Criminals Promised, Volume Delivered: Inside ICE’s Enforcement Model

Donald Trump ran on a simple promise: focus immigration enforcement on criminals and make the country safer. The policy now being implemented tells a different story. With tens of billions of dollars directed toward arrests, detention, and removals, the enforcement system has been structured to maximize volume rather than reduce risk. That design choice matters because it shapes who is targeted, how force is used, and whether public safety is actually improved.

This is not a dispute over whether immigration law should be enforced. The question is whether the policy now in place matches what was promised and delivers the safety outcomes that justified its scale and cost.

What enforcement is optimizing for

In campaign language, “criminals” carried a clear meaning: people who commit violent acts, traffic drugs, or pose a direct threat to communities. That framing implied prioritization, judgment, and a focus on public safety.

Immigration law allows something broader. Most immigration violations are civil, not criminal. Under that legal framework, U.S. Immigration and Customs Enforcement can lawfully detain and remove individuals with no criminal conviction. That authority has existed for decades. What has changed is how enforcement success is measured and rewarded.

Reporting by Reuters documents that ICE field offices are operating under sharply increased daily arrest targets, in some cases tripling prior expectations. When leadership evaluates performance by arrest counts, behavior follows. Officers are rewarded for speed and throughput rather than for time-intensive investigations that require coordination, evidence development, and discretion.

This is not a failure of individual agents. It is the predictable outcome of a management system designed around numbers.

The dragnet in the data

Independent data reinforce this pattern. Analysis by the Transactional Records Access Clearinghouse shows that a majority of people held in ICE detention have no criminal conviction. Recent increases in detention have been driven largely by individuals without U.S. criminal records.

This does not mean the agency cannot distinguish between violent offenders and day laborers. It means the distinction is not central to how the system is currently run.

When removability becomes the dominant criterion and arrest volume the dominant metric, enforcement naturally expands outward toward low-risk, easily identifiable populations. The wide net is not accidental. It is efficient under the existing incentives.

Violence as a byproduct of scale

As operations have expanded, so have reports of aggressive tactics. Investigations by Human Rights Watch document heavily armed ICE raids in residential neighborhoods and workplaces, including the use of force against individuals with no violent history and minimal flight risk. These operations often involve rapid entries, broad sweeps, and limited differentiation among targets.

Federal courts are now examining whether some of these actions exceed constitutional limits, particularly when federal agents operate in public spaces or in conjunction with protest suppression. The pattern is structural, not anomalous. Large-scale enforcement conducted under time pressure and arrest targets increases the likelihood of mistakes, confrontations, and unnecessary escalation.

Here, violence is not the stated goal. It is a foreseeable consequence of enforcement designed for speed rather than precision.

The safety test that matters

If public safety is the objective, success should be evaluated against outcomes communities recognize as meaningful:

1. Reductions in violent crime.
2. Disruption of organized criminal networks.
3. Improved cooperation between immigrant communities and local law enforcement.
4. Lower rates of mistaken identity, excessive force, and community destabilization.

There is little evidence that mass civil arrests of nonviolent, nonconvicted residents advance these goals. Many law-enforcement professionals have long warned that broad immigration raids suppress crime reporting, erode trust, and make communities less safe.

Billions spent on detention beds, transport contracts, and rapid removals may increase deportation totals. They do not, by themselves, produce safer streets.

The policy gap

The central problem is not enforcement. It is a misrepresentation.

If the administration’s true objective is large-scale removals regardless of criminal history, it should defend that policy openly and on its own terms. That would be an honest debate.

Instead, the program continues to be sold as a public-safety initiative while being executed as a volume-based operation. The gap between promise and practice is not incidental. It defines the policy.

What alignment would look like

An enforcement strategy genuinely aligned with the “criminals, not families” pledge would look different:

• Performance metrics would heavily weight serious criminal convictions and de-emphasize civil-only cases.
• Funding would prioritize investigations, intelligence, and interagency coordination rather than detention capacity.
• Use-of-force standards would be transparent, auditable, and consistently enforced.
• Success would be measured by safety outcomes, not arrest totals.

None of this requires abandoning immigration enforcement. It requires aligning means with stated ends.

The bottom line

The United States is spending extraordinary sums on an enforcement system optimized for volume. The cruelty is not accidental, and neither is the inefficiency. Scale without precision produces both. The unresolved question is simple: if safety was the promise, why does the policy reward everything except making us safer?

Edward Saltzberg is the Executive Director of the Security and Sustainability Forum and writes the Stability Brief.