William Galston recently argued that immigration has become so central to American politics, economics, and society that it warrants a new cabinet-level Department of Immigration. He suggests that immigration policy has outgrown the Department of Homeland Security, which was created after 9/11 primarily to fight terrorism. The implication is that a structural reorganization would better align federal institutions with present realities.
I find the argument unpersuasive—not because immigration is unimportant, but because reorganizing government without diagnosing the structural defect it is meant to cure is a category error. When the Department of Homeland Security was first proposed in 2002, I argued against it on two grounds. First, it would add an unnecessary layer of management. Second, consolidating multiple but not all domestic enforcement functions into a single department would make that department insular.
Both predictions have come to pass.
The Redundancy Problem
The creation of DHS merged 22 agencies under one umbrella. But “merged” is misleading. What actually happened was not simplification but accretion. Existing agencies retained much of their internal culture, operational autonomy, and statutory authority. What was added was a superstructure: a new secretary, new undersecretaries, new compliance offices, new review processes, and new coordination mechanisms.
In other words, we added a layer.
In complex systems, corporate or governmental, adding a coordination layer without eliminating underlying complexity rarely produces clarity. It produces reporting overhead. It produces bureaucratic friction. It produces delay. The core statutory frameworks governing immigration, border enforcement, disaster response, and intelligence did not meaningfully change when DHS was created. What changed were reporting lines.
Twenty-plus years later, DHS is one of the largest federal departments. It is not simpler. It is not notably more coherent. It is simply larger.
Reorganizations are seductive because they create the appearance of action. But unless they alter incentives and statutory authority, they mostly rearrange boxes on an organizational chart.
The Insularity Problem
Before DHS, immigration enforcement and related functions were embedded in larger institutional ecosystems. They were part of Justice, Treasury, Transportation. They were subject to broader departmental cultures and cross-pressures.
DHS consolidated many enforcement-heavy agencies into a single department defined by a unifying mission: homeland security. Mission framing matters. When an organization’s identity is tied to preventing catastrophic risk, it naturally evolves toward risk aversion, secrecy, and internal cohesion. Those traits are not inherently bad but they are self-reinforcing.
At the same time, not all enforcement agencies were consolidated. The FBI remained in Justice. ATF remained separate. So we did not create a unified domestic enforcement system. We created a large enforcement bloc with its own culture and intelligence apparatus, operating alongside other enforcement bodies.
That is a recipe for institutional insularity.
Over time, large consolidated departments tend to defend their turf, justify their budgets, and interpret data in ways that reinforce their mission. That is not a moral failing. It is bureaucratic physics.
Does a Department of Immigration Solve Anything?
Galston argues that immigration is no longer primarily a security issue and therefore should not sit inside DHS. He proposes a new Department of Immigration, along with a Bureau of Immigration Analysis to assess economic needs. But what specific dysfunction would this cure?
Is the asylum backlog caused by DHS reporting lines? No. It is caused by statutory design, procedural rules, and insufficient adjudicative capacity.
Are deportation controversies caused by departmental placement? No. They are driven by executive priorities, prosecutorial discretion, and political mandates.
Is the emphasis on family reunification versus economic migration a structural problem inside DHS? No. It is embedded in immigration law enacted by Congress decades ago.
You do not fix statutory architecture by moving boxes.
The proposed Bureau of Immigration Analysis raises a deeper issue. It would assess “the needs of the economy.” But that phrase hides an assumption. Which economy? At what wage levels? Under what capital-labor balance? With what automation trajectory?
If the bureau takes current sectoral dependence on immigrant labor as given, then it is not analyzing needs it is ratifying existing business models. That is circular. It treats present economic structures as optimal and adjusts immigration policy to sustain them.
But immigration levels influence wages. They influence labor force participation. They influence automation incentives. They influence fiscal burdens on states and localities.
Those are political choices. They are distributional questions. They cannot be rendered technocratic simply by creating a bureau.
The Lesson of DHS
The lesson of DHS is not that we need another reorganization. The lesson is that structural consolidation without incentive reform produces bureaucratic insulation. If immigration policy is obsolete and I believe much of it is, then the remedy lies in:
- Rewriting the statutory balance between family and economic migration.
- Reforming asylum procedures.
- Designing workable guest-worker programs.
- Enforcing employer verification consistently.
- Clarifying federal-state responsibilities and compensation mechanisms.
Those are legislative and policy questions.
Creating a Department of Immigration would not answer them. It would elevate immigration symbolically to cabinet status. It would trigger new congressional committees, new political appointments, and a multi-year transition process. It would consume administrative energy. It would not necessarily produce clarity.
In large organizations, structural reorganization is the most expensive and least reliable form of reform.
I opposed DHS because it added a layer and consolidated enforcement culture in ways that encouraged insularity. Both occurred. We should hesitate before repeating the experiment.
If there is a structural change worth considering, it would be more modest and more surgical: separating adjudication from enforcement, creating independent immigration courts, or embedding economic migration functions within departments already oriented toward labor markets and economic analysis.
In complex systems, decomposition by function is usually more stable than centralization by theme.
Immigration is indeed central to our national life. That is precisely why we should resist cosmetic reorganizations and instead focus on the harder work of statutory reform and incentive design.
Rearranging the chart is easier. It is also less likely to work.
