Comparing administrative costs across health systems is very tricky and the data are seriously incomplete, with multiple reasons (not just wage rates) to believe the single-payer savings estimate is too high.

Periodic surges of optimism that national health reform's 'time has come' recur roughly every few years (like periodical cicadas) and pass with much noise but little to show.

Three bills with bipartisan sponsorship—two in the Senate and one in the House (the House bill with about 70 co-sponsors equally split between parties)—exist to encourage states to extend coverage in varied ways.

Being slightly overweight is not bad for you, whereas obesity (being way overweight) is bad for you.

Effective heart disease treatment depends heavily on giving patients beta blockers after a heart attack and putting them on aspirin—inexpensive interventions that not all doctors do—rather than on high rates of expensive procedures.

The US has an extraordinarily complex, non-standardized set of arrangements for paying for health care, where physicians and hospitals must handle many different billing forms and keep records to justify bills, making administration more expensive than any social benefit it generates.

The US underspends on a beneficial form of administration: post-audit evaluation of what medical technologies and treatments are worth relative to cost; agencies that did this had short lifetimes because their findings triggered political backlash.

Comparative effectiveness research cannot deliver complete accuracy given the complexity of the human body and data, but it can provide a better guide than current practice, which often follows rules of thumb.

There is enormous geographic variation (up to eight, nine, ten to one) in the frequency of certain medical procedures across US areas that cannot be justified by variations in epidemiology, and higher procedure rates do not necessarily produce better health outcomes (e.g., coronary bypass and angioplasty).

The study's savings estimate was off by about $50 billion because of an error in handling relative wage rates between the US and Canada—in 1999 the Canadian dollar was worth only about 75-80 cents, so Canada spent less partly because it paid less for administrative personnel.

A single-payer system modeled on Medicare would not capture all of Medicare's true administrative costs, because some Medicare costs—such as collecting payroll taxes and their distortionary effects on work effort—are excluded from reported public Medicare spending.

Single-payer systems can achieve genuine economies because hospitals are budgeted with flat sums rather than billing for each individual service, whereas Medicare still pays individual bills for defined services for about 85% of enrollees.

Even if administrative differences between the US and Canada are real and large, they tell policymakers little, because reform inevitably evolves gradually from current arrangements over decades, while real health spending rises about 5% a year and doubles roughly every 15 years—so any administrative savings would be a small, hard-to-isolate fraction of total spending change.

Health care differs from other industries because the goal of policy is to protect individuals from financial ruin in the event of serious illness; once nearly everyone has such insurance, the price mechanism is short-circuited, leaving only second- or third-best methods of limiting spending and weak incentives for uniformity.

Billing non-standardization persists largely because no single payer has a strong incentive to abandon its own forms and adopt another's, and there is a legitimate role for the public sector to establish uniformity where it is efficient—as was done with telecommunications and standard railroad gauges.

Careful recent studies found that on average Americans receive only about 55% of the recommended care for their condition during a typical physician visit or hospital stay, a figure nearly constant across income, race, sex and age.

The Veterans Health Administration—a government-run organization resembling the British NHS—was transformed over about 15 years after Congress improved incentives and gave administrators more flexibility, and can now be argued to be the most cost-effective health delivery system within the 50 US states.

US out-of-pocket health spending as a share has not only fallen greatly but is now lower than in most other developed nations, because although other countries have complete coverage of major expenses, their total cost-sharing is actually larger than in the US (per Jason Furman's Hamilton Project paper).

Because there is an entrenched ideological division in the US and no single approach commands majority support, the federal government should encourage states to experiment with a wide range of reform approaches (as Massachusetts did under Romney), so the country can learn what works—an explicitly bipartisan, federalist strategy.