Why a Reorganization Does Not Move the Conditions

The proposal arrives already drawn. Boxes, dotted lines, a slide showing before and after, and a paragraph about alignment and efficiency. Somewhere behind it there is a real problem somebody is trying to solve, and it is often nowhere in the document.

Why is this always the answer?

Because it is the only large thing an executive can do without asking anybody.

Consider what the alternatives require. Hiring needs an appropriation and a hiring authority. A new system needs a procurement, an authorization and a year. Changing a statute needs Congress. Changing incentives needs the personnel system. Every one of those has a gatekeeper and a calendar.

A reorganization needs a memo. It can be announced Tuesday and stood up in a quarter, it demonstrates decisiveness immediately, and it produces an artifact that can be shown to an oversight committee. Nothing else in the executive's toolkit has that profile. So the instrument gets selected on availability, and the diagnosis is written afterwards to fit the instrument that was going to be used anyway.

March and Olsen made a version of this argument in 1983 and it has not aged. Studying comprehensive administrative reorganization, they concluded that reorganization rhetoric and ritual affirm an interpretation of political life at least as much as they serve as a basis for short-run decisions. That is not a charge of cynicism. It is an observation that the act carries meaning independently of its results, which is precisely why its results are examined so rarely.

What does GAO actually say?

Almost exactly this, in the plainest language a government auditor uses.

GAO published key questions for assessing agency reform efforts in June 2018, written for Congress and OMB to use on reorganization proposals. The framing is direct. Reorganization is not a cure-all, and at times a careful analysis would have shown that a problem is caused more by faulty processes, a poorly trained workforce or weak leadership than by the structure. GAO also notes that past reform proposals have not always come to fruition and can take years to implement fully.

Read the second sentence beside the first. A structural change that takes years to implement, against a problem that was never structural, spends the years and does not reach the problem. The intervening period is not neutral either, which is the next section.

What does the move itself cost?

Reliability, temporarily, and the cost is almost never on the slide.

Hannan and Freeman argued that organizations accumulate reliability and accountability over time, and that structural change resets that accumulation. The implication is uncomfortable and well supported: the period following a reorganization carries elevated risk of failure rather than reduced risk, because the routines, relationships and tacit knowledge that made the old structure work do not transfer with the boxes. Somebody knew who to call. Now they do not.

There is a second cost that is measurable and usually unmeasured. Bernerth, Walker and Harris developed and validated an instrument for change fatigue, which establishes it as something with a magnitude rather than a complaint. An organization on its third reorganization in six years is not simply an organization that has changed three times. It is one whose capacity to absorb the fourth is lower, and the fourth is the one that might have been necessary.

Which problems does a chart actually reach?

One of the two kinds, and the distinction is the most useful thing in this article.

Gilbert unbundled organizational inertia into two components that behave differently. Resource rigidity is the failure to shift investment, and it is a question of where money and people sit. Routine rigidity is the failure to change the processes that use those resources, and it lives in habits, approvals, review cycles and what people believe they are allowed to do.

A reorganization moves resources directly and reliably. That is its genuine power, and where resource rigidity is the constraint it is the correct instrument. Routine rigidity is untouched by it. The same approval chain, the same review board, the same three signatures reconstitute themselves inside the new boxes within a quarter, because the people who ran them are still there and still believe the same things about what is required.

Gilbert also found that a strong sense of threat eases resource rigidity and worsens routine rigidity, which explains the pattern everyone recognizes: a crisis makes it easy to move money and harder to change how anything is done.

Where does adaptive capacity actually sit?

In capabilities rather than in structure, which is why the chart is the wrong place to look for it.

Teece located the capacity to adapt in the ability to sense opportunity, seize it and reconfigure around it, and he grounded those in microfoundations, the specific processes, decision rules and skills through which an organization does them. None of the three is a property of a reporting line. A structure can make them easier or harder, which is not nothing, and it cannot supply them.

This is what the Four A's calls Adaptability, and the definition is deliberately behavioral rather than structural. It asks whether a change survives contact with the organization. A reorganization is itself a change that has to survive that contact, so it does not sit outside the condition it was meant to improve. It competes with the other changes for the same scarce capacity, which is the argument in The Adaptability Gap.

The test, before the chart

One sentence, written down, that does not contain a box.

State the constraint. It usually takes one of three shapes.

• Decisions on this class of work wait four levels above where the knowledge is.
• Funding for the mission sits in an account whose owner is not accountable for the mission.
• Nobody can name who owns the outcome, so three offices each own a piece and none owns the result.

Now read the sentence and ask whether moving boxes reaches it. The second one is resource rigidity and a reorganization is the right instrument. The first and third are routine and authority problems that will reconstitute under new names, and the honest answer is that the chart does not reach them. GAO-03-669's implementation guidance is built on the same premise, that the transformation is a people and culture problem the structure only frames.

That sentence costs an afternoon. The reorganization it either justifies or prevents costs years, so the asymmetry is not close.

Why this is an Adaptability problem

Because the response is fixed before the diagnosis, which is the definition of a system that cannot learn.

An organization that answers every constraint with the one instrument it can reach alone is not choosing badly each time. It is not choosing at all. And the reason it is worth naming as Adaptability rather than as poor judgment is that the failure is structural in the honest sense: the incentives, the calendar and the gatekeepers make the chart the available move, so anyone in that seat reaches for it. No villain is required.

The diagnostic is one question, asked before the slide is built. What would have to be true for a reorganization to be the wrong answer here? If nobody in the room can produce a version of that, the analysis has not been done, whatever the deck says.

Evidence matrix

What to do with this

Before the next reorganization proposal is drafted, write the constraint in one sentence that names no box, and circulate that sentence alone. If the people who would execute the change cannot agree on what it says, the structure was never the disagreement. If they can, the sentence tells you within a paragraph whether a chart reaches it.