Most conservatives believe that government cannot run any program efficiently. Many economists agree this is true, given the structure of incentives for civil servants versus private sector managers, although most see the question as more nuanced. One popular response is to have government contract with private sector entities when it needs something done rather than do it with public employees and equipment. But even that is complicated.
A tempest in the local teapot of my home state of Minnesota over contracting out Medicaid administration illustrates than clearly. HMOs have administered a large fraction of the caseload for Medicaid and for Minnesota’s own program for people with low incomes but above the Medicaid cut-off levels. Now a new, more competitive bidding process has caused one such HMO to be cut out of the business entirely for 2016. The upshot is a political and legal morass that illustrates the inherent challenges when a government contracts with the private sector to provide public goods like military hardware or transportation infrastructure in addition to health care administration.
Economic theory provides some support for conservative political thought: There are market incentives for efficiency in private organizations that are absent in government. But contracting with the private sector isn’t as simple as it may sound. There are parallel incentive problems.
My state has two large health programs for low-income people. One is Medicaid, which serves just over 1 million people, or about 18 percent of the total population. Total annual payments from taxpayers total $10.6 billion. The federal government pays about 57 percent, the counties that directly oversee it about 1.7 percent. The state bears the rest, or some $4.4 billion for fiscal year 2015.
MinnesotaCare is my state’s program for people who do not qualify for Medicaid but who are too poor to afford private insurance. It is much smaller. Numbers average about 100,000 with a total taxpayer cost of some $550 million. About half of this is paid by the state.
Both programs are administered by private HMOs of which UCare, an entity that includes the University of Minnesota’s hospital system, has had the biggest role. This business makes up about half of this HMOs total revenue.
Now, to lower costs, the governor has pushed a more competitive bidding process.
The preliminary outcome for the two-year period beginning in 2016 is that, based on cost and an assessment of an array of performance indicators, UCare would lose nearly all of its current share of these programs. This would cause a shift in health plan administrators for some 360,000 people in 58 of our 87 counties.
State officials say the new bidding procedure will save some $450 million to the state and federal government over the biennium. The exact criteria for scoring the proposals have not yet been disclosed.
Now think for a minute about how government can provide certain services. One way is to simply do so itself, with government-owned facilities or equipment and government employees.
This remains the model for nearly all K-12 schools and a big fraction of post-secondary education. It also once applied to county hospitals and “poor farms” and much federal defense procurement. Think of the historic arsenals at Springfield, Mass. And Harpers Ferry, Virginia or the Brooklyn Navy yard.
Times change, however.
The bulk of defense procurement now comes from private contractors. Cities and counties still plow snow, patch potholes and fix water-main leaks, but new road construction and major reconstruction projects are contracted out. Charter schools and vouchers still cover a small fraction of all students, but numbers are growing.
Care of dependent homeless, hungry, disturbed and disabled people is often done by private nonprofits funded by government grants or contracts, rather than at government-operated facilities. Moreover, many counties have transferred administration, and sometimes ownership, of their hospitals to private health entities.
The reason is that private entities, nonprofit as well as for profit, often are more effective and more cost efficient in getting the job done.
So what are the problems?
One is defining exactly what job needs to be done. The less standardized the good or service, the harder this is. Casting a 5-ton ship’s anchor or applying 4 inches of asphalt overlay to 12 miles of state highway are pretty straightforward. Designing a new multi-role fighter plane or providing health care is not. And when defining the exact task is difficult, so is measuring the outcome.
When bidding for simple, well-defined tasks, a simple one-round auction can suffice, with selection based a single factor alone — price. But with complex tasks, such as health care, some weighted multiple performance criteria is needed and some degree of subjectivity inevitably enters in. This opens the process to becoming political and even a matter of litigation, as a lawsuit by UCare against the state demonstrates.
Research by the late Nobel laureate William Vickery and others looked at mistakes made by bidders in correctly specifying costs and returns. This is a perennial problem for road builders and other contractors.
In the health care arena, it nearly sank a local private health insurer, whose low rates grabbed a large market share of the state’s health insurance exchange under the Affordable Care Act. This insurer dropped out the second year, its managers now understanding what the term “winner’s curse” means.
One way of overcoming imperfect information on either side is to use multi-round auctions in which an initial set of bids or proposals is opened, compared and disclosed to all potential bidders, which then have a chance to submit revised proposals.
But the state’s new procedures, in which the criteria used and the ratings assigned each bidder are not disclosed until after the decision becomes irrevocable, seems particularly and needlessly opaque.
One danger in frequent re-contracting in this case is that any public-outlay savings are offset by the private transaction costs of hundreds of thousands of people who have to shift their health coverage. This affects actual health care providers and the county level administrators as well as the beneficiary households.
Another danger emerges if the contract selection process causes a long-term reduction in the number of viable bidding HMOs, reducing competition and raising future costs in a way that is difficult to predict in advance or measure when it occurs. If enough bidders drop out, this can lead to a pure monopoly, potentially negating any savings and exerting influence to even skew real market demand.
However, political influence can be exerted to keep an otherwise high-cost or low-performance bidder in the game. That is what skeptics see happening right now as key elected officials rally to defend the threatened organization.
There is no simple answer here, and there won’t be any agreement on the outcome, regardless of what it is.
St. Paul economist and writer Edward Lotterman can be reached at firstname.lastname@example.org.