# One Public Pension Is 25% Funded and Another Is 103%. The Difference Is Not What You Think

Chicago's fire pension is 25.25% funded. Colorado's is 103.5%. After reading 36 valuations across six states, three things predict which end you land on.

Author: J.A. Watte
Published: August 25, 2026
Source: https://jwatte.com/blog/six-state-pension-spectrum-what-separates-them/

---


I set out to compare Colorado and Idaho and ended up reading thirty-six public pension valuations across six states. Colorado, Idaho, Oregon, Florida, California and Illinois. Twenty cities and towns, from Chicago down to Pismo Beach.

The spread is wider than I expected. Chicago's firefighters' fund is **25.25%** funded. Colorado's statewide fire and police plan, measured on base benefits against the actuarial value of assets, is **103.5%**, a $172 million surplus the plan holds as a COLA reserve; book that reserve as a liability, as the actuary's own summary table does, and it reads exactly 100.0%. Same country, same decade, same broad promise to the same kind of worker.

What separates them is not benefit generosity, and it is not which party runs the place. Reading the documents rather than the commentary, it comes down to three things, and the third one surprised me.

I have put [all thirty-six plans on one axis](/infographics/co-id-or-fl-ca-il-pension-spectrum/) with every measurement date and asset basis attached, because a funded ratio without those is not a fact.

## The two ends

**Idaho PERSI** sits near the top: **90.89%** funded with a **$2.42 billion** net pension liability on the GASB market value measure at 30 June 2025, and on its funding valuation a $2.51 billion unfunded liability being paid down on a projected **8.2 year** schedule against a statutory ceiling of 25.

**Illinois's five state systems** sit at **47.4%** funded at 30 June 2025, on the smoothed actuarial value of assets its own funding law requires, with a **$144.5 billion** unfunded liability across five legally separate trusts. On market value they are 47.8% funded and $143.3 billion short. Chicago's four city funds sit below even that: municipal 28.18%, laborers 44.10%, police 26.44%, fire **25.25%**, with a combined net pension liability of **$36.4 billion**.

Between those poles sits everyone else, and each of these needs its basis printed alongside it. Colorado PERA at **75.31%** on the GASB market value measure at 31 December 2025, which is an informational roll-up of five legally separate trust funds rather than one pool of money; no single PERA fund is 75.31% funded and the divisions run from 66.1% to 90.3% on the funding basis. Oregon PERS at **73.1%** on its 31 December 2024 funding valuation, excluding side accounts. Florida's FRS at **82.2%** on the smoothed actuarial value of assets at 1 July 2025. CalPERS was **73.9%** at its last completed valuation, 30 June 2024, and its own annual report puts the current figure at an estimated 79%.

### The first difference: paying the actuarial cost

Idaho's board sets contribution rates against what the actuary says the plan needs, and the plan is far enough ahead that a scheduled 2.50% increase was pushed back to 2028.

Illinois has a statutory contribution that is **knowingly less than the actuarially determined one**. This is not an accusation, it is the finding of the state's own nonpartisan fiscal commission, which names actuarially insufficient employer contributions as the largest single driver of the hole since 1996.

That is the whole first difference. One state pays what the arithmetic requires. The other passes a law saying it will pay less, and the gap compounds.

### The second difference: keeping a valve somewhere

Every well funded plan I read has some mechanism that can move when the numbers go wrong.

Colorado PERA has an automatic adjustment provision that raises contributions and cuts the retiree increase when the projection drifts outside a corridor. It has fired twice, and the annual increase cap now sits at **1.00%**. Colorado's fire and police plan is fully funded on base benefits precisely because the cost of living increase was never guaranteed: load its own 2.50% assumption and the same plan reads **80.1%** funded with a $1.28 billion shortfall. The gap is absorbed by retirees taking a **0.27%** increase rather than by taxpayers funding a bailout.

Illinois has no valve at all, by constitutional design. Article XIII, Section 5 of the Illinois Constitution protects accrued pension benefits, and the Illinois Supreme Court struck down a reform act on that basis in 2015. Illinois can change what it pays in. It cannot change what it owes out.

You can think that protection is right or wrong and the arithmetic is the same either way: a system with no adjustment mechanism has to fund itself entirely from the contribution side, and Illinois chose not to.

### The third difference: not borrowing the ratio

This is the one I did not expect, and it is why I now read every funded ratio twice.

Oregon PERS reports **73.1%** funded excluding employer side accounts and **77.2%** including them. Side accounts are lump sums employers deposited into the system, worth **$4,400.2 million**. Where an employer funded one by issuing pension obligation bonds, the higher ratio was bought with borrowed money. The pension debt shrank and the bond debt appeared somewhere else on the same public balance sheet. Nothing was extinguished.

A ratio can improve because a plan was funded, or because a liability was moved. Those are different events that look identical in a chart.

## Chicago, and the number worth arguing about

Chicago is where the abstraction becomes a bill.

**76.0% of the city's own 2026 property tax levy goes to its four pension funds.** $1,411,865,000 of a $1,857,647,928 gross levy. That is not an outside estimate. The city prints the same split in its own budget book: police 44.0%, fire 19.9%, municipal 9.1%, laborers 3.0%.

But I want to be careful with it, because that figure is easy to misuse and I nearly did.

**76% is the share of the levy the city controls. It is not the share of anyone's tax bill.** A Chicago property tax bill also carries the Board of Education, the school building fund, Cook County, the Forest Preserve, the water reclamation district, the Park District and City Colleges. At the most recent extended rates published by the Cook County Clerk, for tax year 2024, the composite Chicago rate is **6.619 per $100** of equalised assessed value, of which the City of Chicago is 1.615 and the four pension funds are 1.285548.

**So the city's pension levy is about 19.4% of a full Chicago property tax bill.** Roughly one dollar in five. Schools take about 57%.

Both numbers are true. The first tells you what has happened to the city's own taxing capacity. The second tells you what a household pays. Quote the first without the second and a reader will hear the second, which is why both belong in the same paragraph.

One more detail, and it is the one I find most telling: **the pension levy has been frozen near $1.412 billion for five consecutive budgets**, 2022 through 2026, while the total city levy grew from $1,709,387,000 to $1,857,647,928. The pension share of the city levy was 79.6% at the tax year 2024 extended rates and 76.0% in the 2026 ordinance. Not because pensions got cheaper. Because the rest of the levy grew around a line that could not move.

The 2026 appropriation to the four funds is $2,899,698,491, of which **$2,843,221,414** is for pension purposes and $56,477,077 is a provision for uncollected taxes rather than money paid to the systems. That $2.9 billion equals 22.4% of the city's net local funds total and **46.4% of the Corporate Fund**.

## The result that contradicts the usual story

Police and fire pensions get the coverage. In Los Angeles they are the healthy ones.

**Los Angeles Fire and Police Pensions is 100.5% funded, a $158.5 million surplus on $28.8 billion of liability. LACERS, the civilian plan, is 74.6% funded with a $7.0 billion hole.** Same city, same 30 June 2025 valuation date, same actuary, both on the seven year smoothed valuation value of assets at a 7.00% assumed return. On market value the pair reads 105.8% and 76.3%. That is not a measurement artefact.

Two honest qualifications. The 100.5% is the **retirement benefit alone**; LAFPP also runs a retiree health subsidy that is 78.1% funded, so the **combined** plan is 97.6%. Still far ahead of LACERS, but not a surplus.

And I had a second example that did not survive checking. Tampa's fire and police fund reads 96.42% against 87.57% for its general employees, which looks like the same pattern. It is not usable: the two ratios sit at different measurement dates and use **8.50% and 7.00% discount rates**. A 1.5 point difference in the assumption that drives the entire calculation. I pulled it rather than lean on it.

The pattern is real but narrower than it first looked. Even inside Los Angeles it is not a uniformed versus civilian split: the city's other civilian plan, the Water and Power Employees' Retirement Plan, was 100.52% funded at 1 July 2025. And Chicago inverts it completely, with police and fire at the very bottom of the ladder.

Florida does something neither Colorado nor Idaho does. Under **Chapters 175 and 185** a city may tax insurance premiums written inside its boundaries and route the proceeds to its fire and police pension plans. Two separate taxes on two different lines of insurance at two different rates, not one uniform levy.

I wanted that to be the explanation and it is not. **St Petersburg's police plan draws the same subsidy and is the weakest of that city's three plans**, at 86.49% against 89.01% for general employees. The magnitudes cannot carry it either: ten years of state money amounts to 2.67% of Tampa's fire and police plan assets and 6.00% of St Petersburg's fire plan. And above a frozen base the money cannot offset city contributions at all, it buys benefits, which raises the liability alongside the asset.

So the pattern is real and its cause is not established by these documents. Los Angeles, the one clean case here, sits in a state with no such mechanism at all.

## The cities: what the household actually pays

The state comparison is only half of it. At city level the pattern that matters is how little of the bill the entity on the ballot actually controls.

**The Town of Parker, Colorado levies 2.602 mills of an 85.123 mill common stack. That is 3.1%.** On a $500,000 home the town collects about $81. South Metro Fire Rescue takes 12.250 mills, **4.7 times the town's own levy**, entirely outside the town budget, and its voters added 3.000 mills in November 2025.

**Greenwood Village has held its levy at 2.932 mills every year from 2016 through 2025** and has no fire department at all. **Breckenridge's fire district went from 9.531 to 14.019 mills in a single year**, a 47% increase, while the town's own levy is 5.07.

And in Florida you can live somewhere with no municipal government whatsoever. **Apollo Beach is unincorporated Hillsborough County.** No city council, no city budget, no municipal police or fire department. Its households pay a certified **18.2515 mills** across eleven separate ad valorem lines: county general 5.4608, a voted environmental lands levy 0.0604, a **county services levy that only unincorporated residents pay** 4.6163, library 0.5583, local school 3.2480, state school 3.0920, transit 0.5000, the Children's Board 0.4589, the water management district 0.1831 and the Port Authority 0.0737. On top of that, many neighbourhoods carry a Community Development District assessment that is not in the millage rate at all.

A resident can watch their city hold the line and still see the bill climb, because the expensive part of local government is usually a district nobody can name.

## The finding that reframes all of it

Here is the result that made me rewrite my own conclusion.

**Centennial, Colorado has no net pension liability and no OPEB liability anywhere in its audited statements.** It contracts policing from the county sheriff and employs no sworn officers. By every rule of thumb it should be fine.

Its sheriff's contract went from $34,179,580 in 2023 to **$43,589,360** in 2026. That is **+27.5% in three years**, it absorbed **70.2%** of all general fund operating growth, and it now equals **58.9%** of general fund operating spending.

The most pension free city in the study is the most squeezed one in it.

That is worth sitting with, because the pension debate tends to assume the pension is the problem. **The pension is the tail. The payroll is the dog.** Wages, staffing and the market price of a sworn officer drive a municipal budget whether the city employs that officer or rents one from the county. Aurora tells the same story from the other side: its police and fire already consume about **58.8%** of general fund operating spending and took **61.8%** of its growth.

Removing the liability is real and permanent and worth doing. It does not stop the cost.

## What I would actually watch

If you want one number per level of government, these are the ones the documents support.

**For a state**: the ratio of the statutory contribution to the actuarially determined contribution. Idaho pays it and is 90.89% funded. Illinois does not and is 47.4%. Everything else follows from that one habit.

**For a pension plan**: the discount rate, printed next to the funded ratio. Colorado PERA assumes 7.25% and reports 75.31%. Idaho assumes **6.55%** and reports 90.89%, which means Idaho reaches the better number on the harder assumption. Portland's fire and police fund assumes **3.93%** on its own valuation and **5.20%** under GASB, which is why its liability looks enormous next to plans assuming 6.5% to 8.5%. A ratio without its rate tells you nothing.

**For a city**: public safety as a share of general fund operating spending, and the direction it is moving. Not the pension line.

**For a household**: your own city's mill levy as a share of your total bill. In Parker it is 3.1%. In Chicago the city is 24.4% and its pensions are 19.4%. The thing you vote on most attentively is usually not the thing charging you the most.

## A note on how much of this is comparable

Not as much as a chart implies, and I would rather say so than let the chart imply it.

Some of these ratios are GASB accounting measures on market value. Some are funding valuations on smoothed assets. Colorado PERA reports 75.31% on the first basis and 69.1% on the second, a six point spread for the same plan on the same day. Idaho moves 0.3 of a point between them. Measurement dates in this study span 2023 to 2026.

Two plans do not belong on a funded ratio ladder at all. **Portland's fire and police fund is pay as you go by charter** for its legacy members, so a near zero ratio is compliance rather than failure; about a quarter of the bureau's budget now buys prefunded Oregon PERS coverage for members hired after 2006. **Idaho's Firefighters' Retirement Fund is 220.1% funded** because it is closed, has had no active members since 2021, and is running out its obligations against assets that outlived them.

And one structural difference outlives every number here. **Idaho PERSI members participate in Social Security. Most Colorado PERA members do not.** An Idaho public employee holds two independent retirement claims. A Colorado teacher frequently holds one, on a plan that is 71.71% funded with a 1.00% annual increase cap.

That also wrecks the comparison everyone reaches for first. A Colorado state agency pays 21.65% of payroll to PERA against Idaho's 11.96% to PERSI, which looks like 1.81 times. It is not. Colorado pays no employer Social Security tax on those workers and Idaho pays it on top. Counted properly it is **18.16% against 21.65%, about 1.19 times**.

And nearly half of Colorado's 21.65% is not buying a pension at all. **Ten full points are the AED and SAED**, surcharges that exist only to amortise unfunded liability and continue by statute until a division reaches 103% funded. Strip them out and Colorado is paying 11.65% against Idaho's 18.16%. The comparison reverses. Colorado's employer rate is not high because its benefit is rich. It is high because it is servicing a debt.

The honest summary is not that some states are careless and others are prudent. It is that the states at the top kept the ability to adjust something, paid what the actuary asked, and did not borrow their way to a better ratio. The states at the bottom gave up one or more of those, usually decades ago, and are now paying for the decision with interest.

## Fact-check notes and sources

**Colorado**: PERA *2025 GASB Statements Nos. 67 and 68 for Division Trust Funds Valuation* (measurement date 31 December 2025) and the *Actuarial Valuation as of December 31, 2025*; Fire and Police Pension Association *Statewide Retirement Plan Actuarial Valuation January 1, 2026*; Denver, Aurora, Parker, Greenwood Village, Centennial and Breckenridge budgets and annual comprehensive financial reports.

**Idaho**: Milliman *GASB 67 and 68 Disclosure for Fiscal Year Ending June 30, 2025* for PERSI, and the PERSI FY2025 annual report, actuarial section. Note the 90.89% is the **Base Plan**; PERSI system wide including the Firefighters' and Judges' funds is 92.04%.

**Oregon**: PERS 31 December 2024 advisory valuation, which sets no contribution rates. Funded status 73.1% excluding side accounts and 77.2% including, on an actuarial accrued liability of $108,703.0 million with side accounts of $4,400.2 million and an assumed return of 6.90%. Portland Fire and Police Disability and Retirement Fund valuation and city annual report.

**Florida**: FRS actuarial valuation as of 1 July 2025, showing 82.2% on actuarial value and 87.3% on market value, with a $43,250,528,000 unfunded liability at a 6.70% discount rate; the GASB 67 measure at 30 June 2025 is 87.26% and is a different basis. City reports for Tampa, Miami and St Petersburg. Hillsborough County certified millage for tax year 2025.

**California**: CalPERS annual comprehensive financial report; LACERS, LAFPP and Water and Power valuations at 30 June 2025; San Diego City Employees' Retirement System valuation at 30 June 2025.

**Illinois**: Commission on Government Forecasting and Accountability pension briefings and the valuations of TRS, SERS, SURS, JRS and GARS. The **$144.5 billion** figure is from final FY2025 valuations; a widely quoted $144.6 billion comes from preliminary valuations filed earlier. Illinois Constitution Article XIII Section 5. City of Chicago FY2026 Annual Appropriation Ordinance and 2026 Budget Overview; the annual reports of MEABF, LABF, PABF and FABF; Cook County Clerk extended rates for the composite bill. The Chicago Teachers' Pension Fund belongs to the school district and is excluded throughout.

Every figure here was read from a primary document and reconciled against that document's own printed totals. The claims were then put through two rounds of adversarial review whose brief was to refute them. Twenty two claims were tested and twenty two survived, but every single one required a correction first, including one of mine that was wrong by a factor of three. Several statements in this article are narrower than my first draft because of it, and one example, Tampa, was removed entirely.

## Related reading

**[Two states, two budgets](/blog/colorado-idaho-state-budgets-taxpayer/)**: the Colorado and Idaho state comparison this grew out of, including why Colorado's employer contribution rate is smaller than it looks.

**[The city with no pension liability is the one getting squeezed hardest](/blog/colorado-idaho-city-budgets-public-safety/)**: the ten city comparison, and the Centennial result in full.

**[The fire districts nobody votes for](/blog/colorado-fire-districts-outside-city-budget/)**: why a Colorado town can be 3% of your property tax bill.

**[The 2026 entitlement infographics](/blog/entitlement-infographics-2026/)**: the federal programmes these budgets increasingly compete with.

*This post is informational, not legal, financial or investment advice. Pension valuations and appropriations are published figures that change with each valuation cycle. Mentions of named states, municipalities, districts and retirement systems are nominative fair use. No affiliation is implied.*


---

Canonical HTML: https://jwatte.com/blog/six-state-pension-spectrum-what-separates-them/
RSS: https://jwatte.com/feed.xml
JSON Feed: https://jwatte.com/feed.json
Hero image: https://jwatte.com/images/blog-six-state-pension-spectrum-what-separates-them.webp