Anderson (2007) actually relies on Reinhart (2006) for this claim, and Reinhart (2006) in turn offers up this citation:
EIU District 1199 Care for Ohio, "Twice the Price: What Uninsured and Underinsured Patients Pay for Hospital Care," March 2005,{56490583-267C-4278-BC56-A7128CE248A8}/uploads/{374CBAD9-740D-48BC-8536-92ECD76D1444}.PDF (accessed 1 July 2005)
Unfortunately, this source appears to no longer be accessible.

In addition, Anderson (2007) claimed to have confidential information from testimony and documents prepared for trial in lawsuits between providers and "self-pay" patients over their billing practices. While we aren't given any specifics or access to the underlying data, the existence of trials over these practices suggests, at least, a perception that self-paying patients are billed in unfair ways such as being denied discounts similar to the ones insurers get. Since the author himself is the primary source, I will refer to the endnote numbers in which he cites his own confidential communications and observations. In citation 19, Anderson notes that he had access to the testimony of the executives of healthcare providers in these lawsuits, who spoke to the reasons that their firms only offered small discounts to uninsured patients. In citation 22 Anderson notes that uninsured patients typically can get at least a 10% discount, if they ask, which is a tiny fraction of what would be offered to insurers but enough to prove that providers did not withhold discounts for regulatory reasons as some executives had claimed. Citation 23 states

"For example, Hospital Corporation of America's Web site states that HCA 'provides free care for any patient who receives non-elective treatment and whose household financial resources and/or income is at 200 percent or below the Federal Poverty Level.' Some patients above 200 percent of poverty will qualify for 'a managed care like discount.'"
which suggests that some of these providers offered discounts of up to 30 percent of charges, still a lot less than most insurers would get. I don't doubt the honesty of Anderson's reporting; however, this evidence is purely anecdotal. The existence of some hospitals with abusive billing practices is neither surprising nor direct evidence of the claim at hand: that uninsured patients are usually billed at full charge rates. It does, however, get at a few common themes throughout the literature I reviewed: that providers in general offered discounts and charity care to individuals based not on insurance status but rather ability to pay (more on that in a moment), that many providers lack a systematic way to make proper adjustments for patients outside of the insurer framework, and that some providers attempt to gouge uninsured patients by only making them aware of their discount and charity care programs after they refuse to pay full charges.

That leaves us with Thompkins, Altman, and Eilat (2006). They do not merely claim that uninsured patients typically pay full charges (or close to it), but that this has been a deliberate pricing strategy by providers to offset cuts from private and public insurers (as opposed to an unintentional bureaucratic error). They have constructed a whole historical narrative of the charge-to-cost ratio premised on this claim:

"As private third-party payers consolidated in the early 1990s and their market clout grew, they moved away from negotiating with hospitals based on charges and toward contracts based on lower fee schedules or negotiated rates. Accordingly, billed charges defined prices for a shrinking proportion of patients. Hospitals responded by marking up billed charges even faster than the costs of care for such patients. This scenario resulted in an increasing gap between billed charges and the prices paid by most payers. ... This gap has grown steadily since the early 1980s and has accelerated in recent years. This acceleration is attributable to several factors. The employer-sponsored insurance market shifted in favor of managed care, giving those health plans volume and clout to obtain greater discounts from hospitals. Medicaid enrollment grew significantly, providing more stable revenues than the uninsured, but nevertheless paying rates much lower than actual costs. The enactment of the Balanced Budget Act (BBA) of 1997 lowered the growth in Medicare payments, resulting in a cut of roughly $70 billion, or 9.1 percent in hospital payments, over the five-year period 1998–2002 from the pre-BBA Medicare baseline. Thus, payments from public programs and many private third-party payers were increasingly below what hospitals believed to be appropriate for the services provided. All of these forces put pressure on hospitals to squeeze more revenues from a diminishing pool of other sources. The main technique used by many hospitals was to raise their billed charges. This raised to varying degrees the prices charged to payers not under contract, payers with contracts in which most or all payment rates were linked to charges, and services that were outside the scope of fixed- or negotiated-price contracts. Hospitals have continued to rely on charge-related sources of revenue, even from groups that generate very limited amounts of revenues such as self-paying patients."
This is a theory of cost shifting, and as Austin Frakt has written about in other contexts, it is bunk. Granted, this cost-shifting from public and private insurers onto the uninsured is a bit different than the cost-shifting from public onto private insurers that Frakt debunked, but I think the same theoretical framework applies, and any case the Thompkins-Altman-Eilat claim is substantially less credible than the conventional cost-shifting story. Uninsured patients paying the burden for insured patients? Um no. For what it's worth, Anderson (2007) contradicts it: only 10 percent1 of billed charges to uninsured patients are ever recovered. Even the sources offered up by Thompkins et al contradict it: Prottas (2007) suggests that only 7 to 10 percent of uninsured patients' medical debts are ever recovered, and also repeats the result from Anderson (2007):
"When I performed pooled crosssectional regression to examine the relationship between increases in charges and net revenues over the period 1984–2004, I found that hospitals received only a very small proportion of the increase in charges above the rate of increase in costs"
which suggests that increasing charges had no effects on what uninsured patients paid (in my opinion, further weakening the claim that uninsured patients are typically billed for full charges). Thompkins et al went on to say
"In our interviews,[2] hospital officials reported that they were reluctant to bill the uninsured for less than full charges because of insurers' common negotiating practice of insisting on being charged the same as the lowest-paying customers. They also cited Medicare guidelines and kickback regulations as reasons for not offering discounts to self-pay consumers. Related to this, they have operated with the understanding that they were subject to prosecution for fraud and abuse if they gave discounts without prior negotiation to any patients (even if they were poor and uninsured)."
This was a common theme throughout the papers I reviewed: hospital executives claiming that various anti-fraud regulations actually require them to bill uninsured patients for the full charges—an argument undermined by the fact that many of the same hospitals these executives work at do offer at least modest discounts to the uninsured.

The actual data in Thompkins et al is on the time-trend of the charge-cost ratio, not the proportions of charges that uninsured patients pay. For the claim that uninsured patients pay a larger share of charges they cite two other papers. The first is Pryor and Seifert (2003) which attempted interviews with administrators at 30 hospitals, getting only 4 responses. Some of the executives who responded claimed that regulations prohibited discounts for uninsured patients—again, a claim that is untrue but suggests some hospitals do charge full charges to uninsured patients. Pryor and Seifert further suggest that providers that do discount lack effective systems to handle this

"If patients are not identified as qualifying for public programs or charity care, hospitals do not have standard mechanisms for distinguishing among patients in terms of the size of their bills and their ability to pay them, or for informing patients of their ability to negotiate discounts when appropriate."
That suggests that the issue is not so much that hospitals actually do bill for full charges, but that bills to uninsured patients are more likely to be erroneous because of the lack of standardized discounts.3

The other paper cited by Thompkins et al in support of the claim that the uninsured pay full charges was Prottas (2007), which was primarily concerned with medical debt. Prottas (2007) surveyed a similarly small number of hospitals—"a half dozen"—and found that all of the hospitals did have programs to offer discounts of up to 100 percent of billed charges for uninsured patients and others who could not pay. Interestingly, in all 6 cases, these discounts were adjusted based on income and not insurance status, so insured patients could get additional discounts on out of pocket expenses. However, for some income groups discounts were as small as 10 percent. Moreover, in all but one case, billing departments attempted to collect full charges first, and did not tell patients of their discount programs until after they refused to pay.4

Combined, the evidence supporting the New York Times article consists of a non-scientific sample of 10 hospitals and some non-specific, anecdotal claims from confidential trial testimony. Moreover, these samples overwhelmingly say that the providers do, in fact, routinely discount the medical bills of uninsured patients by at least 10 to 30 percent and as much as 100 percent for the poorest patients, although possibly not as much on average as they would discount for those who have insurance. Regulatory uncertainty may contribute to the complexity and opacity of hospital pricing. This evidence does not support the claim that the uninsured pay full provider charges, nor does this evidence support the claim that higher charges imply higher costs to uninsured patients. We need a larger and more representative sample to compare the actual amounts that insured and uninsured patients are billed after discounts and adjustments are made.

As a final point, I want to note that none of this is criticism of any of the papers I reviewed in pursuit of validating the New York Times and Bai and Anderson's claim about the uninsured paying full charges. Most of these cited papers were not primarily concerned with that claim as a research question, and the fact that they do not offer substantive evidence for this claim should not be taken as a suggestion that they do not offer value on the topics they actually aimed to study.

1. The source offered for the 10 percent figure is this. I did not find that claim in that source, but their data is such that it seems plausible the authors would have been able to calculate this figure. I will assume that Anderson had seen results from that study that did not make it into the final paper. In any case, Prottas (2007) provide similar numbers from—as far as I can tell—independent sources.

2. It's not clear to me what interviews the author is describing—the paper did not contain any description of interview data collection. However, one of the papers they cite, also from the Commonwealth Fund, did conduct interview surveys of hospitals—with only 4 respondents—that offered results fitting this description. They also cite this paper, which describes surveys of "a half dozen" hospitals.

3. The high prevalence of hospital billing errors, of course, are a similarly notorious and poorly sourced claim.

4. Note: all providers in each of these papers that surveyed providers did have programs that collected patient's income information and, if they were eligible, helped them enroll in public assistance programs such as medicaid and medicare. Only one of the hospitals surveyed used this information to automatically refer low-income patients who did not qualify for a public program to the hospital's discount programs. Whether the lack of automatic discounts represents a deliberate attempt to gouge patients or is merely an institutional failure, this could be a basis for the widespread perception that uninsured patients have to pay full charges for healthcare.