Independent public-records project. Not an official State of Colorado system.
COLORADO WORKS BETTER
Taxpayer Shareholder Portal · Weekly brief
Weekly brief · 27 August 2026

Colorado’s pension gap moved two ways at once

Every figure here is read from a published document, or worked out from figures that are. Each one is checked again from this page every time it is built.

Two measures of one gap, moving opposite ways

Colorado promises a pension to teachers, state workers, troopers and judges. The fund that pays it is PERA. Once a year PERA works out how far short it is of the money those promises will need, and this week it published that work for the end of 2025.

On the measure PERA uses to set contribution rates, the shortfall grew. It was $28.91B a year earlier and is $30.07B now. The share of the promise already funded barely moved, from 69.2% to 69.1%.

On the measure that appears in the state’s audited accounts, the same shortfall fell, and it fell a long way: from $27.99B to $24.19B.

Both figures are PERA’s own, for the same funds, on the same date. They differ because they value the same investments differently. The accounting figure uses what those investments were worth on the day. The contribution figure spreads gains and losses over several years, so it moves slowly and lags the market. When markets move sharply the two measures come apart, and this year they did.

We publish both and we do not subtract one from the other. Neither is the real number with the other as an error, and a series that mixed them could show a rise where there was a fall. Which basis a figure is on is written beside it every time it appears, on the methodology page and wherever the figure is shown.

The gap grew, and that is not unusual

A figure is easier to read against its own history than on its own. Both pension measures below are shown against every year we hold, with the latest move ranked among the moves before it. The rank is arrived at by counting. It is not a forecast and it is not a judgment about whether the number is good.

The pension shortfall on the contribution basis rose from $28.91B to $30.07B in 2025.

Four years have risen further since 2016.

The share of the promise already funded fell from 69.2% to 69.1% in 2025.

Three years have fallen further since 2016.

What a household pays is now on the site

Until this week this site was mostly about what the state spends. It now also carries what living here costs.

A year of infant care at a center was $20,978. The median rent in Colorado was $1,822 a month. Resident tuition at the University of Colorado Boulder went from $12,312 to $13,566. The state minimum wage is $15.16 an hour.

The school results are there too, and they are not comfortable reading. Chronic absenteeism was 27.7% and is now 28.4%. Chronically absent means missing about a tenth of the year or more, for any reason.

Each of these comes from a publisher who states it. None of it is our estimate.

The legislature recorded no votes this week

This section carries every vote the General Assembly recorded in the week the brief covers, with the tally the legislature printed for each one.

The General Assembly recorded no votes between 21 August 2026 and 27 August 2026. Its most recent recorded vote was on 13 May 2026.

A note on the checking

Every figure above comes from a published document. Each one is read again from this page every time the page is built, and compared against the dataset it came from.

A new valuation replacing an old one is not a mistake by us, and it is not logged as a correction. Things we got wrong are, and this week the corrections log gained entries on four separate days. One of them is worth reading on its own: a chart comparing Colorado’s spending with the 50-state average was drawing its bars too short, so no figure was wrong and the picture still disagreed with them.

Where these figures come from

Every figure above is one of two kinds, and this is which. FACT means a publisher printed the figure and we read it there. CALC means we worked it out from figures a publisher printed, and the arithmetic is on the methodology page. Nothing here is estimated. The vote counts are a third kind and are not listed below. Each one is the tally the legislature printed on its own vote page, which the table links.

  • FACT The pension shortfall on the contribution basis · 2024 and 2025 · Colorado Public Employees’ Retirement Association, Colorado PERA Annual Comprehensive Financial Report Back to the sentence
  • FACT The share of the promise already funded · 2024 and 2025 · Colorado Public Employees’ Retirement Association, Colorado PERA Annual Comprehensive Financial Report Back to the sentence
  • FACT PERA net pension liability (accounting basis) · 2024 and 2025 · Colorado Public Employees’ Retirement Association, Colorado PERA Annual Comprehensive Financial Report Back to the sentence
  • FACT Annual price of infant care, center (Child Care Aware of America) · 2024 · Child Care Aware of America, Child Care Affordability in Colorado — 2024-25 price fact sheet (from "Child Care in America: 2024 Price & Supply Report") Back to the sentence
  • FACT Median gross rent, Colorado · 2024 · U.S. Census Bureau, ACS 1-year, table B25064 — median gross rent, Colorado (data.census.gov table API) Back to the sentence
  • FACT Resident tuition, University of Colorado - Boulder · 2024-25 and 2025-26 · Colorado Department of Higher Education, Finance Division, Tuition and Fees Report — Tuition and Fees for the 2025-2026 Academic Year Back to the sentence
  • FACT Minimum wage, Colorado · 2026 · Colorado Department of Labor and Employment, Division of Labor Standards and Statistics, Adopted PAY CALC Orders, 7 CCR 1103-14 — 2024, 2025, 2026 (plus the 2026 temporary order) Back to the sentence
  • FACT Chronic absenteeism rate · 2023-24 and 2024-25 · Colorado Department of Education, Data Services, 2023-2024 Attendance Rate, Truancy Rate and Chronic Absenteeism by State, Grade, IPST Back to the sentence
← Back to the Taxpayer Shareholder Portal
Weekly brief · 27 August 2026 · Colorado Works Better. Every figure on this page is substituted at build time from the same validated dataset the dashboard reads, so a brief cannot state a number the pipeline does not carry. Corrections are logged in the corrections log.

How to read our labels

FACT Taken directly from an official public record. CALC A labeled formula applied to official records. SAMPLE Illustrative placeholder pending source ingestion. PENDING Real field, left blank rather than estimated — because the publisher does not report it, we have not ingested it yet, or sources disagree and are being reconciled in the open.

WorksBetterThe Honesty Framework(opens in a new tab)PrivacyTerms of UseAccessibility© 2026 Works Better Corporation, a nonpartisan for-profit Delaware public benefit corporation. Our writing, design and analysis only.