Penn Medicine / UPHS: Comprehensive Hospital Investigation

Hospitals examined: Hospital of the University of Pennsylvania (HUP), Penn Presbyterian Medical Center (PPMC), and Pennsylvania Hospital (PAH)
Location: Philadelphia, Pennsylvania
Research cutoff: August 27, 2026
Question: Is Penn Medicine really unable to increase nurse salaries after a 3% raise, an alleged 10% employee-health-insurance increase, and an alleged 20% parking increase?

Visual evidence board: [[Penn Medicine UPHS.canvas]]

This master file consolidates the ownership, hospital profiles, government ratings, finances, executive compensation, nurse/ancillary/physician pay, inflation models, staffing limits, parking and insurance questions, private-capital contracts, governance overlaps, 340B, charity care, acquisitions, sponsorships, lawsuits, controversies, and unanswered-record requests developed during the Penn Medicine investigation.

Visual orientation

The decisive charts

The short answer

The public evidence does not support the phrase “hanging on by a thread.” The University of Pennsylvania Health System (UPHS) had almost $12.0 billion of operating revenue, $247 million of operating income, $10.0 billion of net assets, and an AA long-term credit rating in fiscal 2025. It produced positive operating income in every fiscal year examined from 2018 through 2025.

But Penn is not printing unlimited cash. Its operating margin fell from roughly 5.6%–7.0% in stronger years to 1.6%–2.6% in the last four years. S&P described operations as weak for an AA-rated organization and said Penn had little cushion for disruption. Pennsylvania Hospital also recorded a small operating loss in fiscal 2024 and a deficit on its latest Form 990.

The fairest conclusion is:

Penn Medicine has the financial capacity to discuss and fund higher nurse compensation. Whether to do so is a recurring budget and labor-allocation decision—not proof that the system is insolvent or literally unable to pay. The sustainable amount cannot be calculated from public data alone because Penn does not disclose the number of affected nurses, their payroll, vacancy costs, contract-labor savings, or the proposed pay scale.

A 3% raise also did not preserve purchasing power against Philadelphia’s 5.4% year-over-year inflation in June 2026. Before insurance or parking, that is about a 2.3% real-pay decline: 1.03 ÷ 1.054 − 1 = −2.28%. If the nurse’s insurance and parking deductions also rose, the employee’s take-home position deteriorated further.

Claim-by-claim verdict

1. What Penn Medicine actually is

“Penn Medicine” is a brand for two connected parts:

1. The University of Pennsylvania Health System (UPHS) operates hospitals, clinical practices, and other care entities.
2. The Perelman School of Medicine is the University’s medical school and research enterprise.

The ultimate parent is The Trustees of the University of Pennsylvania, a private nonprofit corporation and tax-exempt 501(c)(3), EIN 23-1352685. It is not owned by private equity, public shareholders, or the Commonwealth of Pennsylvania.

This matters because Penn publishes several different financial views:

Penn’s official financial-report library publishes audited reports back to 1995. The fiscal 2025 report confirms the University/UPHS structure and provides the system summary. The FY2024 UPHS combining statement linked through Penn Medicine’s financial-reporting portal provides the latest publicly retrievable hospital-by-hospital operating table used here.

The three hospitals in plain English

Penn’s FY2024 Facts & Figures reported:

HUP’s 1,308 beds should not be interpreted as 1,308 beds in one tower. HUP has a broader license/campus footprint, including the 504-room Pavilion—renamed the Clifton Center for Medical Breakthroughs—and HUP–Cedar at 501 S. 54th Street.

2. What the government says about quality

The latest CMS Hospital General Information release examined was released August 13, 2026. CMS’s overall rating combines multiple mortality, safety, readmission, patient-experience, and timely/effective-care measures. It is a broad comparison tool, not a declaration that every unit is safe or that every patient will have a good outcome. See the CMS dataset and CMS explanation.

What patients said in HCAHPS surveys

The current HCAHPS period examined was October 1, 2024 through September 30, 2025. “Overall CMS stars” and “patient survey stars” are different scores.

The low response rates and differing patient populations matter. These are useful signals, not direct measures of nurse workload.

Pennsylvania inspections

The most recent Pennsylvania Department of Health occupancy surveys located in 2026 were compliant after review: HUP for a 21-bed Dulles 6 project, PPMC for a lactation-room/camera project, and PAH for a radiation vault. These were narrow project/occupancy surveys—not full retrospective clinical clean bills. The state’s health-facility portal contains facility and survey records.

3. Financial health: eight years in one chart

All figures below are UPHS-wide, not just the three Philadelphia hospitals. Dollar values are rounded from Penn’s audited annual reports.

How to read this

• Revenue rose about 77% from FY2018 to FY2025.
• Operations were positive in all eight years.
• The system’s cushion narrowed materially after FY2021. At a 2% margin, a modest revenue or cost shock can erase operating profit.
• FY2020’s total net-asset change was negative despite positive operations because nonoperating investment activity fell. Investment-market changes are not the same as patient-care operating performance.
• “Profitable every year” and “under cost pressure” can both be true.

FY2025 balance sheet and credit strength

S&P affirmed UPHS at AA/A-1+ in August 2025. Its rating rationale described an exceptionally strong enterprise profile and broad demand, but also said operating results were weak for the rating and left little disruption cushion. That is the most accurate translation of Penn’s condition: financially strong overall, with tight current operating margins.

The three hospitals by themselves

The FY2024 UPHS combining schedule shows:

PAH finished positive after nonoperating items in the FY2024 audited schedule, but its patient-care operations were slightly negative. In the latest public Form 990 period ending in 2025:

Form 990 totals include nonoperating and filing-specific items, so they will not exactly match the audited operating columns. The important pattern is that HUP and PPMC have recently generated positive operating margins, while PAH is the more fragile of the three.

A longer Form 990 screen

• PPMC had negative total results in only 2015 and 2017 among the 2011–2023 structured filing years examined.
• PAH had negative total results in only 2013 and 2014 among those years.
• The University parent reported positive total results in every structured 2011–2023 year examined.

This screen argues against chronic insolvency. It does not guarantee future liquidity or show money available for a particular contract.

4. Where the money comes from and where it goes

UPHS patient-revenue mix, FY2025

The system therefore depends heavily on negotiated insurer rates and federal/state reimbursement. The audited report said compensation, medical-benefit costs, market salary adjustments, pharmaceuticals, supplies, and growth all pressured expenses.

For the three Philadelphia hospitals in FY2024, the largest visible categories were salaries, benefits, clinical supplies/other costs, depreciation, and corporate/inter-entity charges. The combined corporate/inter-entity line was about $1.136 billion, or 18.6% of the three hospitals’ expenses. That is a legitimate transparency question, but it cannot be renamed “administrators and consultants”: the line can contain shared IT, HR, revenue cycle, insurance, system services, physician/academic allocations, and internal eliminations.

Transfers to the University and medical school

UPHS transfers/support to the University and Perelman School rose to approximately $419.3 million in FY2025, versus $32.9 million in FY2024. However, about $279.1 million of the FY2025 amount was fixed assets transferred between related Penn entities. It would be misleading to describe the whole $419 million as recurring cash siphoned away from bedside care.

The correct follow-up is to demand a reconciliation separating:

• recurring cash support;
• physician salaries and shared services;
• capital assets transferred;
• research/education commitments;
• one-time accounting entries.

Charity care

UPHS reported approximately $63.8 million of charity-care cost in FY2025, up from $47.2 million in FY2024. This is the narrow accounting measure of charity cost, not the broader “community benefit” figure hospitals often advertise.

5. What a larger raise might cost

The three hospitals reported $1.704 billion in total salaries in FY2024. As a rough scale illustration:

This is not a bargaining-price estimate. It includes non-nurses, excludes payroll taxes and benefit interactions, and does not account for vacancies, overtime, experience steps, differentials, retention savings, or contract labor. A nurses-only increase would cost less than the all-employee illustration.

The combined FY2024 operating income of the three hospitals was $225.3 million. This shows that a larger raise is financially material but not absurd on its face. The real negotiation should ask Penn to model:

affected RN payroll × proposed incremental percentage + payroll/benefit load − vacancy/turnover/agency savings

Without that model, neither “Penn cannot afford it” nor “the endowment can easily pay anything” is rigorous.

Why the endowment is not a simple answer

Penn’s consolidated investment portfolio was about $27.0 billion at June 30, 2025, including roughly $9.7 billion reported in private-equity funds, $5.3 billion in absolute-return funds, and $2.9 billion in real assets. The endowment was about $24.8 billion and distributed about $1.1 billion during the year.

Those are University consolidated investments—not a checking account owned by HUP, PPMC, and PAH. Much of an endowment is donor-restricted, invested for long-term use, and allocated to schools or purposes. Still, Penn’s consolidated wealth is relevant to the institution’s overall risk capacity and choices; it simply cannot be treated as unrestricted annual hospital payroll.

6. CEO and executive compensation

Kevin B. Mahoney has served as UPHS chief executive officer since July 1, 2019. The table uses Form 990 “reportable compensation” plus “other compensation” reported by Penn-related filing organizations. It is broader than cash salary and can be affected by deferred-compensation timing.

Five-year total: approximately $14.83 million.
FY2021 to FY2025 change: +36.4%.
FY2024 to FY2025 change: +19.0%, influenced by the unusually low “other compensation” reported in FY2024.

Sources are the University, PPMC, and PAH Form 990 filings. This comparison is useful, but it does not prove that insurance, parking, or nurse-pay restraint directly increased Mahoney’s compensation. Some Penn executives receive incentive/bonus compensation; establishing what metrics drive it would require employment agreements, board compensation-committee materials, and incentive-scorecard disclosures.

The responsible incentive conclusion

There is a plausible institutional incentive to protect margins, credit ratings, growth plans, and executive performance targets. There is not public proof that a specific executive personally profits from a nurse’s parking or insurance increase. That distinction should remain explicit.

7. Nurse wages, peer raises, and inflation

Philadelphia RN market

The latest official Philadelphia-area wage table located reports a registered-nurse median of $48.64 per hour, or $101,180 annually:

Source: O*NET/BLS local wage data for Philadelphia RNs.

CMS’s 2022 Occupational Mix Survey files imply the following average RN wage dollars per paid hour:

These are old, hospitalwide averages across RN cost centers—not a bedside base-rate schedule. They may include differentials and paid nonproductive time and should not be used to tell an individual nurse what they “should” earn.

Comparable union settlements

No peer is perfectly identical, but these are large academic/urban systems with public union terms:

• Temple/PASNAP’s October 2025 three-year settlement reported across-the-board raises of 3.25%, 3.25%, and 3%, with experience steps and adjustments producing much larger increases for some nurses. Source: PASNAP announcement.
• Cooper/HPAE’s 2024–2027 agreement reported 4% annual across-the-board raises, plus an annual 1% experience step for eligible nurses below the top experience threshold. Sources: HPAE Local 5118 and its May 31 agreement update.

Union sources understandably emphasize their wins. Temple and Cooper differ from Penn in payer mix, geography, staffing structure, and finances. The useful comparison is that Penn’s alleged 3% is near the low end of peer base increases and lacks the public step/bonus context that makes peer packages more valuable.

Inflation comparison

Philadelphia CPI-U rose 5.4% in the year ending June 2026; food rose 4.2% and shelter 6.5%. The local index can be volatile, so one should also note June 2025 inflation was 3.3% and December 2025 was 3.9%. Source: BLS Philadelphia CPI.

Therefore:

• 3% raise vs. 5.4% inflation = about −2.3% real change before deductions.
• If a nurse earning $100,000 receives $3,000 more gross pay, the actual gain depends on tax, premium, and parking-dollar increases—not their percentages alone.
• A 10% rise on a $100 monthly premium is $10; a 10% rise on a $600 family premium is $60. The plan and tier are indispensable.

8. Health insurance: what can and cannot be verified

The current UPHS employee premium sheets are behind Penn’s HR & You login. Public University faculty/staff rates are a different employee plan and should not be substituted for UPHS hospital-worker rates.

The audit does support the general context: Penn said employee-benefit expenses rose because of medical-cost increases, salary-market adjustments, and workforce growth. It does not disclose whether every nurse’s contribution rose exactly 10%.

Evidence needed to verify the 10%

Obtain the old and new open-enrollment sheets or two pay stubs showing:

• plan name;
• employee-only / employee-plus-spouse / family tier;
• per-paycheck employee contribution;
• deductible and out-of-pocket maximum;
• employer contribution;
• wellness surcharge or tobacco surcharge;
• pay frequency.

Then calculate both:

percentage change = (new deduction − old deduction) ÷ old deduction

and

annual dollar change = (new deduction − old deduction) × paychecks per year

9. Parking: the 20% claim tested

Penn’s posted FY2027 parking rates, effective July 1, 2026, generally rose about 5%:

Sources: Penn Transportation’s rate page and the FY2027 rate announcement.

A nurse moved from FY2026 “everyday campus” at $119 to FY2027 “24/7 campus” at $155.50 would experience a 30.7% increase. Other location or access changes could yield roughly 20%. Thus the nurse’s experience may be real even though the general published price increase was not 20%.

What is not public

• Penn does not separately disclose parking revenue or profit for these three hospitals in its audited statements or Form 990s.
• No verified 2026 ownership or contract change was found that explains a general 20% hospital-parking increase.
• A 2023 Penn story identifies REEF as a HUP garage/valet operator at that time; PPMC’s public information describes its garage as hospital-operated. This does not establish current terms or profit-sharing.
• University Parking describes itself as a self-supporting auxiliary, but University parking accounts should not automatically be treated as identical to all hospital garages.

In FY2024 Penn announced a roughly 5% rate increase and referenced a $10 million multiyear parking-improvement program. Source: Penn Almanac FY2024 rates. A defensible parking-profit answer requires the garage vendor contracts, annual gross receipts, operating costs, debt service, capital allocations, and hospital/University revenue share.

10. Nurse staffing: what the public data can and cannot show

Pennsylvania does not publish California-style mandatory unit-by-unit nurse-to-patient ratios for these hospitals. Medicare cost reports contain paid RN hours, but not the real assignment on a particular ICU, ED, or medical-surgical shift.

A screening model using CMS 2022 occupational paid RN hours, a paid-to-productive conversion factor of 0.559559, and 2023 HCRIS hospital cost-report patient days gives:

modeled productive RN HPPD = paid RN hours × 0.559559 ÷ inpatient days

Grade C means modeled screening evidence, not an observed staffing ratio. The measure mixes units, acuity levels, paid time, outpatient cost centers, and older periods. It cannot establish “safe” or “unsafe” staffing.

To answer staffing credibly, nurses should request:

• budgeted and actual RN hours per patient day by unit and month;
• average and maximum nurse-to-patient assignment by shift;
• vacancy, turnover, and time-to-fill rates;
• mandatory overtime and missed-meal data;
• agency/travel RN hours and expense;
• sitter, tech, LPN, and float-pool hours;
• acuity-adjustment rules;
• workplace-violence and injury logs;
• assignment-objection or unsafe-staffing forms.

11. Administrators, consultants, and job cuts

The statement “more administrators and consultants than ever” could not be validated from public records because Penn does not publish a ten-year UPHS administrative-headcount or consulting-expense series.

What is known:

• The three hospitals recorded about $1.136 billion in FY2024 corporate/inter-entity charges, but those charges cover more than administrators and consultants.
• Penn’s consolidated University financials contain broad salary, benefit, supply, and other categories—not a clean consulting line for UPHS.
• In March 2025, Penn said it would eliminate roughly 300 jobs, more than 100 of which were vacant or expected retirements. Reports said affected categories included office and technical managers and a small number of senior leaders, and Penn said patient services would not end. See Becker’s summary and The Philadelphia Inquirer.

The 2025 cuts are counterevidence to a simple “administrators only rise” narrative, but they do not reveal whether consultants replaced employees or whether bedside administration has grown. The missing records are vendor-level accounts payable, management FTEs by cost center, consulting contracts, and organization charts over time.

12. Acquisition and major-building timeline

Nonprofit hospital combinations often use membership substitution: the parent becomes the sole corporate member, assumes control and liabilities, and records the contributed net assets. It is not always a cash purchase.

Doylestown deserves special attention

S&P reported that Doylestown had lost about $26 million through the first nine months of the period it discussed. Penn’s FY2025 audit shows Doylestown brought $467.8 million in assets but also $316.0 million in liabilities. The pro-forma UPHS-plus-Doylestown operating income would have been about $219.9 million, below UPHS’s reported $246.7 million.

That supports the nurse’s intuition that growth can absorb money and management attention. It does not show Penn paid owners a windfall: Doylestown was nonprofit, no consideration was exchanged, and Penn assumed substantial liabilities.

Lehigh Valley correction

Lehigh Valley Health Network is not a Penn Medicine acquisition. It combined with Jefferson on August 1, 2024. See Jefferson’s official announcement.

Capital versus wages

Buildings are funded over decades through combinations of cash, borrowing, philanthropy, and restricted capital gifts. A donor-restricted building gift cannot simply be converted into annual raises, and debt-financed construction creates future debt service. But capital plans still reflect institutional priorities and risk appetite. It is fair to ask why Penn accepts long-term expansion risk while holding recurring labor growth to 3%.

At June 30, 2025, UPHS reported $5.924 billion of net property and equipment. Penn’s larger University-consolidated reporting universe spent approximately $979 million on property, plant, and equipment during FY2025; that figure includes academic as well as health-system projects and must not be assigned entirely to the hospitals. The public audit does not provide a complete parcel-by-parcel real-estate portfolio, original purchase price, current appraised value, rent roll, or facility-level return. A true property dossier would require Philadelphia and suburban deed/parcel records, bond offering statements, leases, and UPHS’s fixed-asset register.

13. Sports sponsorships and connections

The contracts include branding, sports-medicine services, team care, content, community programs, and employee recognition. The likely business logic—an inference, not a disclosed internal memo—is patient acquisition, sports-medicine referrals, brand visibility, recruitment, and association with prominent teams.

The 76ers relationship also has visible Penn connections: principal owners Josh Harris and David Blitzer are Wharton alumni, and Joel Embiid gave $1.3 million to Penn Medicine’s COVID-19 relief work in 2020. See Penn’s gift story. Alumni and philanthropy are connections; they are not evidence of a hidden quid pro quo.

Penn’s 76ers “Assists for Safe Communities” program announced $76 for each assist. Using a previously cited 2,041-assist season would imply more than $150,000, but the exact audited donation and who ultimately bore it were not located. See Penn’s program description.

What remains unknown

The sponsorship fees, media value, referral revenue, athlete-care reimbursement, renewal terms, bidding process, and conflict reviews are not public. Without them, no one can honestly calculate how many nurse raises the sponsorships equal.

14. Private equity check

What was found

• The University/UPHS and the three hospitals are nonprofit-controlled, not PE-owned.
• Public Penn pages identify Penn faculty and Clinical Practices of the University of Pennsylvania as central physician providers.
• No reliable public source identified TeamHealth, Envision, US Acute Care Solutions, Sound, NorthStar, or another PE-backed company as the systemwide core emergency, anesthesiology, hospitalist, or radiology contractor for HUP, PPMC, or PAH.

What that does not prove

Hospital vendor and professional-services contracts are not fully public. There may be PE-owned vendors in narrower specialties, outsourced services, revenue cycle, technology, supply chain, or system affiliates. A definitive answer requires:

• department-by-department professional-services contracts;
• physician-group tax IDs and ownership disclosures;
• 1099/vendor ledger;
• anesthesia, ED, radiology, pathology, hospitalist, behavioral-health, and revenue-cycle agreements;
• current beneficial-ownership records.

Penn’s University endowment also invests in private-equity funds. That makes the University an investor in PE funds; it does not make Penn Medicine a PE-owned hospital system or prove a PE fund controls staffing at the three hospitals.

15. Controversies and negative signals

These events should be described precisely; none alone proves a systemwide pattern.

HUP–Cedar resuscitation failure

After a November 2023 death at HUP–Cedar, Pennsylvania inspectors found immediate-jeopardy conditions related to a delayed resuscitation response and staff knowledge of how to summon a code. Penn retrained staff, and the immediate jeopardy was lifted after corrective action. HUP–Cedar is a remote HUP campus, not the main 3400 Spruce building. Source: The Philadelphia Inquirer.

Major malpractice verdict

A Philadelphia jury returned a $182.7 million birth-injury verdict involving HUP, and appellate litigation followed. This is a legal finding in one case, not a CMS system rating or proof that all maternity care is unsafe. See the 2025 Pennsylvania Superior Court opinion.

Workplace violence and safety

In 2024, a vehicle struck nurses outside PPMC while they were treating a gunshot victim, killing one nurse and injuring others. This was an extraordinary act by a third party, not established hospital misconduct, but it raises legitimate questions about ED perimeter security and support for staff exposed to violence.

Workforce reductions

The 2025 elimination of about 300 roles is a negative workforce signal and evidence of active cost control, even though Penn said it would not eliminate patient services.

Counterevidence that belongs beside the controversies

• Current CMS overall ratings are strong: 5 stars at HUP and 4 at PPMC and PAH.
• Current mortality summaries show no mortality measures worse than the national comparison at the three facilities.
• The narrow 2026 state occupancy surveys located were compliant.
• UPHS remains AA-rated and operationally profitable.

The evidence supports targeted safety and labor questions, not a blanket claim that Penn’s hospitals are broadly failing.

16. Is Penn the only important hospital option in the region?

HUP is one of the Philadelphia region’s dominant academic/quaternary referral centers. PPMC is a major Level I trauma center, and PAH is important for Center City maternity and adult care. For certain transplant, oncology, neuroscience, and experimental-treatment programs, alternatives may be less substitutable.

But Penn is not the only hospital system:

For emergency care, use 911 or the nearest appropriate emergency department. This report is an institutional investigation, not a recommendation to delay emergency treatment.

17. What nurses appear to be saying publicly

Public Reddit discussions contain mixed, unverified anecdotes: some nurses describe Penn’s benefits, acuity, and academic environment positively; others say pay offers were below competing systems, complain about parking, or compare HUP and PPMC workloads. Individual posts cannot establish a wage scale, staffing ratio, or majority sentiment.

No authenticated private Facebook nurse-group review was conducted, and no private group content should be represented as public evidence. X/Twitter search was technically unavailable during this review. Therefore the social-listening result is partial, weighted toward public Reddit/search-indexed pages, and not representative.

The strongest next step is a confidential, structured nurse survey—not selective screenshots—with hospital, unit type, years of experience, base wage, differentials, benefit tier, parking category, typical ratio, missed breaks, overtime, and whether the respondent is willing to document the claim.

18. The best evidence-based organizing questions

Penn should be asked to answer these in writing:

1. What was the exact FY2026-to-FY2027 UPHS nurse salary-grid change by experience step, and how many nurses received only 3%?
2. What are the old and new employee medical-premium rates by plan and coverage tier, in dollars and percentages?
3. What are the old and new parking deductions by garage, access tier, and shift? How many employees were moved into 24/7 pricing?
4. What are RN vacancy, turnover, agency, overtime, and missed-break costs at HUP, PPMC, and PAH?
5. How much would each additional 1% in RN base pay cost, net of reduced vacancy/agency/turnover expense?
6. Break the $1.136 billion FY2024 corporate/inter-entity charge into IT, HR, revenue cycle, consultants, management, insurance, school/physician support, and other categories.
7. Provide management/administrative FTE and consultant-spend trends for FY2018–FY2026.
8. What recurring cash support moved from UPHS to the University/medical school in FY2025 after excluding the $279.1 million fixed-asset transfer?
9. Disclose sports sponsorship fees, clinical-service reimbursement, referral/marketing metrics, conflict reviews, and renewal terms.
10. Disclose professional-services vendors and beneficial owners for ED, anesthesia, radiology, pathology, hospitalists, behavioral health, and revenue cycle.
11. For every capital project above $25 million, identify restricted gifts, debt, unrestricted cash, expected return, and annual debt service.
12. Provide unit-level budgeted and actual productive RN HPPD and assignment ratios by shift for at least 24 months.

19. Final assessment

What can be said firmly

• Penn Medicine is a University-controlled nonprofit health system, not a PE-owned hospital chain.
• UPHS is not near collapse based on public financials. It is profitable, AA-rated, and has a $10 billion net-asset base.
• Its operating margin is substantially thinner than in stronger pre- and mid-pandemic years, so management’s cost concern is not invented.
• A 3% raise fell below the latest 5.4% Philadelphia inflation rate.
• The general posted parking increase was about 5%, but a category change can make an individual increase much larger.
• Penn’s current employee-health premium sheet, sports-contract costs, parking profit, consultant spending, unit staffing ratios, and exact nurse-payroll cost are not publicly disclosed.
• HUP and PPMC have recently been profitable; PAH has shown modest financial weakness.
• Penn is expanding through construction and nonprofit affiliations while controlling labor costs. That is a priority choice worthy of scrutiny.

The answer to the nurse’s question

No public evidence shows that Penn Medicine is financially unable to raise nurse salaries. The evidence instead shows a wealthy, highly rated system with positive but tightening operating results, major capital and expansion commitments, and discretion over how it allocates recurring dollars. Penn can reasonably argue that margins are thin; it cannot fairly translate that into “hanging on by a thread” without much stronger evidence.

The most defensible labor argument is not “Penn has $25 billion, therefore all of it is available.” It is:

“Penn has remained profitable and AA-rated, is choosing hundreds of millions in expansion and other institutional priorities, and has not publicly demonstrated why a below-inflation 3% nurse raise is the maximum sustainable amount after considering retention, vacancy, overtime, and agency savings.”

20. Evidence grades and limitations

Important limitations

• Financial years and tax-filing labels do not always align; dates are stated as fiscal periods where possible.
• UPHS, University consolidated, and hospital-level values are deliberately kept separate.
• Net assets are not cash; investment gains are not operating income; restricted endowment is not automatically payroll money.
• Membership substitution values are accounting fair values, not necessarily a purchase price paid to a seller.
• The latest hospital-level operating detail publicly retrievable for all three side by side was FY2024; FY2025 Form 990 totals were used as a separate cross-check.
• Social-media review was partial and did not access private Facebook groups.
• No inference in this report establishes corruption, self-dealing, illegal understaffing, or a causal link between executive compensation and nurse deductions.

Core source ledger

1. Penn annual financial reports, FY1995–FY2025
2. Penn FY2025 audited consolidated report
3. Penn Medicine financial-reporting portal for UPHS combining statements
4. Princeton Health financial-reporting portal
5. PPMC Form 990 archive
6. PAH Form 990 archive
7. University Form 990 archive
8. S&P UPHS rating rationale, August 2025
9. CMS Hospital General Information
10. CMS HCAHPS provider data
11. Penn Medicine FY2024 Facts & Figures
12. Penn Transportation FY2027 rates
13. BLS Philadelphia inflation
14. O*NET/BLS Philadelphia RN wages
15. Penn Doylestown affiliation announcement
16. Penn PPMC proton-center announcement
17. Penn Princeton cancer-center announcement
18. Jefferson–Lehigh Valley combination announcement
19. CMS 2022 Occupational Mix Survey
20. CMS HCRIS hospital cost-report data

21. August 2026 update: current wage benchmarks, inflation, and what Penn's 3% claim means

What was already known, and what this update adds

The earlier dossier already contained a Philadelphia RN wage snapshot, CMS-derived 2022 average RN hourly pay for the three hospitals, the employee's reported 3% raise, and a rough 5%/6% raise illustration. This update does not duplicate that work. It adds:

• current 2025 Philadelphia-market wage percentiles and signed 2026 peer-union scales;
• a three-year inflation planning path;
• hospital-specific RN workload-equivalent counts and inflation-adjustment costs;
• a bounded ancillary-worker wage and cost case;
• an explanation of how pay decisions can operate across three hospitals under one system; and
• hospital-specific 340B registration, one-time CMS remedy payments, and a tax-benefit/charity-care comparison with explicit limits.

Q: What is the normal RN wage now?

There is no publicly accessible 2026 Penn Medicine RN wage grid. Penn's current job postings describe benefits but generally do not publish base-pay ranges. The best defensible answer is therefore a market benchmark, not a claim that every Penn nurse earns the same amount.

The latest directly observed Penn figures remain the CMS occupational-mix data based on calendar 2022 payroll: HUP $52.52/hour, PPMC $52.37, and PAH $50.63. These are hospitalwide average paid RN wages, not current starting rates, not a pay scale, and not unique-person salaries.

Interpretation: Penn's older observed averages sat above the current metro median but below the top of the signed Temple and Cooper scales. That does not prove Penn is underpaying every nurse. It does show why Penn should release a current experience-step grid, differentials, and distribution of nurses by step before saying its package is market-leading.

Sources: BLS/O*NET Philadelphia RN wages · BLS Philadelphia May 2025 release · Temple RN agreement · Cooper RN agreement · CMS FY2025 wage-index files

Q: What inflation should nurses plan for over the next three years?

The Philadelphia Fed's third-quarter 2026 Survey of Professional Forecasters projects national headline CPI inflation of 2.3% in 2027 and 2.3% in 2028. It does not publish a Philadelphia-specific 2029 point forecast. For a transparent planning case, this dossier carries the survey's 2.3% long-term annual average into 2029.

This is a national forecast, not a promise and not a local Philadelphia forecast. Actual housing, insurance, parking, food, or healthcare deductions can rise faster or slower than CPI.

• A single 3% raise followed by two frozen years would leave purchasing power about 3.8% lower by the end of 2029 under this path.
• Three consecutive 3% annual raises would finish about 2.1% ahead of the modeled inflation path.
• The employee's reported 3% increase is still below the latest already-published Philadelphia CPI increase of 5.4%; that backward-looking fact and the forward forecast answer different questions.

Source: Philadelphia Fed, Survey of Professional Forecasters, Q3 2026

22. Hospital-by-hospital RN inflation-adjustment case

How many nurses are at each hospital?

Public records do not give a verified current headcount of unique bedside nurses at each campus. The CMS wage-index file does provide paid RN hours. Dividing those hours by 2,080 gives a full-time-equivalent workload measure, not a count of people.

What would merely keeping that payroll level with 2.3% annual inflation cost?

The model holds workforce size and mix constant and increases the 2022 RN salary base by 2.3% each year. “Three-year cumulative” means the sum of additional wages paid in years one, two, and three compared with freezing the base. The 25% loaded column is a planning sensitivity for payroll taxes and benefits, not an observed Penn cost rate.

The first-year direct cost, $12.67 million, equals about 5.6% of the three hospitals' combined FY2024 operating income of $225.3 million. This is not a claim that FY2024 profit is the correct or only funding source; it is a scale comparison. The model also omits vacancies, overtime, agency replacement, wage compression, shift differentials, and retention savings.

Formula: cost in year n = 2022 RN salary pool × [(1.023)^n − 1].

23. Ancillary workers: what is known, what is not, and a minimum cost case

Current Philadelphia-market hourly wage percentiles

Penn has not publicly released a hospital-employee ancillary wage grid or a document confirming what percentage raise each occupation received. The University of Pennsylvania's FY2027 University merit guideline uses a 3% aggregate pool and 0%–5% individual range, but that document is not proof of the UPHS hospital raise program.

Temple's signed allied-professional agreement is a useful peer check: across-the-board raises of 3.25%, 3.25%, and 3.0%, plus experience steps. Examples in the current scale include medical assistants/EEG technicians at $26.62–$29.10, certified pharmacy/monitor technicians at $29.53–$31.98, bachelor's-level medical technologists starting at $38.82, and nuclear-medicine technologists/radiation therapists at $59.92–$68.90.

Sources: Temple allied-professional agreement · Penn University FY2027 merit guideline · BLS/O*NET occupation pages for nursing assistants, medical assistants, pharmacy technicians, surgical technologists, radiologic technologists, and clinical laboratory technologists.

Minimum ancillary inflation case

To avoid calling every hospital salary an “ancillary clinical worker,” the minimum model uses only CMS occupational-mix salary pools for aides/orderlies/attendants and medical assistants. It excludes LPNs, physicians, nurses, and potentially overlapping departmental salaries. It therefore understates the full cost of covering lab, radiology, respiratory, pharmacy, surgical, and other technical workers.

The case for a raise: whether an employee “deserves” a raise is a value judgment. The measurable case is stronger: if nominal pay rises less than prices, purchasing power falls; low- and middle-wage workers are usually more exposed to insurance, transportation, food, and housing increases; and Penn competes against signed peer wage ladders. Penn should disclose starting/max rates, experience steps, vacancies, turnover, overtime, agency use, and actual 2026 raise rules by job and hospital.

24. How pay administration normally works across HUP, PPMC, and PAH

Short answer

It is normal for one health system to combine central compensation governance with hospital-level budgets and local management. It is also legally possible for different hospitals, jobs, bargaining units, or individuals to receive different increases. What is not supported by public evidence is the idea that Penn automatically sets each hospital's nurse raise equal to that hospital's annual profit margin.

The decision chain

1. The UPHS/University governing structure approves strategy, aggregate budgets, capital plans, and senior compensation architecture.
2. System HR/compensation functions set job families, market benchmarks, pay bands, benefits, and merit guidance.
3. Each hospital has its own executive team, payroll, budget, vacancies, service mix, and operating result.
4. Department leaders apply the approved rules to local employees, subject to market adjustments, experience steps, performance systems, scarce-skill premiums, and available budgets.
5. If workers unionize, the employer must bargain wages, hours, and other terms with the certified bargaining representative for that bargaining unit. The final contract—not an individual manager's preference—controls covered employees.

The National Labor Relations Board's acute-care rules recognize separate appropriate units such as registered nurses and technical employees. That means HUP, PPMC, and PAH nurses need not automatically form one systemwide unit, and different units can negotiate different contracts. Conversely, nothing prevents Penn from adopting a common systemwide market adjustment if it chooses.

What can and cannot be claimed

• Supported: PAH's operating loss does not mechanically prove that PAH nurses can receive only a small raise. HUP and PPMC profitability does not mechanically require a larger raise there. Penn's allocation and bargaining choices sit between campus results and employee pay.
• Supported: different raises can lawfully result from different bargaining units, job markets, experience steps, merit systems, or local budgets.
• Not established: Penn publicly uses a formula that ties nurse raises to each hospital's own margin.
• Not established: one campus is legally entitled to another campus's operating surplus.

Sources: NLRB representation and collective-bargaining FAQ · NLRB good-faith bargaining guidance · 29 CFR §103.30 acute-care bargaining units · Penn Medicine employee resources · Penn Medicine careers and total rewards

25. 340B, tax exemption, charity care, and the “gap” question

First: these are four different things

• 340B discount: a federal drug-purchasing discount available to eligible covered entities. It is not the same as a tax exemption.
• Tax exemption: avoided federal, state, and local taxes. Penn does not publish a campus-by-campus audited value of this benefit.
• Financial assistance/charity care: free or discounted care under a hospital's assistance policy, reported at cost.
• Community benefit: a broader IRS Schedule H measure that can include charity care, Medicaid shortfall, education, research, subsidized services, and community programs.

Are all three hospitals in 340B?

Yes. The HRSA Office of Pharmacy Affairs database lists active disproportionate-share-hospital registrations for HUP (DSH390111), PPMC (DSH390223), and PAH (DSH390226). Registration proves eligibility/participation; it does not disclose annual purchase discounts, net revenue, patient pass-through, or profit.

CMS's 2023 remedy file reports one-time corrections for an unlawful Medicare payment policy affecting 340B-acquired drugs from 2018–2022:

These payments are not annual 340B profit. They reverse prior Medicare underpayments and cannot be used to calculate Penn's discount spread.

HUP also underwent a federal HRSA audit covering FY2020. HRSA found inaccurate OPAIS registrations and an inaccurate/incomplete Medicaid Exclusion File; Penn corrected records, terminated ineligible sites, and the audit closed December 14, 2021. HRSA's published result says duplicate discounts did not occur. This is a documented compliance failure and correction—not proof of fraud.

Sources: HRSA 340B covered-entity search · HRSA HUP FY2020 audit result · CMS remedy final-rule page · CMS remedy fact sheet

A bounded tax-benefit versus charity-care screen

The Lown Institute's national methodology models an average nonprofit hospital's tax benefit as 5.9% of expenses. That is an advocacy/research benchmark—not Penn's tax return and not a government valuation. Applying it to FY2024 hospital expenses produces this transparent screen:

That narrow comparison is intentionally severe because it compares a modeled tax benefit against charity care alone. The broader Schedule H picture changes the result:

At the UPHS system level, FY2024 audited expenses were $10.621 billion and narrow charity-care cost was $47.17 million. A 5.9% benchmark would be $626.62 million, creating a modeled narrow gap of $579.44 million; charity would equal 7.53% of that modeled benchmark. But HUP cannot be isolated from that system number, and UPHS's broader Medicaid shortfall/community-benefit presentation is much larger. These definitions must not be mixed.

Sources: Lown 2024 methodology · Lown 2025 national report · AHA critique of Lown method · PPMC Form 990 archive · PAH Form 990 archive

Official version versus the skeptical/nurse version

Penn's strongest official case: wages and benefits are recurring costs; hospital margins are thin; capital gifts and bond proceeds are often restricted; 340B supports underpaid care and mission services; community benefit is broader than charity alone; and annual raises are one component of total rewards.

The strongest evidence-bound skeptical case: Penn has not published the current wage grids, hospital-by-hospital raise rules, vacancy/agency savings, ancillary-worker raises, annual 340B economics, or an actual campus tax-benefit valuation. It asks employees to accept a 3% figure without enough information to test whether that is the maximum sustainable adjustment. PPMC and PAH report narrow charity below 1% of expenses even though broader Schedule H totals exceed the 5.9% model. That makes transparency—not an unsupported fraud allegation—the central issue.

Records that would answer the remaining questions

1. Current RN and ancillary wage grids, experience-step distributions, premiums, and raise memos by hospital.
2. Unique headcount and productive FTE by occupation, campus, unit, and shift.
3. Vacancy, turnover, agency, overtime, missed-break, and recruitment costs.
4. Actual employer health-premium contributions and parking deductions by tier.
5. Annual 340B purchases, payer reimbursement, contract-pharmacy fees, charity/pass-through policy, and net program contribution by covered entity.
6. Campus-specific property-, sales-, income-, and other tax exemptions using local and state tax rules—not a national percentage proxy.
7. HUP's standalone financial-assistance and Schedule H-equivalent community-benefit allocation.

26. UPHS operating entities and the ownership map

The FY2025 audited report says UPHS consists of the following operating entities. This is a better guide to the system than treating “Penn Medicine” as one legal company:

This list does not necessarily include every subsidiary, disregarded entity, joint venture, foundation, property company, or practice name. The complete Schedule R/entity inventory and current corporate organization chart remain important disclosure targets.

Source: Penn FY2025 audited report, note defining UPHS, lines 1734–1740 of the locally extracted report; Penn financial reports.

27. Private-capital and outsourced-contract findings by hospital

The most important correction is that Penn is not private-equity-owned, but private-capital-backed vendors participate in parts of its operations. Public Form 990 contractor disclosures reveal only the five largest independent contractors for PPMC and PAH; they are not a complete vendor ledger or executed-contract database. HUP has no standalone Form 990, making its contractor universe substantially less visible.

Penn Presbyterian Medical Center

The reviewed contractor series documented recurring payments for security and staffing. The canvas aggregation found approximately:

PPMC's latest FY2024 Form 990 contractor table separately reported $3.097M to Allied Universal, $1.754M to OnCall, and $1.133M to GHR. The multi-year totals above aggregate observed top-five appearances, not total contract value.

Pennsylvania Hospital

PAH displays the clearest temporary-staffing exposure in the reviewed filings:

PAH's FY2023 filing alone reported $30.218M to GHR, then the FY2024 filing reported $9.579M. That large swing is relevant to the nurse-retention argument because a hospital can spend heavily on outsourced staffing while debating permanent-worker raises. It does not prove all temporary labor was avoidable or that the same clinical units were involved.

Pennsylvania Hospital anesthesia

Society Hill Anesthesia Consultants received approximately $53.635M across five reported years in the reviewed Form 990 series, including $15.540M in FY2024. Penn identifies its physicians as independent; the practice describes itself as Pennsylvania Hospital's exclusive anesthesia provider. The public record reviewed did not establish PE ownership. The appropriate records request is the anesthesia agreement, subsidy, billing rights, performance measures, beneficial owners, and renewal/competitive-bidding documents.

What was not established

• No public proof of a PE company running the three hospitals' core emergency departments systemwide.
• No complete HUP clinical/vendor ledger.
• No basis for converting every contractor payment into a contract value or a profit figure.
• No proof that PE ownership caused a particular staffing, safety, or patient outcome.

Sources: PPMC Form 990 archive · PAH Form 990 archive · MidOcean/GHR transaction · RC Capital/OnCall · Centerbridge/Medical Solutions · Vistria/Supplemental.

28. Parking ownership and private-capital chain

For HUP, the public record identifies an outsourced parking/valet chain:

• In 2018, Impark operated the HUP garage; valets alleged retaliation during organizing. Five workers were reinstated with back pay, while Impark denied the allegation.
• In September 2023, Penn identified REEF as HUP's garage and valet operator.
• In November 2023, Reimagined Parking, formerly REEF, announced Mubadala Capital as its controlling shareholder.

That establishes private-capital control of the publicly identified HUP operator. It does not disclose the contract price, renewal/amendments, gross receipts, operating profit, revenue split, or responsibility for the nurse-reported rate change. The same finding cannot be generalized to PPMC or PAH without their current contracts.

Sources: Penn identifies REEF · Reimagined Parking/Mubadala announcement · 2018 HUP valet dispute.

29. Executives, interested persons, and sponsorship governance

The latest hospital-level review displayed the following total compensation figures, combining direct and related-organization compensation where reported:

These amounts principally reflect executive, clinical, or administrative employment—not board attendance fees. The five reviewed PPMC/PAH filings did not reveal a Schedule L business transaction in the extracted investigation. That does not prove no relationship exists because Schedule L definitions, thresholds, exclusions, and reporting universes matter.

The 76ers overlap

David S. Blitzer is a Penn trustee and has held a Penn Budget and Finance leadership role; he is also a principal of Harris Blitzer Sports & Entertainment, connected to the Philadelphia 76ers. Penn and the 76ers announced a partnership in 2022. This is a real overlap that makes the contract's valuation, approval, disclosure, and recusal records important.

Penn's purchasing policy states that the President must approve a related-party purchase in writing before the purchase order or procurement-card transaction. Public records reviewed did not show that Blitzer personally received Penn money, approved the sponsorship, or failed to recuse. The evidence supports a records request, not a kickback allegation.

Sources: Penn related-party purchasing policy · Penn/76ers partnership.

30. Physicians: union status, exact resident pay, and the attending-pay gap

The clearest public physician-pay record is the signed UPHS–CIR/SEIU resident and fellow agreement, not an attending-physician compensation grid. It covers qualifying interns, residents, and fellows employed by UPHS at HUP, PPMC, and PAH from October 1, 2024 through September 30, 2027. Attending physicians and several excluded categories are not in this bargaining unit.

Resident/fellow salary scale effective October 1, 2026

The contract provides 3% salary-scale increases for each of its three contract years, along with provisions concerning benefits, four weeks of vacation, sick/personal days, education reimbursement, meals, parking/transit, fatigue-mitigation transportation, and other conditions.

Source: signed CIR–UPHS agreement, Article 38, pages 23–24.

Attending physicians

Penn's public physician-recruitment pages and a current Penn Medicine Clinician posting reviewed did not publish a usable specialty-by-specialty compensation scale. Broad BLS/O*NET physician data are poor substitutes because attending compensation varies enormously by specialty, clinical effort, academic rank, call, productivity/wRVUs, quality incentives, grants, and outside activity. Therefore:

• a current Penn attending-physician pay comparison remains not publicly answerable as one number;
• the correct comparison requires Penn offer letters or compensation plans by specialty and rank, plus comparable Temple/Jefferson/Cooper plans;
• resident/fellow salaries must not be compared to attending compensation as if they were the same labor market.

This is an important transparency gap rather than permission to invent a Penn average.

31. Union and organizing map

Penn's strongest official labor narrative emphasizes Magnet recognition, professional development, benefits, and shared decision-making. The worker-side argument is that shared governance does not substitute for enforceable bargaining rights over wages, staffing, safety, insurance, and parking.

Sources: CIR–UPHS agreement · Penn Medicine at Home United · Penn Magnet/shared-governance account.

32. Comprehensive finding ledger

33. Master records request

1. Current nurse, ancillary, resident/fellow, and attending-physician compensation plans by campus, job, specialty, experience step, shift, and differential.
2. Unique worker headcount, productive FTE, vacancies, turnover, overtime, agency/locum hours, missed breaks, recruitment, and replacement cost by unit.
3. Old/new employee health-premium sheets, employer contributions, deductibles, out-of-pocket limits, parking categories, and approval memoranda.
4. Complete vendor ledger and contracts by hospital and department, including beneficial owners and amendments.
5. GHR, Supplemental, Medical Solutions, OnCall, Allied, Society Hill anesthesia, surgical monitoring, revenue-cycle, consulting, and benefit-adviser contracts/invoices.
6. HUP/PPMC/PAH parking agreements, gross receipts, costs, capital allocations, operator fees, revenue splits, and rate-approval records.
7. 76ers, Flyers, and Philadelphia Union contracts; fees, in-kind services, valuation, referral/marketing metrics, renewals, trustee disclosures, approvals, and recusals.
8. Five-year management FTE, consultant, shared-service, and corporate/inter-entity charge detail, including the $1.136B FY2024 three-hospital line.
9. Executive employment agreements, incentive scorecards, compensation-committee minutes, comparability studies, deferred/severance terms, and recusal records.
10. Complete Schedule R/organizational chart, joint ventures, property entities, captives, physician practices, pharmacies, labs, foundations, and management entities.
11. Annual 340B acquisition cost, reimbursement, dispensing/contract-pharmacy fees, child-site economics, patient-pass-through policy, charity use, and net contribution by covered entity.
12. Actual campus-specific federal, state, and local tax benefits; HUP standalone financial assistance and community benefit.
13. Unit-level staffing grids, assignment sheets, productive hours, acuity, workplace-violence plans, corrective actions, and unresolved employee complaints.
14. Capital-project budgets, restricted-gift terms, borrowing, debt service, expected operating costs, capacity projections, and alternatives considered.

The public record supports a demand for these documents. It does not establish that the undisclosed answers are necessarily adverse.