Certificate-of-Need Laws Still Fail Patients — Even After a Decade of Reform

hospital charges in states without CON are 5.5 percent lower five years after CON repeal 

By Thomas Savidge of AIER. Excerpts:

"Certificate-of-Need (CON) laws require the approval of states’ health planning agencies for health care providers to engage in regulated actions such as opening or expanding facilities or purchasing equipment. Additionally, in many states with CON regulations, the decision to grant a CON is made by a board whose members may work for incumbent providers. This is sometimes referred to as a “competitor’s veto.” [economist Matthew] Mitchell also notes that in all but six CON states, incumbent providers are allowed to participate in the process and object to the application of a would-be competitor and often use the objection as leverage against the potential competitor from encroaching on their territory. Mitchell calls it “a type of territorial collusion that would be a per se violation of the Sherman Antitrust Act were it not facilitated by the state.”

CON laws were first applied to healthcare by New York State in 1964, and by 1970, 25 additional states had similar regulations. In 1974, Congress passed the National Health Planning and Resources Development Act (NHPRDA), where the federal government threatened states into adopting CON regulations by saying they would withhold federal funding for healthcare for any state without CON laws. 

The desire to push CON came from a misguided belief that such regulations could, in Mithcell’s words, “cause hospitals to acquire fewer beds, fill them with fewer patients, and therefore spend less money.”

The threats to withhold federal funding never materialized, but by the early 1980s every state had a CON program for healthcare. As Medicare reimbursement switched from retrospective (hospitals get paid whatever they spend with little incentive to control costs) to prospective reimbursement (hospitals are paid a fixed, predetermined amount for services), CON policies were reexamined. 

Finding that CON laws did little to control costs, especially under prospective reimbursement, policymakers in DC rolled back CON requirements. While the federal CON requirements were repealed, most states maintained CON regulations, which led to greater variation among state CON regulations. By 1990, eleven states had followed suit and repealed CON laws, with only Wisconsin reinstating the program. By 2000, Indiana, North Dakota, and Pennsylvania had repealed most CON laws. 2000 Wisconsin has since re-repealed its CON regulations. In 2016, New Hampshire was the last state to fully repeal CON.

In their 2025 review of the academic literature on CON, Courtemanche and Garuccio find “in at least some cases, CON laws restrict both entry of new competitors and expansion of existing hospitals. The reduction in competitors increases the number of procedures in hospitals.” They continue by noting such findings provide

“[L]ittle evidence that this translates to increased prices or higher hospital profitability, and hardly any research tests for reduced closure rates. Studies on hospital efficiency and quality of care for procedures performed exclusively at hospitals mostly point to null or negative effects, but evidence on quality is more favorable for services that can be provided outside of hospitals.”

While there may be other factors at play (such as government subsidies, the Affordable Care Act implementation, or the variation of CON laws among states), costs can still be passed onto patients without changing prices. Higher costs may look like longer wait times (whether in a hospital or fewer scheduling options), staffing shortages, or the opportunity cost of patients traveling to see a specialist either on the other side of their state or in other states.

If one thing is clear, CON regulations have failed to deliver on the promise of affordable and accessible healthcare.

As of December 2025, 15 states have fully repealed CON regulations. Additionally, numerous states have reformed their CON regulations to shrink the healthcare services that require a CON."

"Arizona, Minnesota, and New Mexico limit CON requirements to ambulatory services while Indiana, Ohio, and South Carolina (as of 2025) only apply CONs to nursing homes. Hawaii regulates the most activities, requiring CON approval for 28 services and technologies. Mitchell finds that nursing home beds are the most frequently regulated, followed by psychiatric services, new hospitals, and intermediate care facilities for individuals with intellectual disabilities. Investment thresholds triggering CON review vary from state to state and are generally lower for non-hospital providers (a $3 million expenditure trigger for ambulatory services in Maine) than for hospitals (excess of $12.365 million in capital expenditures in Maine).

From Mitchell’s findings and his survey of the academic literature, he finds that CON laws are generally associated with high variable costs in general acute hospitals, fewer available hospitals, higher Medicaid costs for at-home care and long-term care, and higher health expenditures. The more stringent and numerous the CON laws in the state, the worse access and affordability for healthcare.

In states that did repeal CON laws, Mitchell found that hospital charges in states without CON are 5.5 percent lower five years after CON repeal. Additionally, safety-net hospitals in states without CON had higher margins than similar hospitals in states with regulation. While repealing or reforming CON will not fix all healthcare policy challenges, doing so can help increase affordability and healthcare access."

TrumpRx: When Government Tries to Build a Market

By Jeffrey A. Singer.

"This evening at 7:00 p.m. Eastern Time, President Trump announced the launch of TrumpRx, the government-run direct-to-consumer (DTC) drug purchasing platform. Essentially, TrumpRx will act less as a pharmacy and more as a portal directing patients to manufacturers’ cash-price sales platforms negotiated by the administration. Supporters view the initiative as a way to eliminate intermediaries and offer patients lower prices.

President Trump is correct that DTC sales of prescription drugs can put downward pressure on prices. We’ve seen this dynamic repeatedly when medications transition from prescription-only to over-the-counter status.

As Michael F. Cannon and I discuss in our Cato white paper Drug Reformation, third-party payment arrangements tend to drive up drug prices. When insurers or government programs are paying most of the bill, patients have little incentive to resist high prices. In fact, they often push back when payers try to steer them toward lower-cost drugs or pharmacies because any savings go to the insurer, employer, or government—not to them. Insurers, for their part, know that denying coverage or refusing to pay list prices can cause backlash from beneficiaries who feel entitled to whatever their plan covers.

Coverage decisions by both public and private insurers greatly influence drug pricing. This inflationary cycle is more apparent in the prescription market than in the over-the-counter (OTC) market because insurance typically covers prescription drugs but usually stops covering them once they become OTC. When that happens, consumers pay out of pocket—and prices often decline because purchasing becomes more sensitive to cost.

When consumers control the money, they comparison shop and weigh price against benefit. When deep-pocketed third parties cover most of the cost, price sensitivity diminishes—and producers face less resistance to higher pricing.

The problem isn’t the DTC model. It’s the assumption that the federal government needs to run it. A growing private marketplace already exists, including platforms such as Mark Cuban’s Cost Plus Drug Company, Amazon Pharmacy, and GoodRx, as well as pharmaceutical manufacturers that sell directly to patients through their own websites. 

PhRMA (the Pharmaceutical Research and Manufacturers of America), the trade association representing the country’s pharmaceutical industry, is also getting into the act. It recently announced the launch of Amer​i​c​as​Med​i​cines​.com, a website that connects patients with manufacturers’ direct-purchase programs.

Injecting government into this space risks crowding out private innovation and inviting the familiar problems of political favoritism, coercion, and regulatory corruption. Some lawmakers are raising concerns about conflicts of interest, transparency, and whether the platform’s structure could violate federal anti-kickback rules—especially given its reliance on partnerships with drugmakers and its connections to existing online pharmacy and telehealth fulfillment channels.

If the administration wants to expand direct-to-consumer drug purchasing, the most effective role it can play is not to build a federal platform but to eliminate policy barriers that hinder private actors from competing, innovating, and lowering prices on their own. That involves rolling back regulations that restrict manufacturer-to-patient sales, removing contractual and regulatory obstacles that prevent pharmacies and wholesalers from offering transparent cash prices, easing restrictions on telehealth prescribing tied to online fulfillment, and lowering the tax and compliance penalties that discourage patients from buying medicines outside third-party payment systems.

As Michael F. Cannon has shown in his work on the tax exclusion for employer-sponsored insurance, federal tax policy actively discourages patients from operating outside third-party payment systems by penalizing those who try to control their health care dollars outside employer- and insurer-managed systems.

The lesson of every market touched by third-party payment is the same: when patients control the dollars, prices fall, and value rises. TrumpRx risks moving policy in the opposite direction—substituting political allocation for consumer choice in a space that is only now beginning to function like a real market—one where patients, not payers or politicians, make the purchasing decisions."

Trump’s Pharmaceutical Plan

By Alex Tabarrok. 

"Pharmaceuticals have high fixed costs of R&D and low marginal costs. The first pill costs a billion dollars; the second costs 50 cents. That cost structure makes price discrimination—charging different customers different prices based on willingness to pay—common.

Price discrimination is why poorer countries get lower prices. Not because firms are charitable, but because a high price means poorer countries buy nothing, while any price above marginal cost is still profit. This type of price discrimination is good for poorer countries, good for pharma, and (indirectly) good for the United States: more profits mean more R&D and, over time, more drugs.

The political problem, however, is that Americans look abroad, see lower prices for branded drugs, and conclude that they’re being ripped off. Riding that grievance, Trump has demanded that U.S. prices be no higher than the lowest level paid in other developed countries.

One immediate effect is to help pharma in negotiations abroad: they can now credibly say, “We can’t sell to you at that discount, because you’ll export your price back into the U.S.” But two big issues follow.

First, this won’t lower U.S. prices on current drugs. Firms are already profit-maximizing in the U.S. If they manage to raise prices in France, they don’t then announce, “Great news—now we’ll charge less in America.” The potential upside of the Trump plan isn’t lower prices but higher pharma profits, which strengthens incentives to invest in R&D. If profits rise, we may get more drugs in the long run. But try telling the American voter that higher pharma profits are good.

The second issue is that the plan can backfire.

In our textbook, Modern Principles, Tyler and I discuss almost exactly this scenario: suppose policy effectively forces a single price across countries. Which price do firms choose—the low one abroad or the high one in the U.S.? Since a large share of profits comes from the U.S., they’re likely to choose the high price:

Pfizer CEO Albert Bourla was even more direct, saying it is time for countries such as France to pay more or go without new drugs. If forced to choose between reducing U.S. prices to France’s level or stopping supply to France, Pfizer would choose the latter, Bourla told reporters at a pharma-industry conference.

So the real question is: will other countries pay?

If France tried to force Americans to pay more to subsidize French price controls, U.S. voters would explode. Yet that’s essentially what other countries are being told but in reverse: “You must pay more so Americans can pay less.” Other countries are already stingier than the U.S., and they already bear costs for it—new drugs arrive more slowly abroad than here. Some governments may decide—foolishly, but understandably—that paying U.S.-level prices is politically impossible. If so, they won’t “harmonize upward.” They’ll follow the European way: ration, delay and go without.

In that case, nobody wins. Pharma profits fall, R&D declines, U.S. prices don’t magically drop, and patients abroad get fewer new drugs and worse care. Lose-lose-lose.

We don’t know the equilibrium, but lose-lose-lose is entirely plausible. Switzerland, for example, does not seem willing to pay more:

Yet Switzerland has shown little political willingness to pay more—threatening both the availability of medications in the country and its role as a global leader in developing therapies. Drug prices are the primary driver of the increasing cost of mandatory health coverage, and the topic generates heated debate during the annual reappraisal of insurance rates. “The Swiss cannot and must not pay for price reductions in the USA with their health insurance premiums,” says Elisabeth Baume-Schneider, Switzerland’s home affairs minister.

If many countries respond like Switzerland—and Trump’s unpopularity abroad doesn’t help—the sector ends up less profitable and innovation slows. Voters may feel less “ripped off,” but they’ll be buying that feeling with fewer drugs and sicker bodies."

Notches in the Tax Code

Why you shouldn’t earn that extra $ if your adjusted gross income is $218,000

By David R Henderson. Excerpt:

"The first is to phase it out over some income range. So let’s say it’s a $1,000 tax credit and people filing jointly with income up to $100,000 qualify. But Congress wants to take it away for incomes above $100,000. Then it can reduce the amount of the credit by, say $1 for every $10 of AGI above $100,000. People with income between $100,000 and $110,000 will get some tax credit. People with an AGI of $110,000 or more get a zero credit.

Notice one important implication for marginal tax rates. For people with AGIs between $100,000 and $110,000, there is an addition of an implicit ten percentage points to people’s marginal tax rates. That can have substantial incentive effects.

This phaseout leads to the term “kink.” If you were graphing net of tax income on the vertical axis against income on the horizontal axis, the line would kink and become closer to horizontal at $100,000 in income.

The Dreaded Notch

But let’s say Congress doesn’t want anyone with income above $100,000 to get the $1,000 tax credit. Then someone who has an AGI of $100,000 will lose the usual tax on income plus $1,000 when he gets that first dollar of income after $100,000. So let’s say his normal tax rate, including the state income tax rate and payroll taxes, is 35%. When he makes that next dollar, he loses 35 cents plus $1,000, for a total loss of $1,000.35. His implicit marginal tax rate is 100,035 percent on that dollar.

Why is it called a notch? Now graph after-tax income on the vertical axis and income on the horizontal axis. When income reaches $100,001, after-tax income falls like a rock. Thus the word “notch.”

I learned all this from Tom McCaleb’s memo.

Now to Today

Congress, or, at least, the Congressional staffers who understand taxes, usually try to avoid notches and settle for kinks.

But not always.

A recent news item in the Wall Street Journal titled “The Medicare Charge That’s Taking a Bigger Bite Out of Social Security Checks,” WSJ, January 23, 2026, gives an instance.

The issue is not taxes but Medicare premiums that are “income-related monthly adjustment amounts,” or Irmaa.

But the principal is the same. The news item’s author, Laura Sanders, gives a table that show the extra monthly Medicare premium you’ll pay when your income reaches a certain level.

So, for example, joint filers with an income of $218,000 will pay an Irmaa of 0. But, if the couple makes one additional dollar, their Irmaa rises to $81.20 per month, which is $974.40 per year. Their implicit marginal tax rate due to Irmaa alone, therefore, is 97,440 percent. (Their actual explicit tax rate for that dollar is rounding error.)

Notches.

By the way, here’s an example of an economically literate and generally very careful economic journalist making the case for a huge notch: “McArdle Advocates 376,537.65% Marginal Tax Rate,” EconLog, January 20, 2010."

The Long Poverty Decline: Before and After the War on Poverty

By Jack Salmon of Mercatus.

"A new paper from Richard Burkhauser, Kevin Corinth and colleagues gives us something rare in the poverty literature: clean historical measurement that actually lets us ask the obvious question. Did Lyndon Johnson’s War on Poverty mark a decisive break in America’s fight against destitution? Or did the big decline in poverty mostly happen before the Great Society arrived, driven not by wealth transfers but by market income?

This new paper finds that the War on Poverty changed the how of poverty reduction (more transfers, more dependency), but it didn’t accelerate the how much. In fact, poverty decreased more before 1963 than after. Below I unpack what the paper does, what it proves, and what it should mean for anyone who cares about liberty, work and sustainable anti-poverty policy.

Comparing apples to apples

A perennial problem in poverty debates is apples-to-oranges measurement. Official statistics historically ignore taxes, in-kind benefits, employer-paid health insurance and other nonwage resources. That makes pre-1964 numbers look worse than they were and post-1964 numbers look better, or at least different, in ways that aren’t comparable.

Burkhauser and Corinth close that gap by building a post-tax, post-transfer “full income” measure extending back to 1939 by carefully imputing taxes, transfers, perquisites and (where available) employer health coverage into the 1940, 1950, and 1960 censuses. The authors then link those imputations to the Current Population Survey Annual Social and Economic Supplements (CPS ASEC) series already used in modern full-income work.

In short, they create a consistent income yardstick that runs from 1939 through 2023. With consistent measurement, we can finally compare the periods before and after the War on Poverty with fewer methodological excuses.

The headline numbers

Using a consistent post-tax, post-transfer measure, the authors find:

· Absolute poverty fell from 48.5% in 1939 to 19.5% in 1963—a 29-percentage-point decline in just 24 years (see the blue line in Figure 1 below).

· Poverty fell again from 1963 to 2023, but only by 15.7 percentage points, reaching 3.7% in 2023.

· If the market value of health insurance is included in income measures, the overall 1939 to 2023 decline is even larger, dropping to just 1.6% in 2023.

 

Crucially, when the authors hold the starting poverty rate constant (so the comparison isn’t biased by vastly different baselines), the 24 years immediately before 1963 improved faster than the first 24 years after it. In other words, the greatest era of poverty reduction came before the War on Poverty.

The poverty reduction mechanism

From 1939 to 1959, market income poverty (i.e., a measure that ignores transfers and taxes) fell by roughly 26 percentage points: nearly the entire post-tax, post-transfer decline for working-age adults and children in that era. Only 2-3% of working-age adults depended on government for at least half their income in the pre-1964 years.

By contrast, from roughly 1967 to 2023, market income poverty hardly budged overall (a small, approximately 4-percentage-point reduction), while post-tax, post-transfer poverty fell much more, meaning transfers filled the gap in poverty reduction that market incomes stopped filling. Dependency rates among working-age adults rose to 7-15% in the modern era (with black working-age dependency peaking much higher in some years). In this context, “dependency” means that a majority of household income comes from government transfers rather than earnings or other market income.

Before the War on Poverty, rising wages, employment and the broad postwar expansion of productive opportunity pulled people out of poverty. After the War, transfers (plus slower market income growth for lower incomes) accounted for a much larger share of measured poverty reduction.

The paper also shows that black Americans and children saw massive absolute poverty reductions before 1964. Black poverty fell from 84.1% in 1939 to 50.6% in 1963; black child poverty fell even more sharply. These gains were driven principally by market income growth, urbanization, rising labor force participation and expanding employment opportunities, not by an expanded transfer state.

Relative poverty tells a different story

If we use a relative poverty line (a threshold that rises with median income), progress is far less dramatic. Relative poverty fell modestly before 1963 and has not improved since. That’s important because modern political rhetoric often cares about relative standing, not absolute material deprivation. Transfers can raise absolute resources for the poor, but they do not easily change the structure that creates relative inequality: differences in skills, employment, family structure and productivity.

What this means for policy

1) For lasting poverty reduction, growth beats transfers. The 1939–1963 story is one of rising market incomes across the income distribution. Policies that encourage capital formation, remove barriers to work and widen opportunity will do more to reduce poverty in the long run than adding layers of permanent income replacement.

2) Transfers are not neutral. They help consumption in the short run, but expanded transfers in the post-1963 period raised dependency rates and, at least for a while, coincided with stagnation in market income poverty. That tradeoff matters for social norms, intergenerational mobility and fiscal sustainability.

3) Program design is incredibly important. One government safety-net policy that actually helped spur market incomes was the 1990s welfare-to-work reforms: Market income poverty fell sharply for children and working-age adults in that period. That’s a reminder that safety nets designed to reward work (or to be conditional on labor-force participation) can avoid the dependency trap.

4) Economists should use honest measurements. We need to include full-income accounting, taxes, transfers and employer benefits when evaluating policy. The official poverty rate, while useful in some contexts, is a bad historical comparator by itself.

There’s an old liberal hope, voiced by Lyndon Johnson, that society could “develop and use [people’s] capacities” rather than simply “support people.” The Burkhauser-Corinth paper shows that, historically, economic development did exactly that for millions before the full modern safety net existed. Transfers have been crucial to reduce measurable hardship, but they were not the engine that drove the 20th century’s greatest antipoverty gains.

If the goal is durable escape from poverty, start with the economy: policies that expand jobs, skills and the social institutions that make work pay. Use targeted transfers to smooth shocks, not to substitute for careers. That’s not just a policy preference; it’s what the long history of American poverty reduction appears to tell us."

Related post:

The Level and Trend of Poverty in the United States, 1939-1979. 

By Christine Ross, Sheldon Danziger and Eugene Smolensky. Published in Demography, Vol. 24, No. 4 (Nov., 1987), pp. 587-600.

"Abstract

A detailed record of the number and characteristics of persons in poverty is available since 1959. This paper provides measures for 1939 and 1949 that correspond as closely as possible to the official poverty statistics. Poverty, as officially measured, fell from 40.5 percent of all persons in 1949 to 13.1 percent in 1979, declining most among the elderly and least among female-headed households. We estimate the effects on the aggregate poverty rate of changes between 1940 and 1980 in the distribution of the population by selected demographic characteristics of household heads and by the employment status of head and spouse. Some changes, such as the increasing proportion of two-earner families, were poverty reducing, whereas others were poverty increasing."

See also Poverty in the United States by Gwendolyn Mink & Alice M. O'Connor. Click here to see inside the book. If you click on the first picture, the numbers are clearer. (Hat tips to commenters Jason Wall and Todd Kreider at this post by Scott Sumner Do poverty programs reduce poverty? Here is a table from Mink & O'Connor's book:

 

There Are No Good Reasons To Subsidize Sports Stadiums. Governments Keep Doing It Anyway.

“If we stop funding all sports stadiums tomorrow, then the world wouldn't change hardly at all," says one economist.

Alexandra Stinson of Reason. Excerpts:

""Almost all stadiums that were built at the beginning of the 20th century were private stadiums," economist J.C. Bradbury recently explained to Reason's Eric Boehm. Up until recently, "owners would have been laughed at if they went to the local city council or county commission and asked for money to help them build a stadium," Bradbury adds."

"In a 2022 paper, J.C. Bradbury, along with economists Dennis Coates and Brad Humphreys, noted that "nearly all empirical studies find little to no tangible impacts of sports teams and facilities on local economic activity, and the level of venue subsidies typically provided far exceeds any observed economic benefits." As the Center for Economic Accountability points out: "Sports compete with other local businesses for consumers' entertainment dollars, rather than creating 'economic development' out of thin air.""

"This public spending comes with tradeoffs. "We see these people spending money in and around stadiums, but what we don't see is their foregone spending, that is, they'd be spending it elsewhere in the community," says Bradbury.

It also comes with corruption. "This pervasive lack of transparency in the planning and negotiation process around stadium subsidies also creates an environment where corruption can flourish," writes John C. Mozena, a senior fellow at Reason Foundation, the nonprofit that publishes Reason. In 2023, former Anaheim, California, Mayor Harish Sidhu "pled guilty to four federal felonies while the city was negotiating a stadium deal with the Los Angeles Angels baseball team" and admitted to "pass[ing] inside information to the team's negotiators and attempt[ing] to influence the city's decisions in favor of the Angels in return for an expected $1 million campaign contribution from the team," according to Mozena. This scandal ultimately led to the cancellation of the Angels' new stadium deal."

""If we stop funding all sports stadiums tomorrow, then the world wouldn't change hardly at all," says Bradbury. "Basically, just these wealthy owners would say, 'OK, I guess I gotta fund it myself.""

How Trump Could Actually Fix Housing Market

Washington needs to unlock capital through deregulation

Letter to The WSJ. 

"Allysia Finley is correct, “institutional investors are a distraction from the real causes of high home prices—government policies” (“Scott Bessent and Gavin Newsom Feud Over a Dumb Idea,” Life Science, Jan. 26).

The market badly needs deregulation to unlock capital. Tax regulations have frozen large swaths of our existing housing stock. And state and local land use regulations lock millions of acres of land out of higher and better uses by making it illegal to build starter homes on smaller lots.

The president, Treasury Secretary Bessent and Congress should consider addressing these capital constraints, which are largely responsible for housing unaffordability.

To account for consumer inflation since 1997, Washington should double the current capital gains exclusion caps for housing sales for homeowners age 65 or older. Currently these caps are $250,000 for a single owner and $500,000 for a married couple of any age. Raising the cap for older owners would unlock an estimated 200,000 homes per year for buyers looking to move up in property quality from the roughly three million senior households above the current exclusions.

Congress should exempt from taxation rental income from newly rented spare rooms. This would unlock an estimated 320,000 rooms per year from the 32 million spare rooms in owner-occupied single-family homes.

Exempt, too, from capital gains taxation starter single-family rentals sold to an occupying tenant in good standing with at least 24 months on-time rental history. This would open the door to homeownership for an estimated 90,000 families per year from the six million tenants living in starter single-family rentals.

Congress should incentivize the building of more starter homes by offering states a small lot bounty program. This would create 200,000 more starter homes per year.

The government should tax profits from newly built, for-sale developments as a long-term capital gain, not income. This would change the tax code from favoring rentals to neutral relative to for-sale homes.

Legalize mortgage prepayment penalties. This is a straightforward way to lower upfront rates by 0.40% to 0.50%—raising home ownership.

What housing markets need is for Washington to unlock capital through deregulation, thereby allowing markets to do what they do best—create and support economic expansion and prosperity.

Ed Pinto

Senior fellow and co-director

AEI Housing Center

Washington"