Not enough public defenders — Oregon dismissed 1,465 criminal cases for want of an appointed lawyer

What is happening?

In February 2026 the Oregon Supreme Court put a clock on criminal cases in which the state cannot find a defense lawyer — 60 days after arraignment for a misdemeanor, 90 days for a felony. 1,465 cases were dismissed under that ruling, 915 of them in Multnomah County and 263 in Washington County. The dismissals were without prejudice, so the charges may be refiled if counsel later becomes available.

The shortage is not confined to one state. In New Mexico 52 of 241 public defender attorney positions were vacant as of July 2026, a 21.6 percent vacancy rate. In Wisconsin 8,445 defendants had waited more than 30 days for a lawyer since 2019, and 12,586 felony cases were backlogged as of 2025-08-01. In Alameda County, California, meeting the national workload benchmarks would take 104 additional attorneys, as reported in April 2026.

The benchmark itself moved. A 2023 national study rebuilt caseload standards that had stood since 1973 and concluded that one attorney can properly handle 59 low-level felony cases a year rather than 150, and 150 low-level misdemeanors rather than 400.

Whose problem is this?

The body that owes the lawyer is not the body that appropriates money for one, and neither of them is the judge holding the case that cannot move.

Where does this problem end?

The boundary matters here because the visible failure is a queue while the invisible failure is a caseload that nobody publishes.

What is the state now, and what should it be?

Now

What it should be — maximum cases per attorney per year, at 2,080 casework hours

Needs a new measurementthe target state is missing at the national level. No source opened here states an official national target for how many people may go without appointed counsel, or a national schedule for closing the attorney gap. Oregon set an agency target of 474 state public defenders, to be reached by hiring 79 a year for six years, and that is one state plan rather than a national one. The 2023 table above is a professional benchmark, not an adopted government target.

How big is it?

Between 2,835 and 4,029 people in Oregon were entitled to court-appointed counsel and had no lawyer, measured on two single days, 2025-05-01 and 2025-11-11. The population file carries the same two numbers.

The number above is a floor and not an estimate of a national total. Oregon publishes a daily count and most states do not, so the number that can be checked is the number from the state that counts.

Two groups are left out. A person who has an appointed lawyer carrying three times the standard caseload is not in this count. Neither is a person whose wait ended in dismissal — in Oregon that was 1,465 cases in one wave. A single-day reading also understates the yearly flow, because everyone who waited and was then assigned a lawyer drops out of it.

For scale in the other direction: at the end of January 2025 Oregon reported 4,178 people unrepresented, of whom 191 were in custody and about 730 had been waiting six months or longer.

Under what conditions does it arise?

Four conditions keep this state stable, and none of them is the absence of a rule requiring a lawyer.

1. Pay sits below the market that competes for the same lawyers. Wisconsin raised the appointed private-bar rate from 40 dollars an hour to 70 in 2020 and then to 100, while the average Wisconsin attorney rate in 2023 was 248 dollars an hour; the state public defender starting salary in 2023 was 56,659 dollars. New Mexico pays contract counsel a flat 220 dollars for a misdemeanor and a flat 6,500 dollars for a first-degree murder case. 2. The standard moved down while volume moved up. The 2023 study cut the felony benchmark from 150 cases a year to between 7 and 59 depending on severity. Over the same period new felony filings in Alameda County rose from 3,266 in 2023 to 4,708 in 2025, an increase of 44 percent. 3. The duty, the money and the deadline sit in three different hands. The constitutional duty falls on the state, the appropriation on an annual legislature, and the deadline on a trial judge who cannot staff the case. The federal tier shows the split in its cleanest form: funding for appointed panel attorneys ran out on 2025-07-03 because a March 2025 continuing resolution froze judiciary funding at the prior year level, and the lawyers already appointed kept working without pay. 4. Declining cases is punished. A California county public defender was held in contempt of court and fined 26,000 dollars after refusing to take new cases one day a week, citing understaffing and a lack of resources to provide due process, as reported in April 2026. Refusal is therefore not a free way for an office to hold its caseload at the standard.

What has been tried?

Oregon is the case that shows the queue does move when money and bodies are added. It is also the case that shows what the queue costs while it moves, because the same period produced the 1,465 dismissals.

What was found?

Why is it still unsolved?

Supply shortage — the number of lawyers willing to do this work at the price offered is smaller than the number of cases, and the party that sets the price is not the party that owes the lawyer.

A shortage of this kind does not announce itself as a shortage. The case is filed, the person is arraigned, and then nothing happens for months while a court clerk works down a list of attorneys. In one Wisconsin county the court attempted contact 10,410 times on a 2023 case and 4,184 times on a 2022 case for the same defendant, who still had no defense attorney more than two years later as of September 2025.

Price is the visible lever and it has been pulled without closing the gap. Wisconsin lifted the appointed rate from 40 dollars an hour to 100 while the general market rate was 248, and still carried 37 unfilled positions, a 10 percent vacancy rate that is better than the 25 percent peak during the pandemic but not near zero. New Mexico, paying flat fees of 220 dollars for a misdemeanor, was at 21.6 percent vacancy in July 2026.

The structural reason is that three decisions live in three places. The duty is constitutional and permanent, the money is annual and discretionary, and the clock is judicial and case-by-case. The federal tier makes this visible without any hiring question at all: the lawyers were already appointed and already working when the appropriation ran out on 2025-07-03, and the judiciary stated that payment would not resume until 2025-10-01 absent supplemental funding. Where a legislature does move, the effect is real but partial — Oregon cut its queue by roughly 30 percent in six months and was still dismissing more than a thousand cases three months later.

What observation would mean it is solved?

Candidates — (a) the count of people entitled to counsel with no lawyer appointed reaches zero and stays there (b) defender offices report caseloads at or under the 2023 national workload standards (c) dismissals for want of appointed counsel stop.

(a) alone is dangerous. It can be reached by handing the cases to lawyers who are already over the standard. Part of the Oregon fall from 4,029 to 2,835 came from paying 42 experienced attorneys to take cases beyond their contracted caseload.

(b) can be reached from the denominator. Charging fewer cases lowers measured caseload without adding a single attorney, which may be good policy but is not the same observation.

(c) is the weakest of the three. Dismissals fall if the deadline rule is narrowed or repealed, and they also fall wherever nobody is counting the days. Read together, (a) and (b) are the pair that cannot be satisfied by moving work around.

What is it connected to?

Fills with researchprosecutor staffing, pretrial jail population and length of stay, trial court backlog, and the general supply of rural attorneys all appear in the same sources, but the relation type and evidence grade were not examined in this round.

What these sources do not say

• How many people nationwide are waiting. The Oregon Judicial Department publishes a dashboard that is updated at 8:30 in the morning each business day and reflects data from the prior day. No comparable national figure was found, and the state figures that do exist are on incompatible bases — a single-day stock in Oregon, a cumulative total since 2019 in Wisconsin.
• How many appointed lawyers are actually over the standard. The 2023 study sets the benchmark, and no source opened here reports the share of offices or of attorneys above it.
• What became of the 1,465 dismissed cases. They were dismissed without prejudice. Whether the charges were refiled, in how many cases, and with what outcome is not reported.
• Whether outcomes differ for those who waited. No source opened here quantifies conviction rate, sentence length, or pretrial jail days for people who spent months without counsel against those who did not.
• Why the Oregon peak figures disagree. A state release gives 4,029 on 2025-05-01 while a March 2026 broadcast gives a May peak of 3,500. Neither explains the gap, and neither states whether the unit counted is people or cases.
• What the shortage costs in money. Jail days, refiling, and repeated hearings are all named in the reporting, and no dollar figure for them was found.

See the evidence

The study itself was not read in full. The publisher returned HTTP 403 to an automated fetch, so the standards table above comes from a September 2023 explainer by a national defender association that reproduces it from the pre-publication Table 7.2, and that PDF was opened and its text read directly. Two claims were confirmed twice: the Oregon dashboard readings of 4,029 and 2,835 appear with the same dates in a state release and in a separate broadcast report, and the 1973 caps of 150 felonies and 400 misdemeanors appear both in that explainer and in the Illinois write-up. One number did not reconcile — the same broadcaster that reports about 1,200 unrepresented in March 2026 gives the 2025 peak as 3,500 where the state gives 4,029, and neither source says whether the unit counted is people or cases, so the March 2026 figure was left out of the population file and is recorded here only as a claim. No source opened here states the date on which the 2023 study was released; the two dates that appear above, 2023-09-12 and 2023-10-20, are the posting dates of the pages that report it rather than a release date, and the publisher could not be reached to settle the point. This is a research-based definition rather than a record derived from one dataset, so observation_refs is empty and provenance_mode is press-derived.

People affected

Estimated range 2,835–4,029 As of 2025-05-01 to 2025-11-11

Sensitivity The interval is not a confidence interval. It is the observed range between two single-day readings of one state dashboard, ten months apart, and each endpoint is a census rather than a sample. The number is a floor in three directions. First, it covers one state: no national count of people waiting for appointed counsel was found, and Oregon appears to be the state that publishes a daily figure, so the number that can be checked is the number from the state that counts. Second, it counts only people with no lawyer at all and therefore excludes the much larger group who do have appointed counsel but whose attorney carries a caseload above the 2023 national workload standards, which cap low-level felonies at 59 a year rather than the 150 assumed since 1973. Third, a single-day stock understates the annual flow, because everyone who waited and was then assigned counsel, and everyone whose case was dismissed for want of counsel, drops out of the count. A later press report puts the Oregon figure near 1,200 in March 2026, but the same report gives the 2025 peak as 3,500 where the state dashboard gives 4,029; the basis of that pair could not be reconciled, and neither source states whether the unit is people or cases, so the 1,200 was not used as the lower bound.

Regional breakdown County counts exist for Oregon in a single month of early 2025 — Multnomah 1,172, Marion 756, Jackson 754, Washington 628, Douglas 253, Coos 99 — but those are counties inside one state, not states measured against a United States total on the same basis. No source opened here gives state-level counts summing to a national figure. Splitting a national number by resident population would be proportional allocation, which is not permitted, and it would be especially baseless here because the shortage tracks state appropriations, pay scales and court rules rather than population.