Roemer’s Law
Capacity doesn’t just meet demand, it creates it.
What Is It?
Roemer’s Law, drawn from health economist Milton Roemer’s research on hospital utilization, is often summarized as “a bed built is a bed filled.” Roemer found that hospital beds, once built and staffed, tend to get used at a fairly constant rate regardless of the actual underlying need in the surrounding population. Adding capacity didn’t just accommodate existing demand more comfortably. It generated new utilization to match the capacity that had been added. The mechanism isn’t fraud or deliberate overuse. Once a resource exists, spare capacity, an open slot, an available budget line, the people around it face a much lower barrier to using it than they would face justifying its creation from scratch. A hospital doesn’t need a documented surge in illness to fill a new wing; it just needs doctors and patients making a long string of individually reasonable decisions that, in aggregate, absorb whatever capacity is on hand. This is a close cousin of, but distinct from, Parkinson’s Law: Parkinson’s begins with existing work and describes it expanding to fill the time available. Roemer’s begins with existing capacity and describes demand expanding to fill it.
Why Does It Matter?
The organizational version of this shows up constantly and is rarely named. Added headcount, an expanded meeting calendar, extra software licenses, a bigger budget line, all tend to fill up to match whatever was provisioned, independent of whether the underlying workload actually grew to justify it. A team that gets two additional engineers usually finds two additional engineers’ worth of work to do, not because the backlog secretly required exactly that much more capacity, but because available capacity is what gets used. New capacity rarely just sits idle waiting to be absorbed. It lowers the cost of asking for things that previously weren’t worth requesting, so demand expands until the new capacity feels every bit as necessary as the old. This creates a specific measurement trap. “We’re at full capacity” feels like hard evidence that demand has outgrown supply. But a system that generates its own utilization will look exactly as full immediately after capacity is added as it did before, which means “we’re at capacity” is weaker evidence for “we need more capacity” than it feels like from inside the system that’s asking for it.
What Changes Once You See It?
You stop treating high utilization as automatic proof that more capacity is needed, and start asking whether the demand was there independent of the capacity, or whether the capacity is what’s generating the appearance of demand. You start distinguishing between two situations that look identical from the inside: real, previously unmet need that new capacity is finally serving, and ordinary slack-absorption that would happen regardless of whether the underlying need actually increased. Both produce full calendars and full budgets. Only one is real evidence the additional capacity was necessary. You also get more deliberate about capacity decisions being justified by evidence collected before the capacity exists, rather than by utilization measured after it’s already available to be absorbed, since the after-the-fact measurement is exactly the one Roemer’s Law predicts will always look full.
Common Misunderstandings
- It is not a claim that adding capacity is usually a mistake, or that organizations should default to scarcity. Plenty of capacity additions serve real, previously unmet need; the law is a caution about how to tell the difference, not an argument against ever expanding.
- It doesn’t mean utilization data is useless. It means utilization measured after capacity exists is weaker evidence for the original need than it appears to be, which argues for measuring demand independently, before the capacity is added, wherever that’s possible.
- It is not limited to healthcare, even though it originated there. The same dynamic shows up anywhere a resource, headcount, storage, budget, meeting time, can be absorbed by ordinary activity without anyone needing to justify the absorption the way they’d have needed to justify the original request.
- It isn’t a strict law of conservation guaranteeing that capacity always fills completely and demand never genuinely exceeds supply. It’s a strong, well-documented tendency worth checking for, not an outcome that’s certain in every case.
- It is not an argument against building slack deliberately. Spare capacity is sometimes exactly what allows an organization to absorb a surprise, run an experiment, or recover from disruption. Roemer’s Law isn’t a case against slack; it’s a warning that unused capacity rarely stays unused for long, so slack built for one purpose should be expected to get claimed for others.
Diagnostic Question
Would this demand have existed if the capacity had never been created in the first place?
Explore Further
Field Notes
None yet.
Related Field Guide
Origin
Milton Roemer, a health economist and physician, documented the pattern in research on hospital bed supply and utilization published in the 1950s and 1960s. The phrase “a bed built is a bed filled” became the popular summary of his findings within health policy and health economics.