High deductibles cut care across the board — the needed with the wasteful

High-deductible health plans are built on a simple theory: make people pay more of the cost of their care up front, and they will cut out the care they do not need while keeping the care they do. The best evidence says the first half happens and the second half largely does not. When people face a deductible, they cut spending — but they cut valuable care and wasteful care at roughly the same rate, and they do almost none of the price-shopping the design assumes.

The natural experiment

The sharpest modern evidence comes from a study published in The Quarterly Journal of Economics in 2017, which examined what happened when a single large firm moved its entire workforce from a plan that covered care in full to a high-deductible plan, all at once [s1]. Because everyone switched at the same time, the researchers could isolate the effect of the deductible itself rather than of which employees chose it.

The switch caused total health spending to fall by between 11.8% and 13.8% [s1]. That is a real reduction, and at first glance it looks like the theory working. The details are where it comes apart.

People cut the good care too

The design's promise is that patients, now spending their own money, will drop the low-value care — the unnecessary scan, the redundant test — and keep the high-value care. The study found instead that consumers "reduced quantities across the spectrum of health care services, including potentially valuable care (e.g., preventive services) and potentially wasteful care (e.g., imaging services)" [s1]. The deductible did not act as a scalpel that removed waste; it acted as a blunt instrument that reduced use fairly indiscriminately.

Nor did people shop for lower prices, which is the other half of the theory. The researchers decomposed the spending drop into price-shopping, cutting quantities, and switching to cheaper alternatives, and found the reduction came overwhelmingly from people simply using less care, not from seeking out cheaper providers for the same care [s1]. Consumers responded sharply to the price they faced at the moment of care — reducing spending by 42% while under the deductible — but they did so by going without, not by bargain-hunting [s1]. Strikingly, even the sickest patients, who would blow through their deductible within the year and face low prices for most of it, cut back early in the year when the up-front price was high [s1]. They were responding to the immediate price, not to their true cost for the year, which is not how a rational shopper is supposed to behave.

An older experiment reached the same core finding

This is not a new or isolated result. The RAND Health Insurance Experiment, a landmark randomized study that assigned families to insurance plans with different levels of cost-sharing, established the pattern decades ago. Participants who faced cost-sharing made one to two fewer physician visits a year and had about 20% fewer hospitalizations than those with free care [s2]. But the reductions again fell on needed and unneeded care alike: the proportion of hospitalizations judged inappropriate was the same — 23% — in the cost-sharing and free-care groups, as was the inappropriate use of antibiotics [s2]. Cost-sharing reduced effective and less-effective care in roughly equal measure [s2].

RAND's overall conclusion was that modest cost-sharing reduced use with, for the average person, negligible effects on health — but with a crucial exception: it did have adverse health consequences for the sickest and poorest participants [s2]. That exception matters, because those are exactly the people a deductible hits hardest and who can least afford to guess wrong about which care to skip.

What the evidence supports

The claim that high deductibles reduce spending is well supported — by roughly 12-14% in the modern natural experiment, and by comparable magnitudes in the RAND trial [s1] [s2]. The claim that they do so by trimming waste while preserving valuable care is not supported: in both studies, people cut valuable and low-value care at similar rates, and the modern study found they barely shopped on price at all [s1] [s2]. The honest reading is that a deductible is a blunt tool. It lowers spending mainly by making people use less medicine of all kinds, and its costs fall disproportionately on the sickest — which is why the same evidence that recommends deductibles as a spending brake also warns against expecting them to distinguish good care from bad. None of this is advice about which plan to choose; it is what the research finds about how deductibles change behavior.