Avik Roy argued a few weeks ago that states have "a huge fiscal incentive" to turn down federal funds to expand Medicaid to all adults living on up to 133 percent of the federal poverty line. This is despite the fact that the federal government has agreed to pay for 100 percent of the expansion in its first years and at least 90 percent of it thereafter.
Roy sees two risks. One is that Medicaid expansion will lead to enrollment by more people who are eligible for Medicaid today but have not signed up. States won't get the enhanced federal match for these enrollees, since they were already eligible; on average, that means the state will pay 43 percent of the cost of covering them, instead of 0 to 10 percent.
The other is that the federal government might reduce its contributions to Medicaid in future years, due to large projected federal budget deficits.
While these concerns are real, they are not very good reasons to turn down the Medicaid expansion.
On the first objection, the "woodwork" problem, the costs simply are not that large. The Kaiser Family Foundation looked at the costs of Medicaid expansion in 2010 and found that, under an assumption that Medicaid expansion would significantly drive up participation by existing eligibles, states would end up paying 7.5 percent of the cost of expanded Medicaid between 2014 and 2019, rather than 4.5 percent.
The cost to state governments of covering those who "came out of the woodwork" would be about $4 billion a year, spread across the country. That figure assumes that about a third of the people who haven't signed up for the Medicaid they're eligible for today would start doing so under Obamacare. These new participants would be, by and large, fairly cheap to cover, since people with extensive need for care are more likely to already be enrolled.
It's also worth remembering that expanded participation is a good thing. The objective of the Medicaid program -- which all 50 states have voluntarily agreed to participate in -- is to provide health coverage to poor people. When eligible people do not participate, it is a policy failure.
On the issue of future reductions of the federal match, it is always possible that the federal government could change the rules in the future. This is true of any state federal program, including Medicaid as it already exists. But state governments are not trapped: if they don't like the new terms, they can always drop out of the program.
Conservatives' likely retort to this is that it's hard to abolish existing entitlement programs. But that's not always true: welfare reform, which reduced the generosity of cash payments to poor people, was popular and therefore politically viable.
The reason Medicaid is likely to be sticky is that it is popular. A Bloomberg News poll last year found that Medicaid cuts were the least popular of 12 deficit reduction options, with support from just 21 percent of respondents. Even cuts to Medicare were more popular.
If state governments become politically unable to drop expanded Medicaid, it will only be because the state's residents consider the program a good use of taxpayer money. That hardly seems like a big risk.
But if any people are really concerned, they should vote for Mitt Romney, who ran for governor in 2002 with the position that the federal government should take on a larger share of Medicaid costs in order to relieve state budgets. He is sure to stand in the way of any efforts, Republican or Democratic, to shift Medicaid costs to the states.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.