Who Actually Signs the Order

There is a version of Washington that appears on camera, and there is a version that produces the documents. They overlap less than most people assume.

A policy announcement is the last step in a process that has already been running for months. By the time a principal stands at a lectern, the decisive choices have been made in rooms full of people whose names will not appear in any story: deputy assistant secretaries, career attorneys in an office of general counsel, budget examiners, the one staffer in a regulatory office who has read the underlying statute more carefully than anyone else in the building.

These people are not shadowy. They are civil servants and mid-level appointees doing exactly what the law asks of them. But they hold a specific and underappreciated kind of authority, which is the authority to write down what an instruction means.

The gap between the decision and the text

A principal decides on a direction. The direction is usually a sentence or two: tighten this, expand that, get it done before the fiscal year closes. The text that implements the direction runs to forty pages and contains a hundred choices the principal never made.

Which entities count. What the effective date is. Whether existing contracts are grandfathered. What the exceptions process looks like and who administers it. Whether the reporting requirement is quarterly or annual. Each of these is presented, when it is presented at all, as a technical matter. Each of them determines who is actually affected.

The person who chooses the definitions has more influence over the outcome than the person who chose the direction.

Lobbyists understand this perfectly and allocate their attention accordingly. A trade association will spend far more effort on a definitional footnote than on the announcement itself, because the footnote is where the money is. Journalists, working on a different clock, generally arrive after the footnote is settled.

Why this class keeps growing

Two forces have expanded the drafting layer. The first is statutory complexity. Modern legislation delegates enormous discretion to agencies, partly because Congress cannot agree on specifics and partly because the specifics genuinely require expertise that a legislative staff does not have. Every delegation creates a drafting job.

The second is turnover at the top. Senate-confirmed positions sit vacant for long stretches, and acting officials are reluctant to make choices they may not be around to defend. Discretion drains downward to the people who will still be there next year. In a department with a dozen unfilled slots, the career staff is not filling a gap, it is running the operation.

The accountability problem is real

It would be easy to write this as an expose, and easy to be wrong. The drafting layer is not a conspiracy. It is mostly people trying to produce something that will survive a court challenge, which is a discipline that generally improves the work.

The problem is narrower and harder. Democratic accountability assumes that voters can trace a decision to someone they can remove. When the meaningful choices are made three levels below the last person anyone voted for, that chain does not break so much as fade. There is no scandal to uncover. There is just a long chain of reasonable technical judgments that add up to a policy nobody campaigned on.

What would actually help

Faster confirmations would help, and everyone has said so for twenty years without effect. Better documentation would help more and costs almost nothing: an agency that publishes the drafting record along with the final rule makes the choices visible to anyone willing to read them.

The most useful change would be in how the work is covered. A press corps organized around principals will keep missing the substance, because the substance is not where the principals are. The story is in the docket, the comment file, and the definitions section. It is duller than a lectern, and it is where the decisions live.