The Affordable Care Act (ACA) got a couple of things right, so right that we should not throw the baby out with the bathwater if we want to move towards a cost-effective, value-driven, and equitable healthcare system.

There are four parts of healthcare insurance reform: private insurance reform; subsidies for low-income individuals who cannot otherwise afford insurance; government coverage (such as Medicare and Medicaid); and operation of markets (exchanges) where insurance is purchased.

The ACA scored high grades on the first three elements and poor grades on the latter. Let’s examine why.

Private insurance reform included adding children up to age 26 in insured groups (also called pools). Including young people improved the risk diversification of the pool, which was good for everyone’s rates as those twenty-year-olds are so healthy. This policy also kept more young people insured.

Pre-ACA private insurance benefited insurers’ bottom lines but left too many people uncovered or under-covered. Insurance reform eliminated health insurers’ ability to refuse to insure (or instead, price out of the market) small employers or individuals with pre-existing conditions or high prior spending.

A change in the Medicare payment system towards incentivizing “value” (highest quality and most efficient care) and not just “number of procedures” was a terrific part of the ACA. The change bent the cost curve, especially once private payers followed and improved upon Medicare’s experiments. The old system maximized healthcare procedures: many of them unnecessary, many too late, and others poorly executed. We need better health, early intervention, and fewer unnecessary procedures and errors. Payment reform is the route. ACA benefit requirements ensured very affordable access to primary care.

The ACA’s Medicaid expansion was an efficient way to dramatically expand coverage. But, when the US Supreme Court allowed Medicaid expansion to be a state-by-state decision, our health care system became less equitable.

Finally, by establishing exchanges, the ACA aggregated small group and individual policyholders into larger group coverage. No longer could a small, older workforce or individual be priced out of the market based on its/her risk profile.

The changes above, coupled with federal subsidies for those who cannot afford health insurance, created insurance access for some 20 million more people. This improvement reduced hospital charity care, saved lives, and mitigated the high cost of delayed care.

Where the ACA fell far short of its promises was in the creation and operation of insurance exchanges.

Their creation did not leverage the considerable administrative capacity in the private sector. Insurers, acting as Third Party Administrators (TPAs)—such as Blue Cross and Aetna—already manage Medicare and self-insured employers benefits. The ACA exchanges relied largely on defense contractors and newly created health co-ops, many of which have gone bankrupt. By not relying exclusively on established TPAs who understand provider contracting and excel in in efficient operations, we got inefficiency, weak provider panels, and inexperience.

Another nagging problem arose from the risk pools for the exchanges, which remained too small and/or had too high a percentage of very sick people. The result was high costs, high deductibles, less choice, and high total out-of-pocket payments for those who used the ACA exchanges. This outcome reduced participation over time by insurers and forced consumers to change plans and doctors.

So how do we improve the ACA without going back to zero, eliminating the advantages presented above?

Large employers, self-insured or not, Medicare and Medicaid populations are large groups that can be insured equitably and, especially as we add payment reform, cost-effectively. Ideally, the Medicaid system could be nationalized and paid for through a progressive tax system, which federal taxes are, lest Medicaid spending continue to displace essential state-provided public goods like infrastructure and education. (Block grants that increase over time accomplish the same thing.) Payment reform should continue for all these groups. The private sector can administer these benefits, as they do now.

Everyone else – small employers and individuals – should be in a large national pool or multi-state regional pools, rather than in multiple, small pools scattered among the states; the US government should continue to subsidize low-income participants as it does now with the ACA, as tax credits are of little help to those with low taxable income, like many elderly and the poor.  We must have a stronger mandate of participation to keep the pools large and diversified by health, making insurance affordable. Otherwise, healthy people “opt out” of insurance until they get sick, which Paul Ryan’s replacement bill will encourage. Every other developed nation mandates participation as they understand the healthy to keep coverage affordable and advance the health of a nation. Yes, the young may pay more than their actuarial cost, but any one individual may get hit by an uninsured motorist or get cancer.  That’s the principle of insurance. Republicans have to just “get over” this unique aspect of healthcare. Economics, not politics, should guide policy in this vitally important area.

This large national pool could be insured by the federal government acting as a single payer, with each person in the group paying his or her fair share and private insurers serving as TPAs. Or, the risk could be born by insurers. Trump’s plan to expand insurance across state lines would be a great addition. Increasing health savings accounts would be great as well. The insurers could create different types of plans to meet the needs of different people just as larger companies offer their employees choices. A minimum set of benefits is required however. Without regulation, there will be a race to the bottom in benefits. Insurance pools will divide by health status, making insurance for those with chronic diseases catastrophically expensive once again.

Trump can improve the ACA then rename it. But let’s not start over from scratch, as doing so will put us on a slippery slope into continued dysfunction in the US healthcare system.