Later this month, teams of volunteers will hit the streets across the United States to count the unsheltered homeless people in their communities. City officials overseeing these “point-in-time counts” hope to emulate the results from Newark, New Jersey, which last year cut its street population by more than half. According to The Economist, Newark demonstrates that progress is possible through a “moneyball” approach to homelessness—meaning data-informed investments in housing assistance. 

Newark’s homelessness challenges are worth comparing with those of its cross-Hudson neighbor. New York and Newark have the same climate, and their real estate markets overlap. One might expect, then, that they face street homelessness problems of equivalent magnitude. Of course, New York has a right to shelter, which advocates defend as a tactic for reducing street homelessness. Whatever benefit New York realizes from its right to shelter is at least partially offset by its uniquely expansive transit system, which runs 24/7 and gives people a viable alternative to entering shelter. Around half of New York’s unsheltered population lives in the subways.

The table below shows the effect of New York’s transit system on the homelessness problem. Apart from its subway-dwelling population, Queens has a trivial problem with unsheltered homelessness.

Source: author’s analysis of data from the U.S. Census, New York City, and Essex (N.J.) County

The table illustrates what The Economist is talking about. Between 2020 and 2023, Newark went from having a higher concentration of unsheltered homelessness than New York’s to a lower one. As The Economist tells it, somewhere toward the end of Covid, New Jersey policymakers got serious about rising homelessness. They started “moneyballing homelessness” by developing “near-real-time” data-management systems to ensure that their public investments were more targeted toward what one scholar called “points of intervention.” Instead of spending heaps of money over the course of years on fully staffed shelters and other expensive programs, the argument goes, smart policymakers will provide people with just the right kind of help, at just the right time, to forestall their deterioration. The classic example of this approach is short-term rental aid to get someone through a temporary breakdown in his housing situation.

This approach only goes so far, though, since no homeless American has just one problem; nor was there a single point, along his trajectory into homelessness, where “it all went wrong.” Co-morbidity is life in the social services. That goes especially for those chronic street cases with whom Newark has been supposedly so successful. Major homeless-services agencies have extensive experience with short-term rental assistance, which can go by different names, such as “homeless prevention programs” and “rapid rehousing.” These programs are relatively cheap and can help some people. But at scale, they won’t end homelessness. Indeed, experts’ traditional view is that temporary rental assistance is best suited for more functional “episodically homeless” cases, whereas “chronically homeless” street cases need permanent supportive housing, a more expensive and complicated intervention. America’s homeless crisis cannot be reduced to one of skinflint landlords unwilling to yield on their unreasonable demands for tenants’ back rent.

Social policy debate would benefit if everyone agreed to retire the term “moneyball.” The original moneyball team was the Oakland Athletics, who are now Major League Baseball’s most pathetic franchise. They will soon relocate to Las Vegas. The lesson the A’s supposedly taught, 20 years ago, was that spending smarter on player salaries could be more effective than spending more. Given that baseball has just seen its first $700 million contract, it seems safe to say that the moneyball revolution was oversold. In both social services and sports, you can get a temporary boost from analytics. Then the effect fades, and you must confront the reality that technocracy will never suffice. 

Technocratic solutions are sometimes an appealing alternative to partisanship, but they can also function as an attempt to change the subject. One solution to homelessness with a partisan reputation is law enforcement, a theme only lightly touched on in the Economist article. Enforcement is, admittedly, also a simplistic approach to a complex problem. But enforcement’s track record as an “evidence-based practice” (to the extent that that term means anything) is underrated. Cities that have made their streets more orderly via enforcement pushes include Phoenix, Portland, Oregon, Los Angeles, and New York. If that sample size seems too small, consider the experience of the “silent majority” of U.S. cities whose street conditions never make international headlines because they have never been allowed to deteriorate to the point of outrage. Most cities maintain a tolerable degree of social order because they consistently enforce ordinances pertaining to behavior in public places. Cities have not driven themselves into bankruptcy by doing this, despite progressive advocates’ insistence that enforcement is ruinously expensive. Rental assistance has played a modest role in this success, at best.

Enforcement-as-prevention can’t easily be applied in San Francisco, Seattle, and other cities already dealing with entrenched crises. But enforcement can function everywhere as a kind of harm reduction. It makes street homelessness less unmanageable, thus creating a smaller problem to attack with data-informed rental assistance programs. Reckoning with street homelessness requires robust police involvement. Any “data-informed” approach to homelessness that doesn’t account for enforcement cannot be considered a “truth-informed” approach.

Photo: Paul Bradbury/iStock


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next