Paragon PIC

They say a picture is worth a thousand words. Well, data visualization is worth at least that much. Paragon Pic makes health policy easier to see and understand by demonstrating with images what researchers would normally try to explain with a lecture or text. Check for a new figure every week.

They say a picture is worth a thousand words. Well, data visualization is worth at least that much. Check for a new figure every week.

October 6, 2026

In 2025, Over 90% of SDP Spending in TX and AZ Scheduled for Non-Rural Hospitals

Hospital lobbyists argued that the reforms to provider taxes and state-directed payments (SDPs) in the One Big Beautiful Bill (OBBB) would decimate rural providers. For example, the National Rural Health Association (NRHA), in its comment on CMS’s proposed rule implementing the OBBB’s SDP provision, stated it would “disproportionately harm rural hospitals and jeopardize Medicaid enrollee access to care in rural communities.” This claim has often been repeated in the media without evidence.

In this analysis, we examine publicly available provider-specific scheduled SDP payments in Texas and Arizona in 2025 and show that in both states, more than 90 percent of SDP payments were for non-rural providers. This is consistent with a recent study by the Assistant Secretary for Planning and Evaluation (ASPE) at the US Department of Health and Human Services, which found that only 5 percent of SDP funds in 2025 went to rural inpatient hospitals.

Using the 2025 Federal Office of Rural Health Policy (FORHP) hospital-site designation, rural providers account for 2 percent of spending under Texas’s principal hospital SDP, CHIRP, and 8.3 percent under Arizona’s principal hospital SDP, HEALTHII. The figures describe what the hospitals were scheduled to receive under these SDPs. They do not reflect final cash receipts, all hospital SDPs in either state, or hospitals’ net gain accounting for their financing contributions through Medicaid provider taxes or IGTs. Texas and Arizona use different annual windows. Both Pics use the same FORHP definition.
We also examined whether those numbers changed under different definitions of rurality. Texas’s rural share did not change significantly under different definitions (2.3 percent and 2 percent under Texas’s and the U.S. Department of Agriculture’s (USDA) respective definitions). However, Arizona’s rural share varied more under different definitions (19 percent and 4.4 percent under Arizona’s and the USDA’s respective definitions).
We also examined spending per available hospital bed to see if differences in hospital capacity could explain the heavy skew toward non-rural providers. We found that Texas still spent more per available hospital bed on non-rural providers than rural providers, while Arizona spent more per bed on rural providers than non-rural providers.

Methodology

We selected Texas and Arizona because each publishes a hospital-level fiscal year 2025 schedule that can be checked against individual providers. They are illustrative cases, not a representative sample of states. AI assisted the data preparation and chart production. We manually checked a sample of its calculations and found no errors.

Texas: Rural Providers Account for 3.2% of Scheduled Hospital SDP Dollars in 2025

Texas SDPs

For Texas, we examined each hospital’s scheduled payments under the state’s Comprehensive Hospital Increase Reimbursement Program (CHIRP), which Texas describes as its “statewide program that increases Medicaid payments to hospitals for inpatient and outpatient services provided to persons with Medicaid.” CHIRP has three components: Uniform Hospital Rate Increase Payment (UHRIP), Average Commercial Incentive Award (ACIA), and Alternate Participating Hospital Reimbursement for Improving Quality Award (APHRIQA). The state posts projected payments for all three components in its Final Payment (SFY2025) Scorecard under “APHRIQA Scorecards.”

Between September 1, 2024 and August 31, 2025, CHIRP was scheduled to spend $6.05 billion (out of $6.48 billion it was approved to spend). We calculated this total by combining the projected UHRIP and ACIA amounts with updated figures for the APHRIQA. The updated APHRIQA figures include interim payments reported as having been received and final payments or recoveries calculated after hospitals’ quality results were known. These figures replace the preliminary APHRIQA estimate to avoid double counting. Because UHRIP and ACIA remain projections, the combined total is not a final accounting of actual payments. Texas has other directed payment programs, but they serve different provider types or lack comparable provider-level spending data, so we excluded them from the analysis.

Arizona: Rural Providers Account for 8.3% of Scheduled Hospital SDP Dollars in 2025

Arizona SDPs

Our data for Arizona covered the Hospital Enhanced Access Leading to Health Improvements Initiative (HEALTHII) model for October 1, 2024, through September 30, 2025. Across 117 providers, it lists $3.031 billion in SDP payments (out of $3.09 billion it was approved to spend). Arizona used its 2023 hospital claims to estimate its base payments in 2025, adding an allowance for claims not yet reported but assuming no growth in hospital use between 2023 and 2025. Provider amounts could change with actual FY2025 managed-care utilization and final reconciliation.

Arizona has other less costly SDPs that were not included in our analysis because there were no publicly available provider-level estimates that would have permitted a rural/urban comparison. These programs seem targeted to non-rural hospitals, including some examples that seemed to target single providers located in the urban area of Phoenix.

The Safety Net Services Initiative (SNSI) is an SDP that was approved by CMS to spend $388.5 million on “the eligible public safety net hospital established by the state for inpatient and outpatient hospital services.” That hospital is Valleywise, Maricopa County Special Health Care District, located in the city of Phoenix.
Another SDP that seems directed towards just one provider is the Pediatric Services Initiative (PSI), which was approved to provide $59.12 million “for inpatient and outpatient hospital services at freestanding children’s hospitals with more than 100 licensed pediatric beds.” It seems directed to Phoenix Children’s Hospital (PCH). The City of Phoenix’s summary of its June 18, 2025, Formal Meeting described PSI as “a partnership between the City of Phoenix, AHCCCS and PCH” and that its purpose “is to help provide financial sustainability to PCH as one of the country’s largest and busiest pediatric systems.”

Another major SDP that we excluded is Arizona’s Access to Professional Services Initiative (APSI), which was approved to provide $310.4 million in payments to hospitals based on professional services. We excluded this SDP from our analysis because the state did not publicly disclose provider-specific payments. However, one AHCCCS document from before 2022 listed 11 hospitals as APSI recipients. While it is unclear whether those providers have changed, none of those 11 hospitals are considered rural under any of the definitions we used.

Rural vs. Non-Rural Classification

Both analyses classify each hospital under the 2025 Federal Office of Rural Health Policy (FORHP) rural designation, as compiled by UNC Sheps Center. For any unmatched sites, we used the Health Resources and Services Administration’s (HRSA) rural-area point lookup to find the correct FORHP classification.

In addition to the FORHP, we also considered two other rural vs. non-rural definitions. First, we looked at the states’ own designations (Texas CHIRP’s own provider classes and a limited Arizona AHCCCS APR-DRG/RHIF category schedule). Second, we used the U.S. Department of Agriculture’s (USDA) 2023 county metro/non-metro codes.

Findings

Under FORHP 2025, 136 Texas CHIRP rows classified rural account for $191.6 million of $6.050 billion (3.2 percent). Seventeen Arizona HEALTHII rows classified rural account for $252.7 million of $3.031 billion (8.3 percent). The other 247 Texas and 100 Arizona rows are non-rural under this site designation. All 383 Texas and 117 Arizona provider rows, including zero-dollar rows, remain in the denominators. None has an unknown FORHP class.

Numbers Using Different Definitions of Rural

We also examined how the results changed under different definitions of rural. Texas’s results changed little, while Arizona’s varied more substantially. Texas’s rural designation under CHIRP yields 2.3 percent. Using USDA’s county metro/non-metro classification, non-metro hospitals account for 2.2 percent of selected dollars.

While Arizona’s rural share under USDA (4.4 percent) was smaller than its share using the FORHP definition, the state’s own definitions showed a much higher percentage of rural providers. Under Arizona’s own designations, contained in a separate limited AHCCCS APR-DRG/RHIF schedule, rural hospitals account for $576.6 million (19.0 percent) while $117.8 million (3.9 percent) went to hospitals with unknown state-category status. The schedule’s “All Other” category is not proof of a non-rural location.

These sensitivity shares all use the full state selected-dollar totals, including unknown classifications.

Adjustment for Hospital Size

One possible explanation for the patterns we observe is that non-rural providers receive larger SDP payments because they serve more patients. To account for capacity differences, we also examined spending per available hospital bed. These numbers were taken from hospitals’ Medicare cost reports only for providers that had full-year Medicare cost reports and known geography (about 99.6 percent of selected Texas dollars and 98.3 percent of Arizona dollars).

The denominator is all available beds, not Medicaid patient volume. Also, some cost-report periods do not coincide with the SDP year.

We found that even adjusting for hospital beds, Texas SDP spending was still significantly skewed towards non-rural providers, while the opposite was true in Arizona. In the separate matched comparison using USDA non-metro/metro county status, rural dollars per average available bed were about $33,300 versus $102,600 for metro Texas hospitals, but about $504,800 versus $181,500 in Arizona.

These findings only describe selected gross provider allocations before provider taxes or intergovernmental financing. They do not show whether provider taxes help or hurt any individual hospital, whether all rural hospitals participate, or how the OBBB’s reforms will change access or employment.

CONCLUSION

These findings are relevant for both the SDP reforms and the changes to provider taxes in the OBBB. As Paragon has previously written, states often combine SDPs with provider taxes. They first use provider taxes to raise funds from providers, then use SDPs to pay providers back while triggering larger infusions of federal Medicaid money. Both the state and providers often net even greater funds (see this video explaining the financing mechanism).

The fact that states skew their resources towards non-rural providers adds another piece to the puzzle of why states with provider taxes experience decreased rural hospital employment, as research by Liam Sigaud and Niklas Kleinworth found. That decrease in provider-tax states happened even though the authors found no corresponding decline among urban hospitals. Our finding provides one possible explanation for this seeming contradiction: states direct nearly all their Medicaid SDP dollars towards non-rural hospitals.

Ultimately, Congress and the Administration are right to focus their efforts to reform these payments and stop diverting resources from the most vulnerable populations.

6Aw Cms Cbo Sdp Federal Before After A0Wuu000005Vb9Dyaa

September 14, 2026

Higher SDP Savings from OBBB Reforms Reflect a Higher Spending Baseline

Hospitals and their lobbyists are falsely implying that the Centers for Medicare and Medicaid Services (CMS) is going beyond its statutory authority by imposing steeper Medicaid payment reductions than Congress intended.

In 2025, Congress passed limits on state-directed payments (SDPs), through which states override negotiated payment rates and direct Medicaid managed care plans to pay providers specified amounts. These arrangements have pushed Medicaid payments well above what Medicare pays for the same services. In response to this problem, the OBBB ties payment limits for certain services to Medicare rates (and 110 percent of Medicare rates in non-Medicaid expansion states). Though the bill temporarily allows some existing arrangements to remain above those limits, these grandfathered arrangements must begin phasing down in 2028.

As we argued in our comment letter supporting CMS’s rule implementing these reforms, they will ultimately benefit seniors and incentivize lower costs. These reforms are also projected to save taxpayers hundreds of billions of dollars. Herein lies the controversy.

When the OBBB was being debated, the Congressional Budget Office (CBO) estimated that the SDP limits would reduce federal Medicaid spending by roughly $149 billion over ten years. However, in May 2026, CMS’s Office of the Actuary estimated that CMS’s proposed rule would reduce federal Medicaid spending by $510 billion over ten years. Hospitals have pointed to that difference as evidence that CMS is going beyond the statute.

Hospitals’ argument is wrong.

The different estimates primarily reflect that CBO significantly underestimated the federal cost of SDPs before the reform and did not anticipate their explosive growth from 2024 to 2026.

CBO and CMS OACT measured savings against substantially different spending projections. CBO used its January 2025 Medicaid managed care baseline, while CMS’s actuaries used newer budget projections incorporating SDP data through December 2025. In 2025, states submitted new SDPs that raised the baseline. The newer information available to CMS showed massive SDP growth. OACT estimates total SDP spending at $143.8 billion in fiscal year (FY) 2025 and projects $167.4 billion in FY 2026 absent the new restrictions. CMS’s higher estimate of spending without the OBBB, therefore, better reflects reality.

This PIC shows a $444 billion gap over fiscal years 2024–2034 between CMS OACT’s 2026 baseline projection—the spending trajectory without the reforms—and CBO’s 2025 projection. It also shows projected SDP spending after the reforms. Those post-OBBB projections are virtually identical. That is strong evidence that the underlying policy is the same and that the larger savings estimate results from CMS having more up-to-date information about what SDP spending would otherwise have been. After applying each agency’s estimated savings from the OBBB, projected annual SDP spending is nearly the same by 2034.

CMS’s proposed also extended its reforms beyond the OBBB by also applying limits to fee-for-service supplemental payments and some additional SDP categories not specifically mentioned by the bill. Those extensions were small, only accounting for about $5 billion in additional savings from 2029 to 2035. More importantly, CMS had clear legal authority to enact these extensions given its long-standing statutory authority under the Social Security Act to oversee fee-for-service supplemental payments as well as SDPs.

Ultimately, hospital lobbyists have causation backwards. The wide gap between CBO’s earlier score and OACT’s estimate is not evidence that CMS departed from the OBBB statute. Rather, it is evidence of just how rapidly SDPs grew while policymakers were debating and implementing reforms—and proof that those limits were essential to restore fiscal integrity to Medicaid and reduce rampant corporate welfare in the program.

4Aw Nearly 500 More July 25 26 A0Wuu000005Tc9Pyaa

September 1, 2026

Nearly 500 More Health Care Facilities Serving Rural and Low-Income Americans are Operating than Before the OBBB

In the weeks surrounding the passage of the One Big Beautiful Bill (OBBB) in July 2025, some critics of the law predicted an immediate crisis for health care providers who serve low-income and rural patients. The National Association of Community Health Centers warned that “[m]any [community health centers] are already making tough decisions about cutting back services, reducing essential staff, and closing sites” and claimed that the OBBB would worsen these trends.

Last fall, House Minority Leader Hakeem Jeffries asserted that “hospitals and nursing homes and community-based health centers are closing all throughout America,” blaming the OBBB.

Some media reporting has reinforced this narrative with misleading and incomplete information. A piece in Axios in September 2025 stated that “a string of recent rural health clinic closures is threatening to further reduce access to care in outlying areas as health systems brace for cuts in the in the Republican budget law,” listing several examples of providers shutting down or consolidating.

It is true, of course, that some providers have exited the market over the past year. But individual closures are a normal feature of any economic sector, including this one. An accurate accounting must consider both sides of the ledger – the facilities opening as well as the closures. By that measure, there is no evidence the law has undermined access to safety-net providers.

This PIC, drawing from comprehensive CMS enrollment data on Rural Health Clinics (RHCs) and Federally Qualified Health Centers (FQHCs), does not support the picture of a smaller health care safety net. Together, these facilities deliver outpatient medical services to millions of low-income and rural Americans. RHCs are specifically designated to address health care provider shortages in rural areas, while FQHCs are community-based centers that provide comprehensive care to primarily to low-income patients and communities with limited access to health care.

Since the OBBB’s passage, the number of RHCs and FQHCs nationwide has grown. According to CMS data, between July 1, 2025 — three days before the OBBB was signed — and July 17, 2026, the latest data release, the number of RHCs enrolled in Medicare rose by 148 (or 2.7 percent), and the number of FQHCs increased by 338 (or 3.1 percent). That is a net gain of nearly 500 facilities serving rural and lower-income communities. The data also show that the increase in the number of RHCs and FQHCs pre-dated the OBBB and has continued at a similar pace since the law’s passage. Compared to the fourth quarter of 2023, the earliest available data from CMS, the number of RHCs has grown more than 400 and the number of FQHCs has increased by nearly 1,400.

Nearly 500 More Health Care Facilities Serving Rural and Low-Income Americans are Operating than Before the OBBB

In part, the recent increase in safety-net facilities may reflect a core feature of the OBBB, the Rural Health Transformation Program, which began distributing $10 billion in annual federal funds to states in early 2026. Many states have begun soliciting grant applications and disbursing funds to clinics and hospitals.

To be sure, some of the OBBB’s biggest reforms — community engagement requirements and limits on state-directed payments — have yet to take effect, and it is premature to empirically judge the law’s long-run impact. But alarmist claims that the law’s passage caused the collapse of safety-net providers are simply untethered from reality.

4Aw Spending Outpaced Growth In Oecd Countries A0Wuu000005Tjvnya2

August 31, 2026

US Health Spending Has Significantly Outpaced Growth in OECD Countries Since the ACA Took Effect

One of the recurring claims made about the Affordable Care Act (ACA) is that it “bent the cost curve.” The evidence usually offered is the slowdown in U.S. health spending growth that began around 2010, when the law was enacted. (The actual slowdown began several years prior to then.) But attributing causality based on such a pre/post comparison is simplistic and often misleading. What if broader, global forces would have dampened the growth of U.S. health spending even in the absence of the ACA?

One way to test the ACA’s role is to compare the trajectory of U.S. health spending against other high-income countries that did not enact similar major health policy changes in or around 2010. This PIC indexes real per capita health spending (converted at purchasing power parity) to its 2000 level for the United States and for the 36 other OECD countries with complete data through 2025.

By 2025, U.S. spending per person had roughly doubled relative to 2000. The comparison countries grew about 85 percent at the mean and about 80 percent at the median. Weighting by GDP (so that Germany and Japan count for more than Iceland), per capita health spending in the rest of the OECD grew about 65 percent.

Two features of the figure are crucial for assessing the ACA’s impact.

First, growth in per capita health spending from 2000 to 2010 was virtually identical in the U.S. (40 percent) and across other OECD countries (mean of 40 percent, median of 41 percent). The series continued to track closely through 2013, when per capita health spending in the U.S. was 43 percent higher than in 2000, the OECD mean was 42 percent above its 2000 level, and the OECD median had grown 45 percent.

Second, from 2014 onward U.S. spending growth accelerated and pulled away from the comparison group of other OECD countries, and the gap has continued to widen through 2025. This timing coincides precisely with the implementation of the ACA’s principal coverage provisions which have substantially increased federal health spending. By 2019, the U.S. was 69 percent above its 2000 level, while the OECD mean had grown 62 percent and the median had risen 57 percent.

Admittedly, other OECD countries do not constitute a perfect control group. Different countries faced a range of macroeconomic conditions, deployed a variety of cost containment measures to address health costs, and adopted different fiscal policies in response to the pandemic. Despite these limitations, it is notable that the divergence in health spending between the U.S. and other OECD countries occurred just as the ACA entered full implementation. Looking across countries, this pattern is more consistent with the ACA increasing health spending than decreasing it.

While the ACA almost certainly reduced cost pressures within Medicare through its hospital payment reductions, these savings were more than offset by increased spending elsewhere — namely, through the expansion of Medicaid, generous exchange subsidies, and accelerated industry consolidation spurred by the law’s regulatory burdens. This trajectory of higher spending was further exacerbated by the COVID-19 pandemic and subsequent Biden administration policies that emphasized enrollment over proper eligibility determinations. While U.S. health spending surged at the onset of the pandemic in 2020, international responses briefly narrowed the spending gap in 2021 and 2022 before U.S. per capita spending re-accelerated from 2023 to 2025, reestablishing a widening divergence from international peers.

2Mh Prescription Drug Prices Drop Even Faster A0Wuu000005Ptivyai

August 18, 2026

Prescription Drug Prices Drop Even Faster Than in First Trump Administration—Hospital Price Inflation Still Hot

The media described the Consumer Price Index (CPI) report for July as showing inflation cooling for the second month in a row. One major exception is hospital prices, which continue to rise rapidly. Hospitals account for 31 percent of health spending, and prices for inpatient and outpatient services rose 5.2 percent and 5.9 percent respectively over the past year, nearly on par with what we found in last year’s Paragon PIC discussing inflation for the twelve months through July 2025. Prices for physician services grew about one and a half percentage points slower than prices for all items less medical care.

Most remarkable is the significant decline in prescription drug prices, a drop of 3.1 percent over the last twelve months—the largest decline in over six decades.¹ Although retail prescription drugs only account for 9 percent of health spending, this is a welcome trend. It’s also reminiscent of the first Trump term. In the March 2019 Economic Report of the President, the Council of Economic Advisers (CEA) explained that “For the first time in 46 years, the Consumer Price Index for prescription drugs fell in both nominal and real terms during a calendar year” (p. 198). The CEA showed that prescription drug prices had been increasing about five percentage points faster than Personal Consumption Expenditures (PCE) from 2013 to the end of 2016. Beginning in 2017, they grew by only one percentage point faster than PCE (p. 236, fig. 4.7).²

The figure shows that the fall in drug prices started in February 2026. One notable development that month was the launch of TrumpRx, the Administration’s direct-to-consumer (DTC) website where patients can buy discounted medicines. TrumpRx launched with 43 branded medicines available and has since added many more drugs, so that it now offers over 800 medicines.

TrumpRx is a storefront that operates through vendors like GoodRx, Mark Cuban’s Cost Plus Pharmacy, and manufacturers’ own discount programs. These cash-pay channels bypass many of the “middlemen” that drive up drug prices and have gradually increased their own market share, up to about 9 percent of transactions in 2022. Even patients who have insurance coverage for outpatient drugs often find it less expensive to pay cash.

According to one market expert, “As cash-pay prices appear on a federally promoted platform, a new layer of public transparency sets in. Patient expectations around what a drug should cost could begin to anchor around those figures, creating internal pressure on manufacturers to rationalize the relationships between list price, net price, and DTC price.” Another industry insider credits a clarification from the U.S. Department of Health and Human Services Inspector General that manufacturers can sell directly to Medicare and Medicaid enrollees without triggering the Anti-Kickback Statute.

The benefit of DTC is most apparent with GLP-1s, the anti-obesity medications that have become very widely used. GLP-1s were among the first drugs available on TrumpRx in February, and the Administration also launched a demonstration called BRIDGE that allows Medicare beneficiaries to access them for a $50 copay. However, Medicare Part D insurers cover GLP-1s for diabetes or some other conditions but not obesity alone. Only 36 percent of employer-sponsored plans cover GLP-1 drugs for obesity—usually only with prior authorization—and that share is falling. The result is that the market for GLP-1 drugs is a consumer-driven market.

It is too early to estimate how much of the recent drug price reduction can be attributed to DTC platforms like TrumpRx. Since generics account for more than 90 percent of prescriptions, the steady approval of generic applications (particularly first generics) is an important driver of lower prices. Others are attributing a portion of the decline to the Inflation Reduction Act (IRA). Medicare implemented negotiated prices for the first ten drugs subject to the IRA in January 2026. However, the timing and persistence of the decline make it difficult to attribute much of the trend to the IRA. The first negotiated prices took effect at a single point in January and applied to only ten drugs, but there was no significant price reduction that month, and the negotiated prices would not have kept trending down.

The precise causes of the historic decline in prescription drug prices will take time to disentangle, but consumers are clearly benefiting from greater competition and new ways to purchase medicines. The contrast with hospital care is striking: while drug prices are falling at their fastest rate in decades, hospital prices continue their relentless rise.

Footnotes

1↑ The accuracy of the CPI for prescription drugs is limited by a two-year lag in the basket of drugs used to estimate prices. Prices for 2026 are based on a basket of drugs available in 2024, and many new drugs are introduced every year. Prices represent transaction prices between the pharmacy, patient, and third-party payer, if applicable. The index excludes prices for prescriptions with no patient payment, such as hospital inpatient services or outpatient infusions or injections.

2↑ PCE and CPI are different measures of inflation produced by different federal agencies, the Bureau of Economic Analysis (BEA) in the Department of Commerce (PCE) and the Bureau of Labor Statistics (BLS) in the Department of Labor (CPI). The PCE is mostly compiled from business surveys and the CPI from consumer surveys. The PCE responds more quickly to substitution because it uses a chain-weighted index. So, when consumers switch to less expensive goods, it is reflected in the next month. The BLS updates the basket of goods and services in the CPI annually, with a two-year lag. (The 2026 CPI uses 2024 weights.)

3JRG National Health Expenditures A0wUU000005oX7ZYAU

August 12, 2026

Health Care Spending Continues to Outpace the Economy

Earlier this summer, the Centers for Medicare and Medicaid Services (CMS) released its latest National Health Expenditures projections, and the outlook is alarming. National health spending is growing at more than 7 percent annually—about two percentage points faster than GDP. When health care consistently grows faster than the economy, it consumes an ever-larger share of national income. CMS projects that health care spending will exceed one-fifth of the U.S. economy early in the next decade.

Much of this growth reflects expanding federal involvement in health care. During the Biden administration, Medicaid enrollment increased dramatically because of pandemic-era continuous eligibility requirements, ACA subsidy expansions, and widespread improper enrollment in both Medicaid and the ACA exchanges. At the same time, states increasingly exploited financing arrangements such as Medicaid state-directed payments to draw down additional federal matching funds, largely to fund corporate welfare for large, politically powerful hospital systems.

Hospital services are the largest driver of spending growth. Hospital expenditures increased 8.9 percent in 2024 and are projected to rise another 8.2 percent in 2025—both well above overall health care spending. As John R. Graham documented in recent Paragon research, hospitals remain the least efficient major sector of American health care, yet they continue to command an ever-larger share of national health spending through market consolidation, government payment policies, and regulatory advantages. These trends reinforce the urgent need for reforms that promote competition, transparency, and more efficient delivery of care.

The implications extend well beyond health policy. Despite historically high federal revenues, annual budget deficits continue to approach $2 trillion because federal spending continues to rise even faster. In a 2023 Paragon research paper, Paul Winfree—now an advisor to Federal Reserve Chairman Kevin Warsh—explained how rapidly growing federal health care programs, together with rising interest costs on the national debt, are steadily consuming America’s fiscal space and limiting policymakers’ ability to respond to future economic, fiscal, and national security challenges. The latest National Health Expenditures projections reinforce that warning.

3AW Medicare Site Of Service Payment A0wUU000005fkX3YAI

July 21, 2026

Medicare’s Site-of-Service Payment Gap is Projected to Keep Growing

Medicare often pays more for the same service when administered in a hospital outpatient department (HOPD) than an independent doctor’s office. This Paragon Pic shows that the disparity between the two sites of service is wide and growing.

We analyzed data for 4,281 services across HOPDs and physician offices and found that Medicare’s site-of-service payment gap has grown sharply. In 2011, HOPDs received 109 percent more than physician offices for the same services on a volume-weighted basis; by 2026, they received 247 percent more. In nominal dollar terms, average HOPD payments rose from $258 to $355, while physician-office payments fell from $123 to $102. If this trend continues through 2035, HOPDs will receive $430 on average while physicians’ offices will only receive $92–a 369 percent advantage for hospital-based care for services that are virtually indistinguishable.

This gap is the result of two separate payment systems that Medicare uses to pay for services. The first is the Outpatient Prospective Payment System (OPPS), which ostensibly compensates hospitals for the capital and labor expenses involved in providing services. The second is the Physician Fee Schedule (PFS), which pays physicians for services rendered, regardless of where that service is performed. A service that is performed in a HOPD by a physician will receive two payments: an OPPS payment for the hospital—often called a “facility fee”—and a PFS payment for the physician. Meanwhile, that same service performed by a physician in an independent physician’s office will only receive one payment from the PFS. Medicare pays a higher non-facility PFS rate than the facility PFS rate because the physician incurs practice expenses that the hospital otherwise bears. However, the OPPS facility fee that hospitals receive is generally equal to or larger than the PFS rate. Because most hospital-employed physicians are salaried, hospitals usually receive both the OPPS and PFS payments. As demonstrated in the Paragon Pic, the combined hospital and physician payment typically exceeds the office-based physician payment by a wide margin.

Medicare awards this extensive reimbursement for outpatient procedures even though outpatient facilities are relatively low-cost compared to inpatient facilities. The Medicare program functionally recognizes this fact in the slightly larger PFS rate paid to independent physicians, who must maintain a facility that covers the same services as a HOPD. This payment differential is effectively a subsidy benefiting hospitals that inflate operational costs and reducing their incentives to improve efficiency. To make matters worse, Medicare automatically adjusts the OPPS facility fee for inflation but does not do the same for PFS payments to physicians due to the different designs of the payment systems.

These payment differentials create significant distortions, including incentives for vertical consolidation. They also mean more costs for taxpayers, and the distortions and waste will only grow the longer Congress allow them to persist. Introducing a Medicare site-neutral payment reform would reduce incentives for hospitals to acquire physician practices and lower costs for patients and taxpayers.

19MH PIC Almost Entire Obamacare Premium Inc A0wUU000005hNkrYAE

July 20, 2026

Taxpayers Pay for Almost the Entire Increase in Obamacare Premiums

Obamacare, including its perverse subsidy design, continues to drive up insurance premiums and health care costs by largely insulating subsidized enrollees from premium increases. The premium increases for Affordable Care Act (ACA) plans are concealed from enrollees since the enrollee’s share is capped at a percentage of their income. This subsidy design means that when insurers raise premiums, the cost is borne by the taxpayer.

This Paragon PIC illustrates how the cost of a benchmark ACA premium has been divided between taxpayers and a representative enrollee—a 50-year-old with income at 200 percent of the federal poverty level—since the exchanges began. The orange portion represents the enrollee share, and the navy amount represents the share borne by taxpayers.

The ACA’s major coverage provisions took effect in 2014 and significantly increased individual market premiums. In that first year, taxpayers covered 68 percent of the premium. Between 2014 and 2020, the annual premium increased from about $4,500 to $8,000. Yet, the enrollee amount stayed nearly flat during this period as taxpayers absorbed almost the entire premium increase. By 2020, taxpayers covered 80 percent of the premium—a significant increase from 2014.

From 2021 to 2025, COVID-era subsidy boosts (shown in light blue) replaced much of the enrollee share with expanded subsidies. That increased the government’s share of the premium to 93 percent of the cost. Despite the COVID subsidy boosts expiring after 2025, the underlying ACA subsidy still covers more than 80 percent of the premium for this representative enrollee.

Some reports show preliminary premium increases of 14 percent in 2027, prompting claims that enrollees will face significant cost increases. As this PIC shows, that narrative is misleading. For the representative enrollee shown here, the annual out-of-pocket premium will increase by only about $40 from 2026 to 2027—roughly $3 per month. The taxpayer’s share will increase by $1,426. For this enrollee, the government will pay more than 82 percent of the premium in 2027.

Moreover, this representative enrollee pays a larger share of the premium than most subsidized marketplace enrollees. Approximately 73 percent of 2026 ACA open enrollment period sign-ups claimed income below 200 percent of the federal poverty level. Because premium contributions decline as income falls, most subsidized enrollees pay even less toward their premiums—and taxpayers pay even more—than this figure illustrates.

3MH ACA Spending 64 Percent Higher Tha COVID A0wUU000005SJeLYAW

July 6, 2026

Affordable Care Act Subsidy Spending Will Remain Above Pre-Biden Projected Levels, Per CBO

CBO’s most recent estimates show that federal spending on Obamacare subsidies will remain well above pre-pandemic estimates. The One Big Beautiful Bill made important changes to improve the integrity of the exchanges.

This is especially important given the massive extent of improper enrollment in the exchanges. The CBO’s 2026 baseline projection reduces the rate of growth of Obamacare spending by $212 billion over ten years, 2027 to 2036, versus the baseline projection published in 2025. However, the 2025 baseline for the same period jumped by $583 billion from the 2021 baseline.

The main driver of the Biden spending increase was a set of administration policies that prioritized increasing Obamacare enrollment at any cost along with an expansion of subsidies as a COVID-era boost. These policies weakened program integrity measures, stopped income verification for many enrollees, and essentially opened a year-round enrollment period—and created large incentives for enrollees, brokers, and insurers to misstate applicant income to maximize subsidies. Paragon estimates that 6.2 million people were improperly enrolled in Obamacare in 2026.

(The 2021 baseline stops in 2031; and the 2025 baseline stops in 2035. Those baselines are extrapolated through 2036 using common-sense assumptions, as indicated by the dotted line in the figure.)

Paragon PIC

Footnotes

Reforming Government. Empowering Patients.