GLP-1 Weight-Loss Drugs Keep Getting Cheaper: Thank Competition

By Gale Pooley of Cato. Excerpt:

"When Wegovy’s one-of-a-kind treatment debuted in 2021, a month’s supply cost $1,349. At the time, blue-collar employees were earning an average of $26.12 an hour, putting the time-price at 51.65 hours per month, or more than six days’ wages.

Employer insurance plans balked, and the government’s health programs were barred from covering weight-loss drugs unless there was a specified medical complication. More than half of American adults (137 million) meet the medical criteria to benefit from the FDA’s approved uses of GLP-1s: the cost of covering them all is enormous.

Even with mostly cash payers, demand outstripped supply, putting Wegovy on the FDA’s drug-shortage list. With that designation came a legal path for pharmacies to compound the product themselves, and offer lower-priced alternatives through telehealth and cash-pay programs.

Two years later, a second major company gained FDA approval for an injectable drug for chronic weight management. Eli Lilly’s Zepbound is pharmacologically different, but its effects are similar and it competes for the same patients and insurer budgets. The list price undercut Wegovy, at $1,060.

Here’s where things really get interesting.

Eli Lilly knew few of the potential beneficiaries of its medication would qualify for the medication through their insurers. So it offered a special $550 cash price for patients whose insurance didn’t cover GLP-1s. In 2024, it launched LillyDirect: a direct-to-consumer, cash-pay distribution for its competing product. Direct-to-consumer cash-pay pathways allowed patients to bypass asking “which drug does my insurance cover?” and forced manufacturers to compete in a more retail-like environment.

And compete they did. Novo Nordisk introduced its own cash-pay option for Wegovy — NovoCare. Its lowest dose was priced at $499/month. In April 2025, with the FDA’s shortage resolved and still facing innumerable gray-market and privately compounded copies of semaglutide, Wegovy’s price was reduced again, to make the authentic brand more price-competitive for patients, who Novo knew could easily switch. Standard self-pay injectable Wegovy dropped to $349/month, with a $199/month introductory offer.

In response, Eli Lilly cut Zepbound’s cash pricing, offering the typical maintenance dose for $299 by December 2025. Prescribers and patients became more comfortable switching between major branded therapies that filled essentially the same medical need. That increased the price pressure on drugmakers who were accustomed to negotiating with insurer formularies and formal rebate systems.

Almost immediately after Zepbound’s price reduction, the FDA approved an oral formulation of Wegovy. Clinical trials show the once-a-day pill is almost as effective as the injectable, and the pill form expands the available market to include patients unwilling to self-inject GLP-1s.

Today, the oral maintenance dose of brand-name Wegovy costs $299 a month for self-paying patients. Self-pay pricing for injectable Wegovy settled at $349. Fifty percent of sales of both Wegovy and Zepbound are direct exchanges with cash-paying patients.

Our “average” blue-collar worker from 2021 is now earning $32.53/hour. The time price of his monthly weight-loss treatment has fallen by 75 to 80 percent.

Over the same period, according to BLS reports, hospital services have risen 26.6 percent, and medical care prices, 12.9 percent. Overall consumer inflation (CPI) has increased 22.5 percent."

The Stealth Tax in Proposition 40

By David R Henderson. Excerpts:

"California’s Proposition 40 would impose a one-time tax of 5% on people with a net worth of $1 billion or more."

":The initiative’s Section 50310 states: “The Legislature may amend the 2026 Billionaire Tax Act, by statute passed in each house of the Legislature . . . [with] two-thirds of the membership concurring, if the statute is consistent with and furthers the purposes of the 2026 Billionaire Tax Act.”"

"California’s heavily Democratic Legislature could use the law to increase taxes on people who aren’t close to being multimillionaires—let alone billionaires."

"Most personal property is taxable at the same rate as real property (1%), while taxes on stocks and bonds are capped at 0.4%. The authors of Proposition 40 apparently understood this and wrote Section 50310 to supersede Article XIII [of California’s Constitution]." 

"After the 16th Amendment authorized the federal income tax, it started small. In 1913 the tax rate on married couples filing jointly was 1% on income up to $20,000 (equivalent to some $674,000 today). The top rate was 7%, on income above $500,000 ($16.8 million today). Five years later, in the midst of World War I, the rate was 6% on income up to $4,000 and 77% on incomes over $1 million."

In São Paulo entrenched interests lost out due to less restrictive housing regulations; prices went down & new households benefited

See Who benefits from upzoning? A 2016 zoning reform in São Paulo increased the ability of developers to supply housing units by lifting limitations on permitted densification on a block-by-block basis by Tyler Smith of the AEA.

It looks like the eighborhoods receiving the largest reductions in regulations saw the biggest increase in supply and the biggest decrease in prices. See the text in red near the end.

"The share of the world's population living in cities rose from 33 percent in 1960 to 56 percent in 2019 and is projected to reach 68 percent by 2050, with much of that growth concentrated in developing countries. Whether cities can provide sufficient housing for those future residents depends on local land use rules, which determine where and how developers can build.

But measuring the effects of these rules has proven difficult. Data limitations, complexity, and endogeneity have stymied efforts to estimate the value of zoning reforms.

In a paper in the American Economic Journal: Economic Policy, authors Santosh Anagol, Fernando Ferreira, and Jonah Rexer  examine a 2016 zoning reform in São Paulo and related detailed administrative data to assess the impact of zoning policies. 

"The really hard thing about studying zoning is getting fine-grained information on the actual rules that are enforced on the ground," Anagol told the AEA in an interview.

The central regulatory parameter for land use in São Paulo is the built area ratio (BAR)—floor space divided by lot area. The 2016 zoning reform stripped the city's 32 neighborhood governments of the power to set the maximum BAR, assigning new values block by block with the aim of densifying corridors served by metro, rail, and bus lines. 

Across roughly 45,000 blocks, the average maximum BAR rose from 1.55 to 2.09, permitting an average of 36 percent more construction on a given parcel. However, the distribution of reforms was uneven. Forty-five percent of blocks received an increase of 1 BAR or more while nearly half saw no change or a reduction.

The authors exploited this block-level variation using a boundary discontinuity design. Blocks whose maximum BAR increased were used as treatment groups, and the nearest blocks where the maximum BAR held constant or fell were used as controls.

The authors supplemented the building permit records with property tax records, real estate listings from 2019 to 2022, formal sector wage data, and a 2017 commuting survey of 24,800 households. They then estimated an equilibrium model in which developers choose how much to build, and households choose among 329 commuting zones, based on price, commuting time, access to high-paying jobs, building age, density, and neighborhood characteristics.

According to the authors’ estimates, developers responded quickly to zoning reforms. A 1.4-point increase in maximum BAR raised multifamily permit filings by roughly 66 percent relative to nearby control blocks, with differences emerging about a year after passage of the reform and growing thereafter. Notably, single-family permits were unaffected, consistent with BAR easing constraints primarily on larger structures. 

Between 2019 and 2022, listings for sale grew about 10 percent faster in treated blocks. At the commuting-zone level, a 1-point BAR increase was associated with 10.9 percent higher listings growth and a 5.7 percentage-point reduction in prices.

By simulating the model for ten years following the reform, the authors estimated that the reform added roughly 39,000 units—a 1.6 percent increase in the housing stock—and reduced average prices by 0.4 percent. The authors suggest that this modest average can be explained by BAR rising only 0.54 overall. But neighborhoods receiving the largest increases saw up to 17.5 percent more units and 3 percent lower prices. 

"While in the long run it may bring all prices down, the welfare gains, at least initially, may be concentrated amongst higher-income, higher-educated groups," Anagol said. 

The reform produced a nominal wealth transfer of 11.33 billion reais, or 1.62 percent of city GDP, from current homeowners and landlords to future real estate owners. The size of the transfer suggests that it may generate political repercussions. 

"The entrenched interests lose out from these housing reforms when prices go down, whereas new households benefit," Anagol said.

The work shows that targeted upzoning produces real supply responses and localized affordability gains. But the concentration of losses among existing owners may help explain why reforms of this kind remain difficult to pass.

♦

“Estimating the Economic Value of Zoning Reform” appears in the August 2026 issue of the American Economic Journal: Economic Policy."

Wealth of Generations: Update Through the First Half of 2026

By Jeremy Horpedahl. 

"It’s been a while since I updated my generational wealth chart, and we now have estimates through the 2nd quarter of 2026, so here’s the latest chart:

 

Wealth for younger Americans continues to grow substantially, but let me make two caveats:

1. Yes, I know median data is better. I’m writing a book that uses median wealth data! But the latest median wealth data from the Fed’s SCF is currently only available through 2022, so it’s not super relevant to current conversations. We should have the 2025 data soon.
2. Because of the way the data in my chart is produced in the Fed’s DFA, it groups everyone under age 45 together. That’s a mighty big group, and it because it encompasses both Millennials and a lot of Gen Z, it makes it hard to directly compare to earlier generations.

So, until we have 2025 median wealth data, and until the Fed’s DFA starts breaking out Millennials and Gen Z, here is my current best compromise chart:

 

In Figure 2, I have used the Fed DFA data for age groups, which are still pretty large groups, but you can consistently compare them over time. The average wealth level of both the 18-39 group and the 40-54 group have seen substantial gains. In fact, the gains for younger cohorts have been even better than middle-aged Americans, though both saw substantial gains.

And this chart shouldn’t be affected by the lack of household formation among some younger Americans: I am using the full population as the denominator, so if anything, this will understate growth rates. Even so, the growth rate from the depths of Financial Crisis in 2010 have been substantial: 228 percent growth from 2010 to 2026 for ages 18-39. The growth rate for ages 40-54 was less dramatic, though they also didn’t experience as large of a slump from 2007-2010.

While we can always hope and work towards growth rates being better, average wealth for Americans of working age is currently at record highs, having fully recovered from both the Financial Crisis and the inflation slump of 2022."

Federal Film Tax Credit Could Cost Up to $50 Billion

By Adam N. Michel of Cato.

"After President Trump’s endorsement, a bipartisan, bicameral group of lawmakers recently introduced the Motion Picture, Television, and Entertainment Revitalization Act. The bill would create the first direct federal subsidy for Hollywood movie studios through a tax credit covering 20 to 30 percent of what film and TV productions pay their workers. 

Dozens of states and countries have experimented with subsidies for the film industry. The overwhelming evidence is that the subsidies mostly don’t create new production, don’t meaningfully increase jobs or wages, don’t build an industry that survives without the subsidy, and are fiscally costly. 

Using the film industry’s own analysis of the newly proposed federal film tax credit, I estimate the program could cost between $33 billion and $49 billion over ten years. That works out to between $54,000 and $81,000 for each additional industry job. Under less optimistic assumptions, the cost per job could rise to as much as $125,000. 

State and International Film Subsidies Don’t Work 

Today, 39 US states, Washington, DC, and Puerto Rico offer film incentives. Patrick Button investigated these state programs and found that adopting one increased the number of television series in the state by at most 1.5 series. He found no meaningful effect on feature films, employment, wages, the number of businesses in the film industry, or businesses in related industries. Professor Michael Thom finds that, between 1998 and 2013, state film incentives did not raise the industry’s share of the economy or its concentration in the state.

Another journal article found that film incentives can attract movies but found no strong evidence that they create a permanent movie industry in the state. A study of California’s film credit lottery similarly found that receiving a subsidy increased the probability that a film would be made in the state by 16 percentage points. However, the California Legislative Analyst’s Office concluded that even with the increase in production activity, “there is weak evidence that expanding the tax credit would benefit California’s economy as a whole.” It concluded that expanding the credit only makes sense if preserving Hollywood’s market share is treated as an end in itself. 

The conclusion in California matches similar academic research. John Charles Bradbury looks at US state economies over a period when film incentives were being adopted, suspended, and repealed. He finds no evidence that film incentives have a positive impact on state economies. A Canadian cost-benefit analysis similarly concludes that, despite increased employment in the film industry, the film incentives made Canadians poorer overall, as the increase comes at the expense of economic activity in other sectors of the economy. In a similar cost-benefit analysis, Ireland’s Department of Finance estimated its film incentive resulted in a €72 million net cost to Irish society in 2016. 

The most favorable academic work examines the early film incentive-adopting states of Louisiana, New Mexico, and Rhode Island, finding real gains in wages and employment. However, the same research finds that film employment and wages fell in states that later repealed or capped their programs. This shows that tax credits don’t build self-sustaining industries; they build industries that exist only to collect taxpayer subsidies. 

Louisiana provides anecdotal evidence of how fragile a subsidized industry is. In 2015, the state capped how much in credits it would pay out each year and temporarily suspended its buyback of credits. The state’s economic development agency data indicate that film business fell by about 75 percent. Again, amid uncertainty over repeal and reform in 2024 and 2025, production in the state fell to a single active project in May 2025. It rebounded after lawmakers increased the subsidy amount that any single production can claim, showing the fragility, high costs, and dependence film subsidies create. 

The positive assessments of film incentives almost universally come from industry-funded studies that rely on unrealistic assumptions. For example, a series of Ernst & Young studies concluded that film subsidies pay for themselves, returning as much as $1.90 in new revenue for every dollar subsidized. In a review of 29 assessments by independent government agencies in 23 states, Professor Thom found all but one cost the state more than it returned in new revenue. 

Federal Film Tax Credit Could Cost Billions a Year

The Hollywood subsidy bill is led by Sen. Tim Scott (R‑SC) and Rep. Nathaniel Moran (R‑TX) and joined by Adam Schiff (D‑CA) and Linda T. Sánchez (D‑CA), among others. It would create a new federal tax credit for 20 percent of a qualifying production’s labor costs. The credit could rise to as much as 30 percent with bonuses for filming in rural opportunity zones or disaster areas, for independent productions, for spending across many states, and for increasing the number of domestic productions. The credit is transferable, so a production with little or no federal tax liability can sell it for cash to a company that has federal tax liability to offset.

Relying on a study commissioned by the Motion Picture Association (MPA), supporters claim the new federal film subsidy could create nearly 145,000 new jobs a year and $250 billion in new economic value added between 2027 and 2035. 

Using the study’s own assumptions, I estimate the new federal film credit could cost between $3.3 billion and $4.9 billion per year, or up to $49 billion over a decade. The low end assumes every project claims the lower 20 percent credit, and the high end assumes all qualifying compensation receives the maximum 30 percent credit. The bill has no per-production cap, no limit on star salaries, and no cap on total cost. 

At the high end, this new subsidy would spend as much subsidizing Hollywood movie studios as the National Park Service spends each year, at about $5 billion in annual budget authority.

The MPA estimate rests on several strong assumptions, including that without a federal subsidy, the US share of global TV and movie film production will fall below 30 percent, a decline of 13 percentage points for TV and 9 percentage points for movies, by 2035. With the credit, it assumes the US share rises to 65 percent, a significant change in both level and trend.

The study’s model also assumes no supply constraints for labor or other inputs and does not account for economic activity displaced elsewhere in the economy. To the extent the subsidy actually creates new wages, higher federal income and payroll tax receipts would offset some of the credit’s gross cost. But the empirical research surveyed above on film credits suggests that much of the apparent job creation reflects workers shifting between industries rather than net new employment. Excluding those assumed economy-wide spillovers, the study’s own model implies about 60,600 additional full-time-equivalent annual jobs directly in the film industry. 

The $3.3 billion to $4.9 billion in new subsidies works out to between $54,000 and $81,000 a year for each additional full-time film industry job the MPA claims would be created. Because the credit would be primarily used by productions already filming in the United States, the fewer new jobs created, the more costly each additional job becomes, even as the total cost of the program falls. If the credit brings in half the new jobs and production expenditures the MPA assumes, the cost per additional job rises to about $83,000 on the low end and $125,000 on the high end. This range is consistent with a 2016 Massachusetts Department of Revenue evaluation finding that the state’s film incentive costs $109,762 per new Massachusetts resident job created. 

Film industry jobs are also not low-wage jobs. The MPA study estimates that film and television production jobs pay about 50 percent more than the comparable national average. Higher-wage and higher-skill workers also tend to have more employment options, strengthening the empirical finding above that subsidies largely pull these workers away from unsubsidized productions and other industries.

Hollywood’s problem is not a shortage of subsidies. Thirty-nine states and dozens of countries already pay studios to little effect. A new federal film tax credit would primarily put taxpayers on the hook for multimillion-dollar Hollywood studio productions that would have largely been filmed here anyway. Congress should protect taxpayers from subsidizing Hollywood by rejecting the federal film credit, and states should repeal their programs, too. If filming in the United States costs too much, the answer is lower taxes and easier regulations for all businesses, not billions of dollars in subsidies for a single industry. 

Notes: The labor cost estimate follows the report’s assumption that 53 percent of the $277.5 billion in total industry spending is labor costs and assumes that all domestic labor income qualifies for the credit. The figure for industry jobs created applies the 2.33 employment multiplier and 0.919 full-time equivalent conversion ratio to the study’s 153,700 reported total annual jobs."

Single-Family Homes Are the Largest Source of Affordable Rentals in the United States

By Arthur Gailes & Edward J. Pinto of AEI.

"Myth: Affordable housing is popularly conceived as primarily consisting of subsidies for people living in apartment buildings.

Market Reality: Nationwide, most of the least expensive rental housing comes from single-family 1-4 and mobile/manufactured homes.

Fifty-five percent of the people paying these least expensive rents live in single-family homes (1-4 units) or mobile homes, not in multifamily buildings.

That distinction between homes and people matters. Single-family homes have far more bedrooms on average—about 3.2 nationally, compared with 1.4 for large multifamily buildings. So even where multifamily buildings provide a large share of low-rent units, single-family homes can still house more of the people benefiting from least expensive rents.

This story holds true throughout the country. In a high-priced metro like Los Angeles, most of the least expensive rental units (54%) are multifamily. But most people (53%) in these units live in single-family 1-4 and mobile/manufactured homes.

The reason is mostly bedrooms. People, when they get together to form households, disproportionately live in single-family homes because they have the most affordable space. Most people live in homes with at least two bedrooms, and that is just as true for people with low incomes as it is for people with median or high incomes."

"Across most metros in the country, once we look at the population living in the least expensive rental units rather than simply counting rental units, single-family homes provide most of that housing (green).

The counterintuitive result is that not only do single-family 1-4 and mobile homes constitute the largest source of affordable rentals in the United States, but policymakers and voters should be skeptical about narratives that treat single-family rentals as a source of housing-cost problems. This includes animosity towards institutional investors. The homes that are most affordable for families to rent are disproportionately single-family 1-4 and mobile/manufactured homes."

Does Costco Cause Cancer?

By Alex Tabarrok.

"In December 2025, researchers led by Yazan Alwadi at Harvard’s T.H. Chan School of Public Health published a paper in Environmental Health that claimed to find that cancer incidence increased for people living closer to nuclear power plants in Massachusetts. In March, the same researchers published an expanded nationwide study claiming a similar result—this time looking at cancer mortality rates, rather than incidence—in Nature Communications. This was followed by a paper in the Journal of Exposure Science & Environmental Epidemiology that looked at associations of lung, breast, and colon cancers. Most recently, a study of total mortality, not just cancer, was published in the European journal Environmental Epidemiology.

The problem? Using the same methods pretty much everything causes cancer. An amazing takedown from Deric Tilson and Adam Stein:

For instance, living near a private four-year university is associated with a 15-fold increase in cancer mortality when compared to living near a nuclear power plant.

Costco has the largest effect of all the locations we have tested. Over 2.2 million cancer deaths can be attributed to Costco; that’s more than 20% of all cancer deaths between 2000 and 2018. Hot dogs, bulk spices, and reasonably priced clothes come with a cost.

What went wrong?

[The authors] chose nuclear power plants because a story could be built around that framework. When the researchers got positive results across our nation’s nuclear power plants, they didn’t check what their shiny new methodology would do using other landmarks. This is their pitfall: by taking the easy way out—getting results and making up a story around those results without double-checking their method—the authors could have no idea that what they were actually capturing was the methodology itself.

…The attributable number of deaths from this methodology is probably zero, but the attributable number of bad papers is at least four."

Here are two more excerpts from the Deric Tilson and Adam Stein paper:

"But proximity is not exposure. We have methods of measuring exposure for nuclear power plant workers. While the radiation exposure of nuclear workers will always be greater than or equal to that received by the surrounding public, most of the closely monitored US nuclear workforce receive no measurable annual dose. When workers are exposed to radiation, the average dose received is only 2 percent of the occupational limit. If operators and workers who are on-site at nuclear power plants receive an annual dose between zero and one-fiftieth of the occupational limit, how is it possible that residents 5, 10, 25, 50, 120, or 200 kilometers away would receive any measurable dose from the same plant?"

"In our original rebuttal, we remarked that the studies can’t prove their assertions because they lacked proper control. It seems they also neglected to do any placebo testing. They chose nuclear power plants because a story could be built around that framework. When the researchers got positive results across our nation’s nuclear power plants, they didn’t check what their shiny new methodology would do using other landmarks. This is their pitfall: by taking the easy way out—getting results and making up a story around those results without double-checking their method—the authors could have no idea that what they were actually capturing was the methodology itself. In our replication and expansion, we have shown that choosing a number of sites and landmarks from capitals to warehouse stores can all yield a positive result without any reasonable pathway for American citizens to become exposed to radiation, develop cancer, and then die. The variations are random noise, all sloping in a positive direction. Even when using randomized outcomes, the methodology outputs positive results.

The papers by Alwadi et al. don’t show a novel mechanism by which nuclear power plants meaningfully contribute to cancer mortality; instead, they show an association of data points across 400 km-wide circles. 48,519 square miles, about the size of Mississippi, is a massive area to claim any kind of exposure. Most studies measuring distance-based exposures look at much smaller distances, such as under 10 km (an area of 113 square miles). The methodology is blind, which can be a feature in research areas needing to avoid bias, but in this case it is a bug; it doesn’t understand radiation exposure or dose. All it understands are its inputs: coordinates, proximity, covariates, and cancer deaths. From these, it can give you a number and even a positive result, but it cannot explain why that result exists. It is still an ecological study, one with a sophisticated statistical technique, but not a very useful one."