The one-sentence version. The system measures and funds inputs — cases filed, charges stacked, dispositions cleared, per-case fees paid — and almost never measures the outcome that matters: did the person stop cycling through? What you pay for is what you get.
1. The standard: outcomes, not cases
A “case” is an input. It is opened when a charge is filed and closed when a disposition is entered. You can generate more of them, stack more charges onto each one, and clear more of them — and none of that tells you whether justice was done, whether a victim was made whole, or whether the defendant ever came back.
An outcome is what actually happened to a person and a community after the system was done with them. Did the cycle stop? Did the underlying problem — addiction, housing, mental health, an unpaid fine that snowballed into a warrant — get addressed, or just processed? We measure the system by outcomes, because that is the only measure that can tell the difference between working and merely being busy.
The system's incentives reward generating and maintaining caseloads, not resolving them. The money — flat per-case defense fees and caseload-justified court budgets — follows volume, not outcomes. The result is a churn that financially sustains the actors while the defendants' underlying problems go unaddressed, and they cycle back to generate the next case.
2. The Medicare parallel
This is not a novel idea; it is a well-documented failure mode. When Medicare cut the reimbursement paid per service, doctors did not earn less — they ordered more services. The unit of payment was the service, not the patient's health, so the system produced more services.
Whatcom's justice system has the same design. The unit of payment is the case — a flat fee per assigned defense case, a court budget justified by caseload volume. Lower the price per case, or raise the pressure to clear them, and the system does what it is paid to do: it generates and maintains cases. The thing no one is paid to produce — a person who never comes back — is the thing the system does not produce.
3. What a “case” hides
Counting cases makes hard work look like results and makes gamesmanship invisible. Two mechanics recur in the record:
Over-charge, then dismiss
Charges are stacked at filing — inflating the apparent workload and, for fee purposes, the size of the matter — and then quietly dropped. In one documented protection-order matter, a majority of the counts were dismissed on the city's own motion. The charge count was theater; the resolution was the reality. If you count cases and charges, this reads as diligence. If you count outcomes, it reads as inflation.
Clear the case, not the problem
A case can be “resolved” on paper — a plea, a dismissal, a failure-to-appear warrant issued — while the person walks back into the exact circumstances that produced the charge. The case is closed. The outcome never changed.
4. Why repeat defendants are the “best clients”
In a volume system, a person who cycles through fifteen times is not a failure — they are recurring revenue. Every re-arrest is a new booking, every new charge a new case, every new case a new fee or a new line in the budget justification. There is no line item for “never came back.” A system honestly measured by outcomes would treat that fifteen-time cycle as its single loudest alarm. A system measured by cases treats it as steady business.
5. The metrics we use instead
Wherever the data lets us, we replace case counts with outcome measures:
- Recidivism / re-booking rate — the share of people re-booked within 6, 12, and 24 months, and how fast. This is the direct test of “did it stop?”
- Cost per cycling defendant — not dollars per case, but dollars spent on people who still re-offended. This is the Medicare-waste figure: money that bought activity, not outcomes.
- Cost and spend per unique person — billing ledgers exploded and de-duplicated by person, so “$X for N cases” becomes “$X for M unique people, re-booked K times.”
- Appearance and resolution — how often scheduled hearings are actually held versus failed, and how many hearings a single matter consumes before it truly ends.
These are built from public records — court dockets, jail booking histories, and government billing ledgers — joined at the level of the person, not the case.
6. The discipline we hold to
An outcome-based argument is only as credible as its restraint. Our rules:
- Structure and results, not motive. We document how the incentives are built and what they produce, and we let the reader draw conclusions about intent. We do not assert that anyone runs a “racket for profit” — that is a claim about a person's state of mind, and we do not make it.
- People are protected; officials are named. We never publish the name or re-identifiable record of a defendant or private individual. Judges, prosecutors, defense attorneys, and public officials — who act in a public capacity — are named. Case studies are composites or de-identified.
- “It's already public” is not our license. When the court points to public records to justify exposing individuals, we reject that argument; we do not turn around and adopt it ourselves. Compilation destroys practical obscurity, so we aggregate.
- Resistance to measuring is itself evidence. When an institution funded by caseload volume declines to report the outcomes that volume is supposed to buy, that silence is a finding. You do not hide the scoreboard if the score is good.
7. The sources
The reporting on this site is built from primary public records, not press releases:
- Court dockets and minute entries (the official record of whether hearings were held or failed).
- Jail booking histories, aggregated by person to measure re-booking.
- City of Bellingham and Whatcom County billing ledgers for indigent defense.
- Adopted budgets and staffing tables, public-records disclosures, and court filings.
Where a figure depends on a judgment call — a definition, a match between records, a coverage gap — we say so on the page that uses it. If we get something wrong, we correct it in place.