A homeless couple set up a small tarp for shelter against the elements on West 36th Street and 11th Avenue, near the Javits Center, Tuesday September 9, 2025. Luiz C. Ribeiro for New York Post New York City spends $8 billion a year on social services like homeless shelters, addiction programs and senior centers — billions more than it spends on its police department.
The Department of Social Services, for example, cuts sizable checks each year: nearly $660 million to the Institute for Community Living and about $600 million to RiseBoro Community Partnership, for post-shelter affordable housing.
The Department of Health and Mental Hygiene has funneled more than $1.2 billion to Public Health Solutions.
The Department of Homeless Services just ballooned its deal with the Hotel Association of New York to a staggering $1.9 billion over the next three years.
Under the current system, NYC doesn’t pay the nonprofit contractors that provide these services for results.
It pays them for expenses: rent, salaries, office supplies and so on.
As long as the paperwork checks out, the check gets cut.
It doesn’t matter if the shelter helped keep someone off the street.
It doesn’t matter if the addiction program got anyone clean.
It doesn’t matter if the job training led to a job.
Get selected as the service provider, spend the money, do the work, file the forms, get paid.
Instead of asking “how can we stop drug addiction?” the city asks, “how many social workers are working?”
A nonprofit can burn through its entire contract, produce minimal results for the community and still be made whole by the taxpayer.
That’s not a bug — it’s how the system operates.
And it means there’s zero structural incentive for any contractor to do anything more than comply with the paperwork.
Worse yet, because the city is paying for expenses rather than outcomes, it must verify every single expense.
That means extensive oversight, mountains of paperwork and a contracting system so slow that it often takes multiple years simply to begin work.
The mayor’s own Office of Contract Services acknowledged back in 2019 that this system has no built-in mechanism to hold providers accountable for results.
Here’s what that looks like in practice: Last year, more than 90% of human-services contracts were registered late.
Some nonprofits waited over 200 days after starting work to receive their first payment.
At the Department of Housing Preservation and Development, the average wait was 765 days — that’s over two years without payment.
City Hall has proposed some solutions, including an improved dashboard to track the delays, more bureaucratic staffing and a mechanism to enable partial payments.
These aren’t terrible ideas, but they’re patches on a broken system.
It’s called performance-based contracting, and New York City already uses it — for charter schools.
Charters get public money based on their enrollment, and renewals are tied to student outcomes.
Deliver results, keep your funding; don’t deliver, lose your charter.
It’s simple, and it works: City charter-school kids consistently outscore their peers on state tests.
That model should apply to nonprofits across the board.
Give them enough upfront funding to get started, then tie the rest to real results — people housed, people employed, people in recovery.
This approach does something the current system never could: It removes the need for overbearing administrative oversight while giving contractors a reason to perform.
Right now the city spends enormous resources verifying that money was spent correctly without real needs being met.
Under a performance model, it would spend those resources verifying that people were actually helped — a better use of everyone’s time.
Other states have already figured this out: Illinois and Kansas have both used performance-based models for human-services contracting, with real results.
The Mayor’s Office of Contract Services took a step in the right direction this year when the Department of Correction introduced the Challenge-Based Procurement model to award contracts for education and other inmate assistance programs.
While funding is not yet directly tied to results, the new approach shifts the focus away from bureaucratic process — and that’s meaningful progress.
Yes, measuring social-service outcomes is harder than measuring test scores, and some of the people these programs serve face challenges no contractor can fully fix.
A badly designed metric could end up punishing organizations doing hard work in hard circumstances.
But “it’s complicated” is no defense for a system that burns through billions with little to no accountability.
The status quo has its own victims — nonprofits buried in administrative burdens, communities waiting years for services and taxpayers financing a machine that runs on documentation but is blind to results.
The people these programs are supposed to help deserve better.
Josh Appel is a Manhattan Institute policy analyst.
