Twelve percent more files reached resolution last month. That number represents a record for the housing authority. It does not mean twelve percent more people moved into apartments. This figure describes the movement of paper. It tracks the migration of digital records from the active queue to the electronic archive.

The supervisor stood before the whiteboard during the monthly review. He circled the percentage with a green marker. He spoke about the increase in team productivity. The clerks sat in their ergonomic chairs and looked at the floor. They understood the nature of the achievement. They knew how the numbers reached their current height.

A closed file is a completed task. It counts as a victory on the departmental dashboard. The system does not distinguish between a successful placement and a technical denial. A family receiving a voucher closes the file. A family failing to provide a bank statement also closes the file. Both actions satisfy the metric.

A clerk identifies a missing signature on page fourteen. They issue a denial notice immediately. This action takes five minutes of administrative labor. Helping the applicant find a pen and understand the form would take forty minutes. The dashboard rewards the five-minute interaction.

Ahmed S.-J. understands the mechanics of small components. He works as a watch movement assembler. He spends his days fitting gears into brass plates. Each wheel must align with its corresponding pinion. A single millimeter of deviation stops the entire mechanism. He knows that forcing a part into place might make the assembly look finished. The watch will still fail to keep time.

The Damaged Escapement

Ahmed views the housing bureaucracy as a damaged escapement. An escapement regulates the release of energy in a watch. It controls the pace of the hands. When the escapement moves too quickly, the watch loses its relationship with reality. It records hours that have not yet passed. The housing authority has tuned its escapement for maximum speed. It creates a record of success that does not exist in the physical world.

The intake process follows a rigid sequence of events. A clerk receives an application and checks for basic eligibility. They verify income and household size against federal limits. If a document is missing, the clerk sends a request for information. The applicant has ten days to respond. The system marks the case as closed if the deadline passes without a reply. This action clears the queue.

The pressure to close cases changes the behavior of the staff. They stop calling applicants to remind them of deadlines. They find a missing utility bill and issue a denial notice. This action is efficient. It produces a result that the management can measure. It removes a name from the waiting list.

High numbers on the list indicate a failure of the social safety net. Low numbers suggest the system is functioning correctly. The administration seeks to lower the numbers. They achieve this by purging the list of “inactive” participants.

A purge cycle begins with a mass mailing. The authority sends a postcard to every person on the list. The postcard asks the applicant to confirm their continued interest. They must return the card within fourteen days. People who have moved or lost their mail do not respond. Their files are closed in bulk.

The dashboard turns green during a purge month. The productivity metrics reach their peak. The clerks process thousands of closures with a single keystroke. The administration reports a significant reduction in the backlog. They do not report on where the families went. They do not know if the families still need housing.

Measuring productivity in social services is a complex task. We assume that measuring the process improves the service. This assumption is often incorrect. When the metric is “cases closed,” the fastest path is closing them unfavorably. The rational dashboard quietly incentivizes the opposite of the mission.

Clean Data, Messy People

The software used by the authority was designed by consultants. These consultants prioritize legibility. They want the data to be clean and easy to read. Clean data requires simple outcomes. A “denied” status is a simple outcome. It requires no further follow-up. It represents a clean break in the narrative.

A successful placement is messy. It involves inspections and lease negotiations. It requires communication with landlords. These tasks take time. They slow down the movement of the file. A file that stays open for months is a red mark on the dashboard. It suggests that the clerk is struggling with their workload.

The families searching for housing must navigate this mechanical indifference. They need to find open section 8 waiting lists before the clock starts ticking. They need to know the deadlines before the clerk looks for a reason to close their file. The applicants are responsible for their own survival in a system that rewards their exclusion.

“She succeeded in getting him into an apartment. Her supervisor criticized her for that hour. He told her she was falling behind the team.”

– The Story of Sarah, Clerk

The clerk named Sarah used to work differently. She spent an hour helping an elderly man find his birth certificate. She made phone calls to the state records office. She succeeded in getting him into an apartment. Her supervisor criticized her for that hour. He pointed to her low closure rate. He told her she was falling behind the team.

Sarah now follows the metrics. She sees an incomplete application and clicks the denial button. She does not feel good about the action. She feels like a gear in a watch that keeps the wrong time. She watches the green bars on the dashboard grow taller. She knows that every tall bar represents a family that was not helped.

The institution has sacrificed the ground-level craft of assistance. It has replaced help with processing. Processing is the act of moving information from one state to another. Help is the act of changing a person’s circumstances. The two are not the same. The metrics cannot measure the quality of the help.

We see this pattern in many modern organizations. A call center measures the number of calls handled per hour. The operators hang up on difficult customers to keep their average low. A hospital measures the speed of patient discharge. The doctors release patients before they are fully recovered to free up beds.

When an institution optimizes for a legible metric, it sacrifices the unmeasured. The unmeasured is where the humanity of the work resides. It is the extra ten minutes spent explaining a rule. It is the second phone call to a struggling applicant. These actions are invisible to the dashboard. They are treated as waste.

The logic of the machine is internally consistent. It seeks to maximize its own efficiency. It views the applicant as raw material. It views the clerk as a processor. If the processor moves the material to the “finished” bin, the machine is happy. It does not matter if the “finished” bin is actually a trash can.

Ahmed S.-J. recently matched all his socks. He performed this task with the same precision he uses for watches. He aligned the heels and the toes. He folded them into neat squares. This task provided a sense of order. It was a simple system with a clear outcome. He wishes the housing authority could achieve a similar alignment between its goals and its actions.

The housing crisis is not just a shortage of buildings. It is also a failure of administration. We have built systems that are too efficient at saying no. We have rewarded the gatekeepers for keeping the gate closed. We have mistaken the movement of paper for the movement of people.

The clerk moves the file to the archive to prove the desk is empty. The desk is empty because the people have been erased from the data. The data is clean and the dashboard is green. The mission is dead.

The supervisor will give another presentation next month. He will show a new set of charts. The numbers will likely be even better. The clerks will continue to click the denial button. They will continue to meet their targets. The gap between the metrics and the reality will continue to widen.

We must ask ourselves what we are actually measuring. If we measure the speed of closure, we will get fast closures. If we want to measure housing, we must measure how many people are in homes. This requires a different kind of data. It requires a system that values the messy work of staying open.

The watch on Ahmed’s wrist is accurate because every gear is perfectly aligned. It does not try to move faster than the sun. It does not skip seconds to reach the end of the hour. It respects the time it is meant to track. The housing authority should learn from the watchmaker. It should respect the lives it is meant to serve.

The files should stay open until the people are inside. This would be a slow metric. It would be a difficult number to explain to the board. It would be a messy dashboard with many yellow and red bars. It would also be a system that actually works. We have enough empty desks. We need more full apartments.