Fix Medicare's Bizarre Auction Program

Ian Ayres

Crosspost from Freakonomics:

Here’s a piece co-authored with auction guru Peter Cramton, a professor of economics at the University of Maryland:

Fix Medicare’s Bizarre Auction Program

By Ian Ayres and Peter Cramton

Harry Truman once quipped, “Give me a one-handed economist! All my economists say, ‘On the one hand, on the other’” Often even a lone economist has difficulty making a recommendation. While true on certain matters, there are many issues where economists do agree about the right and wrong course of action. A case in point is competitive bidding for Medicare supplies.

Economists and other auction experts agree that using administrative prices from 25 years ago to set Medicare prices is a bad idea, and that a much better approach is to price Medicare supplies in competitive auctions. That is not surprising. What is surprising is the degree of consensus that Medicare’s shift to auctions is fatally flawed and must be fixed for the Medicare auctions to succeed in lowering costs while maintaining quality for medical equipment and supplies.

For the last ten years, the Centers for Medicare and Medicaid Services has been testing an auction approach that is incredible in the inefficiency of its flawed design. This policy brief lays out a number of weaknesses with the auction procedure but it is sufficient to focus on the interaction of just two:

Bids are not binding commitments

In the Medicare auction bidders are not bound by their bids. Any auction winner can decline to sign a supply contract following the auction. This undermines the credibility of bids and encourages low-ball bids in which the supplier acquires at no cost the option to sign a supply contract. This aspect of the current system has led to the predictable outcome where a number of bidders, realizing that prices were set below their costs, have refused to sign contracts.

Flawed median-bid pricing rule

As is standard in multi-unit procurement auctions, bids are sorted from lowest to highest, and winners are selected, lowest bid first, until the cumulative supply quantity equals the estimated demand. Non-standard is that the current system sets reimbursement prices using the median of the winning bids rather than using the clearing price. Since most providers are small, they lack the resources to invest in information and strategy in preparing bids. For them an effective and easy strategy is the low-ball bid, as any one firm’s impact on price is negligible. The low-ball bid is appealing to these firms because it is a winning bid with a negligible effect on the price. However, with many firms following this strategy the median-bid price is significantly biased downward and possibly below the cost of all suppliers. This possibility is not a problem for the low-ball bidders since, as described above, suppliers have the option of not signing the contract in such an event. Equally troubling is that fifty-percent of the winning bidders are offered a contract price less than their bids.(emphasis added)

There are good reasons why we have never seen a median pricing rule combined with withdrawable bidding. It is not likely to elicit serious signals of value. You can read more about the auction rules, the relevant portion of the Federal Register explaining the final rules, a journal article showing some of the problems, and the official bidding form, eligibility requirements, and quality standards).

One of us recently asked a group of auction experts (mostly prominent economists but also computer scientists and engineers) to be signatories of a letter to Chairman Stark, House Ways and Means Health Subcommittee, advocating the use of auctions to price Medicare supplies, but sharply criticizing the government’s proposed auction approach, which the administering agency (the Centers for Medicare and Medicaid Services) has been testing for the last ten years in several metropolitan areas. In less than 48 hours, 167 experts signed the letter—including multiple Nobel prize winners and members of the National Academy of Sciences.

Medicare should junk the flawed procurement auction rules and take advantage of the enormous advances that have been made in auctions and market design to fix the auction rules. “The appropriate bidding mechanism would arise from a collaboration of government officials, industry representatives, and auction experts,” wrote Peter and Brett Katzman in the policy brief mentioned above. “It would emphasize transparency, good price and assignment discovery, and strategic simplicity. The result would be sustainable long-term competition among suppliers that reduces costs while maintaining high quality.” This approach has been used with great success in other complex settings such as government auctions of radio spectrum and modern electricity markets. The engagement of auction experts should not be surprising. The economic problems being solved are far from trivial. The government would never consider building a bridge without the input of bridge experts on its design. Similarly, the government should enlist the help of expert auction designers when structuring complex auction markets. (In full disclosure, we have provided paid advice to government and non-governmental entities on how best to structure auctions.)

The mystery is why the government has failed over a period of more than ten years to engage auction experts in the design and testing of the Medicare auctions. The letter confirms that any expert would be able to quickly identify fatal flaws in the Medicare competitive bidding program. We suspect the problem is that CMS initially did not realize that auction expertise was required, and once they spent millions of dollars developing the failed approach, they stuck with it rather than admit that mistakes were made. This bureaucratic inertia is seen not just in government but in all organizational decision making.