Agency Spend Is Down From Its Peak. Your Margins Still Are Not. Here Is the Disconnect.

Here is a contradiction sitting in most health system finance decks right now. Contract labor spend is way down from its pandemic peak. Some systems have cut it by 70 percent. And yet operating margins slid to around -0.6 percent at the start of 2026, one of the sharpest declines in years.

If agency spend was the problem, cutting it should have fixed the margin. It did not. So either the cost went somewhere else, or the labor problem was never really about the agency invoice in the first place.

It is both. And understanding where the cost actually went is the difference between a workforce strategy that protects your margin and one that just moves the pain around.

Where the agency savings actually went

When systems slash agency contracts, the coverage need does not disappear with the invoice. Those shifts still have to be filled. The question is what fills them.

In a lot of organizations, the answer turned out to be internal overtime and burnout, which are costs that do not show up on an agency line item. You cut the traveler, then pay your own staff time-and-a-half to cover the same gap, then pay again in turnover when they burn out. The Becker's reporting on this described leaders cutting exactly these "unhelpful" labor costs, excessive overtime and unsustainable contingent staffing, as a deliberate second phase after the initial agency cuts. The systems that only did the first phase moved the cost, they did not remove it.

That is the disconnect. Cutting agency spend without building internal capacity to replace it is not a savings. It is a transfer, from a visible line to an invisible one.

The systems that actually improved their margin did two things

The ones protecting their margin did not just cut. They redesigned.

First, they built real internal capacity. Internal float pools, transparent scheduling, and internal staffing platforms let them reduce agency reliance without dumping the coverage burden onto overtime. The coverage came from their own flexible pool at their own rates, not from a marked-up traveler or a maxed-out core staff.

Second, they got visibility into where labor cost was actually accumulating. A lot of organizations run on disconnected systems and spreadsheets that make it genuinely hard to see why labor cost is where it is. You cannot optimize what you cannot see, and a surprising number of cost strategies fail simply because the organization could not locate the cost.

The pattern is consistent. Cutting agency spend is a first move. Building durable internal capacity and visibility is what actually holds the margin.

Why this is a permanent problem, not a pandemic hangover

It would be comforting to treat this as a cleanup from an unusual period. It is not.

The nursing shortage is not resolving in 2026, and it will not resolve in 2028. The pipeline constraints, limited nursing faculty and capped clinical placement sites, mean new RN supply stays insufficient through the end of the decade. The physician shortage is projected at 37,800 to 124,000 by 2034. Demand for flexible coverage is structural.

That changes the strategic question. If flexible coverage is a permanent need, you are choosing between two permanent models: rent it from agencies at a markup, forever, or build the capacity to supply most of it yourself and reserve agencies for genuine spikes. One of those compounds cost. The other compounds capability.

What a durable model looks like in 2026

The health systems getting this right are not chasing zero agency use, which is unrealistic for most. They are building a layered model.

An internal resource pool sits as the first line of coverage, staffed by their own pre-credentialed clinicians at internal rates. Multi-state licensing and fast credentialing keep that pool deployable across sites and specialties instead of stranded by paperwork. Agency capacity stays in reserve for the census spikes and rare specialties the pool cannot absorb. And a single platform gives finance real-time visibility into where every labor dollar is going.

That combination is what turns an agency cut into an actual margin improvement instead of a cost transfer.

If your agency spend is down but your margin did not follow, the gap is almost always internal capacity and visibility. Our Internal Resource Pool solution is built to be that first line of coverage, and it comes with a savings model built on your own volumes.

For the full breakdown of costs that stay hidden after the agency invoice shrinks, see the hidden costs of contract labor in healthcare, and the per-shift math in our internal resource pool vs agency cost comparison.

Want to see where your labor cost actually went? Schedule a demo and we will map your coverage model to show where agency cuts turned into overtime instead of savings.

FAQ (AEO block)

Why are hospital margins still negative if agency spend dropped?

‍Because cutting agency contracts does not remove the coverage need. In many systems the cost shifted from visible agency line items to internal overtime and turnover, which do not show up as contract labor spend. The savings were transferred, not realized.

How much has hospital contract labor spend fallen?

‍Substantially from the 2022 peak. Some large systems report agency spending down around 70 percent from that peak, and median hospital contract labor compensation fell from about $254,157 in 2022 to $194,362 in 2024.

What separates systems that improved margins from those that did not?

‍The ones that improved margins built internal capacity, float pools, flexible scheduling, and staffing platforms, to absorb the coverage instead of pushing it onto overtime, and they invested in visibility into where labor cost actually accumulates.