Twenty-Six Minutes

Austin City Council begins final adoption of the fiscal year 2026-27 budget on August 12, and the vote is scheduled to run through August 14. The proposal increases the Austin Police Department budget by about $24 million, bringing it to roughly $550 million. The same proposal cuts $5.4 million from social service contracts and eliminates 24 of them outright, the first installment of a $16.8 million reduction target city staff have spread across two budget years. The Travis County Sobering Center loses roughly $2.9 million in ongoing city funding. The Summer Youth Employment Program, funded at about $652,000, is eliminated.

The department published its own line-item accounting of that increase on August 4, in response to a budget question from Council Member Mike Siegel. The largest piece is a 5 percent base wage increase for all sworn officers. The rest is pension, health insurance, facilities and technology allocations, fleet maintenance, and $2.5 million in additional Liability Reserve, which is the money held against claims.

None of that is a public safety intervention. It is the cost of keeping an existing institution running, which is a legitimate thing for a city to pay and an entirely different thing from what people mean when they say a budget invests in public safety. When the research on policing gets cited in defense of this increase, and it will be, it is being cited in defense of a wage settlement.

The increase to the Liability Reserve alone, the money Austin holds against claims arising from the conduct of its own employees, is roughly half the size of the entire social services cut.

People will argue about this the way they always do. Police or social services. Enforcement or prevention. One side will say Austin tried defunding the police and it failed, and the other will say policing has never made anyone safer.

Neither argument is particularly useful. The evidence says police can reduce serious violence. It also says housing, mental health services, youth programs, and other interventions can prevent the crises that eventually become police calls, arrests, jail stays, emergency room visits, and victims.

But Austin is not making those investments on a level playing field. Texas law places a floor under the police budget that does not exist under social services, and that changes the question.

Council does not have to settle which system deserves to exist. It has to decide what the next dollar buys in each one, and what happens when that dollar is taken away.

The case for police, made properly

The strongest evidence for policing is real, and there is no reason to dodge it.

For about twenty years, economists have studied what happens when cities hire more officers, using federal hiring grants as the source of variation. Evans and Owens examined the Clinton-era COPS program, Mello the 2009 stimulus round, Chalfin and McCrary the measurement error in the underlying data. The estimates vary in size and point the same direction: larger forces are followed by less crime, with violent crime more responsive than property crime. A 2022 paper by Chalfin, Hansen, Weisburst, and Williams added race-specific estimates, finding that each additional officer prevented roughly a tenth of a homicide, with the reduction running about twice as large per capita for Black victims. The same paper found more low-level arrests, falling disproportionately on Black residents, and that cities with larger Black populations did not receive the protective benefit equally.

I take all of that seriously. If Black lives lost to homicide are a public safety emergency, and I believe they are, then a body of research showing that officers reduce those deaths belongs in my argument rather than somebody else's.

All of that research estimates the effect of hiring officers. That is not what Austin is being asked to fund, and I am not proposing that it should be. Austin already cannot staff the positions it pays for, and I would put the money elsewhere in any case.

The point is narrower, and it cuts against the argument this evidence usually gets used to make. The studies describe hiring. This budget funds a wage settlement, a pension payment, and a facilities allocation. Whatever the literature establishes, none of it transfers to a line item nobody has claimed will change anyone's safety.

Police respond downstream

Police are a downstream institution. Something has happened, or somebody believes something is about to, and the city responds. Some of those situations require an emergency response and legal authority a social worker does not have.

Prevention works earlier, and one crisis prevented can eliminate several later costs. Housing somebody who has been arrested repeatedly does more than put a roof over their head. It can also eliminate future police calls, citations, arrests, jail bookings, shelter stays, detox admissions, and emergency medical responses.

The evidence for this is unusually strong. The Urban Institute studied 724 people experiencing chronic homelessness in Denver, all of them extremely high users of police and jail services. Participants had been arrested at least eight times during the previous three years. Researchers randomly divided them into two groups. One group was offered permanent supportive housing. The other continued receiving services normally available in the community.

Three years later, the people offered supportive housing had:

• 34 percent fewer police contacts.
• 40 percent fewer arrests.
• 30 percent fewer jail stays.
• 27 percent fewer jail days.
• 40 percent fewer shelter visits.
• 65 percent fewer detoxification visits.

Six categories of government activity fell at once, around a small group of people.

The limitations matter too. Only 79 percent of those offered housing were located, engaged, and housed, the reduction in emergency medical service use was not statistically significant, and the study population was unusually high-utilization.

So the finding is narrower than the slogan and more useful. Permanent supportive housing substantially reduces police, jail, shelter, and detox use among people who cycle repeatedly through all four, which is exactly the population Austin spends enormous sums responding to after something has already gone wrong.

Austin has already tested another approach

We do not have to go to Denver for everything. Austin has already run one of the cleaner local experiments in public safety.

When someone calls 911 here, the operator can ask whether the caller needs police, fire, EMS, or mental health services. The last option matters. A clinician can help triage the call, resolve it by phone, or send behavioral health responders from Integral Care rather than automatically sending a police officer.

Austin's Chief Medical Officer Mark Escott reported that 86 percent of eligible calls diverted to the collaborative mental health response between 2022 and 2024 were resolved without police involvement. Researchers from the LBJ School of Public Affairs and Cornell University later examined more than 20,000 mental health related calls and found that using mental health responders reduced police time on scene by more than half. The average reduction was about 26 minutes per call.

That may be the most important number in this entire debate.

Twenty-six minutes.

Start with what those minutes are for the person on the other end of the call. Twenty-six additional minutes of an armed officer standing inside a psychiatric emergency is not neutral time. It is time in which a frightened person can be restrained, arrested, hospitalized against their will, or killed. Shortening that exposure is the reason the program matters, and in Austin most of those calls now end without any police involvement at all.

There is a second argument, and I make it because it reaches people the first one does not, on premises I do not hold. Somebody who believes police capacity is what produces safety should want this program expanded, because every crisis a clinician takes is an officer returned to other work. Austin has moved tens of thousands of cases away from law enforcement this way.

I do not believe police capacity produces public safety. Housing, income, treatment, and the presence of adults who know a young person's name produce public safety. But the diversion program is worth defending on either account, and an argument that works on somebody else's premises is still worth making to somebody else.

Anyone preparing to call that anti-police has it backwards. The program exists because armed response is the wrong instrument for a mental health emergency. The staffing effect is a consequence of that, not the purpose of it.

Travis County shows the same pattern from the jail side, where Dell Medical School researchers report that at least 40 percent of people held have a mental health diagnosis. In the county's diversion pilot with Integral Care, 94 percent of participants had not been arrested again 30 days after entry, 72 percent had not gone to an emergency department, and 66 percent had not experienced another crisis episode. Early numbers from a young program, and worth treating as such, pointing where the rest of this points. People enter police and jail systems because another system failed first.

What prevention can actually do

Youth programs are another place where advocates sometimes overstate the evidence. There is good evidence. We do not need to exaggerate it.

Researchers studied Chicago's Becoming a Man program through randomized trials. Early studies found total arrests fell between 28 and 35 percent while participants were in the program, violent crime arrests fell between 45 and 50 percent, and graduation improved. Later research pooled four trials. The combined estimate was smaller but still substantial, with violent crime reductions reaching as high as 37 percent, and educational gains more modest than the earliest studies suggested.

The finding survived the scrutiny. An inexpensive youth intervention produced measurable reductions in violent offending, at a return of roughly two to ten dollars for every dollar invested. Summer employment programs show something similar: supported summer jobs repeatedly reduced violent crime arrests, including after the summer ended, without consistently improving later employment or schooling, and with some property crime increasing.

The finding matters because programs do not always work for the reasons we think they work, and the question worth asking is whether they produce the outcome we are paying for. If a summer jobs program reduces violent behavior, that is a public safety outcome whether or not it turns every teenager into a permanently employed adult. Which is why eliminating Austin's Summer Youth Employment Program deserves more scrutiny than treating $652,000 as a line item that can disappear quietly.

Community violence intervention deserves more caution. Programs modeled on Cure Violence have reduced shootings in some neighborhoods and failed in others. Implementation matters. Targeting matters. Local conditions matter.

Funding something because it carries a progressive label makes no more sense than funding something because it carries a badge. The standard should be the same. Does it reduce harm?

What Baltimore proves, and what I will not recommend

Baltimore launched a Group Violence Reduction Strategy in its Western District in January 2022. The approach focused on a small number of people believed to be driving a disproportionate amount of group-related gun violence. Those individuals were offered services and support. They were also told clearly that continued shootings would bring focused legal consequences.

Independent researchers Max Kapustin and Aaron Chalfin estimated that homicides and nonfatal shootings fell by roughly one-third during the first 18 months, and carjackings fell about 40 percent. Violence did not move into neighboring districts. Overall arrests did not increase, and neither did low-level arrests.

The last finding is the one that matters here, and it is narrow. A large reduction in homicide occurred without an increase in arrests. Whatever produced it, volume of enforcement was not the mechanism. That single fact does more damage to the case for a larger police budget than any argument I could construct, because it comes from the same researcher whose work is cited to support one.

I am not going to recommend that Austin build one.

Focused deterrence works by making the threat of coordinated prosecution credible and immediate. It requires police, prosecutors, and outreach workers operating as a single apparatus, and what it builds is a more efficient machinery of punishment aimed at fewer people. Mariame Kaba and Andrea Ritchie describe this kind of intervention as the sort of reform that extends the life of the thing it improves, and on this one they are right. A strategy that reduces shootings by tightening the coordination between a police department and a district attorney has expanded the state's capacity to punish while lowering the number of people it punishes. The second half is real. So is the first.

Reform is worth doing when it moves toward replacement. Reform that entrenches an institution further while performing improvement is how the institution survives scrutiny, and I have watched enough of it in reentry and behavioral health to recognize the shape.

So I take the finding and decline the model. What Baltimore establishes is that the homicide benefit is not a simple function of how many officers a city employs. What Austin should build from that is a question I would put to the people who live where the shootings happen, not to a partnership of agencies that would design it for them.

Every system is built to produce what it produces

There is an old line among people who work on systems, usually credited to Arthur Jones and repeated by Paul Batalden, that every system is perfectly designed to get the results it gets.

I have found nothing more useful for reading a budget.

Austin's public safety spending produces what it is arranged to produce. It generates arrests, because arrests are what the largest department is built to make. It generates jail days, because the jail is the receiving end. It generates emergency room visits at three in the morning, because at three in the morning the emergency room and the jail are the doors that open. None of that is a failure of the system. It is the output the arrangement was designed to yield, and it will keep yielding it regardless of what any department says it values.

Consider how the money is sorted. One column holds police. Another holds homelessness services. Another holds behavioral health, and another youth programs, and another the jail, which is the county's column and not the city's at all.

Those categories are administrative. They are not real in the way the budget pretends.

The person in the diversion call at three in the morning, the person in the jail bed the following week, the person on the housing queue, and the person generating repeat calls in the police line item are frequently the same person, moving through departments that account for them separately and bill them to different funds. Austin does not have a homelessness problem and a behavioral health problem and a public safety problem. Austin has people whose lives cross every one of those lines, encountering a government organized to see only one line at a time.

Which is why the pyramid matters more than any single line item.

Picture public safety as a shape with a wide base. At that base sit housing, income, education, health care, behavioral health, youth development, treatment, transportation, and the ordinary density of relationships in a neighborhood where people know each other. Those conditions determine how many people ever reach a crisis. Above them sit the targeted interventions: crisis response, street outreach, intensive case management, domestic violence services. Above those, emergency services. And at the narrow point, the small set of situations that require somebody with legal authority to use force.

Austin has built the shape upside down, and so has nearly every American city. The narrow point carries $550 million. The base is where the $5.4 million came from.

The Sobering Center is the clearest illustration in this budget. It exists so that somebody too intoxicated to be safe has somewhere to go besides a jail cell or an emergency room bed. It is the receiving end that makes diversion possible, which is another way of saying it is what has to exist before an officer can decline to make an arrest. The proposal takes $2.9 million out of it in the same document that adds $24 million to the department whose alternative it is.

Where I stand

I should say plainly what I am, because it changes how the rest of this ought to be read.

I am an abolitionist. The word carries a caricature I want to set aside first, which is that police vanish on a Tuesday and nothing takes their place. That has never been the argument. Abolition, as I understand it and as I practice it, begins with a question rather than a demand. What produces harm, and what would have to change for less of that harm to occur? Removing a coercive system without building the capacity it was carrying leaves a vacancy, and vacancies get filled by whatever is nearest.

I think the institution is flawed past the reach of any training curriculum. Austin's own department descends from a City Marshal whose duties included enforcing the city's slave curfew, and over the following century the job accumulated responsibilities nobody designed it for, from truancy to mental illness to homelessness to school discipline. Angela Davis has spent decades asking a question that budget season never asks: what was this institution built to do, and for whom. A department that grew out of controlling a population, and then absorbed every social failure the city had no other answer for, is not a neutral instrument that happens to be misapplied. The misapplication is the history. None of that entitles me to misread evidence. The Chalfin findings are inconvenient for the position I hold, and they are still the findings. I could discount the study because I dislike where it points, and plenty of people who share my politics do exactly that. An argument that requires me to suppress a result will not survive contact with anyone who has read the literature, and it should not.

So here is the position, including the part I cannot prove.

I accept that additional officers reduce homicide. I also believe the aggregate harm policing does to the communities that absorb most of it exceeds that benefit. I cannot demonstrate the second claim the way the first has been demonstrated, and I will not pretend the two rest on equal footing.

The difficulty has nothing to do with the harms being unstudied. They are studied well. There is a substantial literature on what an arrest does to employment and housing, on the health consequences of police stops, on what a parent's incarceration does to children, on deaths during police contact. The benefits are studied well too, and reported in a common currency: crimes prevented, homicides abated, dollars of victimization avoided.

What I have not found is the subtraction. The cost-benefit work in this field prices an officer against crimes avoided and stops there, leaving the harm column outside the arithmetic. If somebody has published an accounting that puts both sides in the same units and computes the difference, I want to read it, and I will say so here when I do.

Until that exists, my claim and its opposite are the same kind of claim. The person who tells you policing nets out positive is making a judgment about a subtraction nobody has performed, exactly as I am, usually without mentioning it.

What follows from that is not paralysis. If the ledger cannot be netted, the sensible course is to keep the parts shown to work, replace the parts shown to harm, and build the alternatives that make replacement possible.

That last clause carries a condition I hold strictly. Reform counts when it is a step toward replacement. A reform that makes an institution more tolerable, better trained, more efficient, and therefore more permanent has not moved anything. It has bought time. The test I apply to any proposal in this piece is whether it shrinks the territory policing occupies or makes the occupation more comfortable, and I would rather lose an argument than pass the second thing off as the first.

What I want is not exotic. I want a woman walking to her car at two in the morning to be unafraid. I want a mother's son to come home. I want a person in psychosis to be met by somebody trained to sit with them rather than by somebody carrying a weapon and a stack of other calls. I have not met many people who want something different from that. Almost every disagreement here is about how, and almost every one of them gets argued as though it were about whether.

Getting there requires capacity Austin does not have yet. Until it exists, people have to survive the system we actually have. Somebody is walking out of the Travis County Jail on a Wednesday morning with a paper bag and a bus pass. Somebody is sleeping behind a building tonight. They cannot wait for the better system, and I'm not willing to ask them to.

So I hold both. Build the replacement, and keep people alive while it is being built. Nearly everything I am asking Council to do in this piece is survival work inside a system I would like to see replaced. Nobody reaches the better system by failing to survive this one.

Holding both does not mean writing the Police Department a blank check. I support funding police functions that produce improved safety, and defunding the ones that do not. The test I would apply to a nonprofit contract applies here, and it should be applied to named units and practices rather than to the department as an abstraction.

Austin's budget is not a level playing field

This is where the current Austin debate becomes different from the debate we had in 2020.

In 2021, the Texas Legislature passed House Bill 1900. The law applies to cities with populations above 250,000. If a city reduces police appropriations from one fiscal year to the next and the governor's Criminal Justice Division determines that the city has "defunded" its police department, serious penalties can follow. The city can face restrictions on property tax revenue. State law can divert sales tax revenue to cover state law enforcement costs. The city's annexation authority can be restricted. Areas annexed during the previous 30 years can receive an opportunity to vote on disannexation.

There are exceptions in the law, including provisions involving overall budget reductions and certain approved adjustments. But as a practical matter, the statute makes reducing a large city's police budget extraordinarily difficult. Austin's police appropriation, $525.4 million as amended for the current fiscal year, functions much like a floor.

I want to be direct about the law itself. House Bill 1900 is bad policy, and the damage it does runs well past the police budget.

The protection compounds. Each year's appropriation becomes the following year's floor, and the floor ratchets upward. A department that receives a large increase during a strong revenue year carries that increase into every lean year afterward. Austin's budget office projected a $26.4 million deficit for fiscal year 2026-27. A rising floor inside a shrinking budget means every dollar of shortfall has to come from somewhere with nothing underneath it, which is exactly what happened after Proposition Q failed.

What the law does to evaluation is the part I find hardest to accept. A budget that cannot go down has a limited ability to respond to evidence. If Austin studies its police department and finds a unit producing nothing, the finding cannot return a single dollar to the rest of the city. Measurement disciplines an institution when it can move money, and the statute permits that money to move only inside one department's walls. Every other department in Austin gets evaluated against the possibility of losing funding. One does not, which means the least examined spending in the city is also the best protected.

The law punishes reorganization it was never aimed at. Moving forensic services, victim services, or 911 call-taking out of a police department and into a civilian department lowers the police appropriation on paper while the work continues unchanged. Austin's interim police chief, Joseph Chacon, testified against the bill in 2021, telling legislators that moving money out of his budget did not mean the function had stopped, and that dollar figures do not always track with public safety. The chief of the department the bill was written to protect asked the Legislature not to pass it.

And every one of those penalties lands on people who made no decision at all. None of it falls on a council member. It falls on residents, including the residents who wanted the police budget left exactly where it was.

The floor was a policy choice made in 2021 by legislators who wanted this particular city to stop making its own. It deserves to be named as a choice rather than treated as weather.

What the law does not govern is the composition of that appropriation. Nothing in Texas statute requires Austin to keep paying for a school policing contract, to maintain militarized capacity for routine warrant service, or to fund the enforcement activity that generates low-level arrests without touching homicide. Where a police function produces no public safety gain, the money behind it can move to a function that does. It can move toward homicide clearance, toward the forensic backlog, toward victim services, toward putting a clinician beside an officer on the calls that need both.

Anyone who says Austin cannot cut police spending is right about the total and wrong about everything inside it.

There is no equivalent floor underneath social services, and that difference matters enormously.

After Proposition Q failed in November 2025, Austin had to remake its budget. Social services took reductions. Parks took reductions. Homelessness programs took reductions. Other city functions took reductions. The police budget did not suddenly become equally available for reconsideration, because state law does not treat it the same way.

Now Austin is entering another budget cycle. The proposed police budget rises by about $24 million to roughly $550 million while social service contracts take a $5.4 million reduction and 24 eliminations.

There is a live decision attached to that number. After the Travis Central Appraisal District certified the tax rolls, the city found $4.7 million more in property tax revenue than it had projected. Council members Zo Qadri and Mike Siegel have signaled support for using it to restore the social service funding, and nonprofit staff and clients filled council chambers last week to ask for exactly that.

The unexpected revenue covers all but about $700,000 of the cut. The police increase is roughly five times the entire cut. One of those numbers is being debated in public this week, and the other arrived as a given.

Broken out, the increase is an $11.7 million wage settlement at full year under the Meet and Confer agreement, $4.5 million more for facilities management support, $3.7 million more for city technology, $2.5 million more for the retirement contribution, and assignment pay, health insurance, fleet, administrative support, and payroll taxes for most of what remains.

The department is not proposing to grow. It proposes to pay more for the force it already has, and to cover the citywide allocations charged against it. Anyone describing this budget as an investment in more policing is describing a raise, a pension payment, and an overhead transfer.

There is a further number in that document worth sitting with. Against $210.8 million in civil service wages, the proposal carries a personnel savings line of negative $33.8 million. That is the department crediting back the money for positions it does not expect to fill. Roughly a sixth of sworn payroll, budgeted in advance as unfillable.

The same increase also funds expanded recruitment. Austin is spending money to fill positions it has already conceded, on the facing page, that it will not fill. Savannah Lee of Equity Action made the sharper version of this point in July, noting that the generous police contract Council approved in 2024 was itself sold as the fix for staffing, and that advocates warned at the time it would come out of future social services budgets. Equity Action publishes an alternative spending plan each year called the Community Investment Budget, which is worth reading beside the city's own.

Meanwhile the overtime lines tell you where the work is actually going. The current-year estimate for civil service overtime is $33.6 million against $26.0 million budgeted, with reimbursed overtime running $8.4 million against $2.5 million budgeted. Austin will spend something near $44 million on overtime this year and has budgeted about $30 million for next year, which is close to the figure it has already exceeded.

That is the structural reality behind this debate. One budget has a floor. The other does not.

What I would ask Council to do

Saying "move police money into social services" describes where someone wants policy to go. It does not describe what Council can do in August 2026. Three tools remain, and all three are available next week.

The first is the size of the increase, which is still a policy decision even when the direction is not, as is where new revenue goes.

The second is shared responsibility. Austin's own budget analysis raises a legitimate point: the city carries a larger share of social service spending than some peer jurisdictions and taxing entities. Fine. Then turn the observation into a plan. Homelessness, behavioral health, substance use, and crisis response do not stop at the city limit, and Austin, Travis County, Central Health, and Integral Care should hold durable agreements about who pays for what rather than leaving nonprofits to fight for one-year contracts every summer.

The third is the simplest. Do not cut a program with evidence behind it without calculating what losing it costs somewhere else.

Before approving a social service reduction, estimate what the reduction is likely to do to police calls, EMS responses, jail use, emergency department visits, shelter use, and other city costs. Maybe the estimate shows that a program really is not producing enough value. Then cut it. But if Austin saves $500,000 on one contract and creates $900,000 in police, jail, EMS, and hospital costs somewhere else, that is not fiscal responsibility. It is moving money between columns.

Restore the $652,000 Summer Youth Employment Program. Youth employment does not guarantee that a young person will never commit a crime, and nobody should hold it to that standard. Randomized studies have repeatedly found reductions in violent offending from supported youth employment programs, and for $652,000 in a $6.6 billion city budget, that deserves serious consideration.

Put the composition of the police budget in front of a community body with actual authority. The total is fixed by state law and the composition is not, which makes internal reallocation the one lever available to Council that requires nothing from the Legislature. What I will not propose is that the department rank its own units. No institution empowered to use force against people should be the body that evaluates whether that force produces anything, and a department asked what it would give up will name what it was already prepared to lose. The evaluation belongs to a community oversight structure with subpoena power, budget authority, and members chosen by the neighborhoods that absorb the most enforcement. Anything less produces a report, and Austin has a shelf of those.

Keep expanding mental health diversion. It works here, and it works first for the person in crisis, who meets somebody trained to sit with them rather than somebody carrying a weapon.

Protect permanent supportive housing and the services that make housing successful. Housing should be measured not only by how many people receive apartments, but by what happens afterward. Police contacts. Jail days. Emergency department visits. Housing retention. Repeat homelessness.

Fund community violence intervention run by organizations rooted in the neighborhoods where the shootings happen, with the design authority held there rather than by a partnership of agencies. The evidence for this work is uneven, which I said plainly earlier and will not soften here. Implementation and credibility determine the results. Both are properties of who is doing the work, which is an argument for putting the money and the decisions in the same hands rather than for withholding them until a cleaner study arrives.

Measure safety, not activity

This may be the change I care about most.

Police departments are often measured by activity. Calls answered. Arrests. Citations. Response times. Social services are often measured the same way. People served. Beds filled. Referrals made. Appointments completed.

Those numbers tell us what government did. They do not necessarily tell us whether anybody became safer.

Put both systems on the same scoreboard. Homicides. Nonfatal shootings. Reported victimization. Domestic violence injuries. Repeat crisis calls. Jail days. Emergency detentions. Emergency department transports. Police use of force. Housing retention after one year and three years. Youth violent offending.

Then ask what each public dollar bought. A police unit that makes fewer arrests while shootings decline is succeeding. A homelessness program that houses fewer people but dramatically reduces repeat jail and emergency use may also be succeeding. A nonprofit that serves 10,000 people without changing any measurable outcome may not be.

The badge should not protect a program from evaluation. Neither should the word "community."

I could be wrong

I want these claims to be testable.

If Austin maintains or expands mental health diversion and police time spent on mental health calls stops declining, then part of my staffing argument fails. If Austin builds a focused violence reduction program, runs it competently for several years, and shootings do not move, then Baltimore may not translate here. If permanent supportive housing expands and repeat police contacts and jail stays among participants do not decline, then the Denver findings may not describe Austin's population. If restoring youth employment produces no measurable effect on violence, school attachment, employment, or other agreed outcomes, then we should reconsider that investment too.

I would rather know I was wrong than protect a program because it matches my politics. That same standard should apply to policing.

Council votes next week

Council cannot repeal House Bill 1900. It cannot rebuild Texas municipal finance in five days. It cannot put a floor under anything the Legislature declined to protect.

It can direct the $4.7 million in certified revenue to the social service contracts rather than anywhere else, which covers all but about $700,000 of the cut. It can ask what the $24 million buys before approving it, given that the department's own line items answer with a wage settlement and an overhead transfer rather than anything shown to make a person safer. And it can decide whether $652,000 for young people's jobs is disposable in a budget that found $2.5 million more for the Liability Reserve.

Ask afterward what that decision saved.