Congressman Steven Palazzo, Republican of Mississippi, has proposed an amendment supporting Federal government subsidies to insurance: “We have one common goal—to make sure that insurance remains available and affordable to everyone.”

No, the ardent Obamacare opponent has not been bewitched by a Democratic spell. I left out of his statement one key word: “flood.” It is flood insurance that Palazzo considers essential and worthy of government subsidies, not health insurance. In fact, he finds it so essential that neither likely-to-balloon expenditures nor damaging distortions to insurance and real estate markets dissuade him.

Flood insurance subsidies are now being reduced—sharp premium increases began on Oct. 1. Palazzo’s and other Republicans’ opposition shows that they are not the anti-government ideologues they claim—and perhaps aspire—to be. And if they only cared to look, their support for flood insurance could provide a window of understanding into the reasons for Obamacare.

The National Flood Insurance Program (NFIP) began in 1968 after flood losses, particularly those due to Hurricane Betsey in 1965, made private insurers largely stop selling flood insurance. The Federal government went into the flood insurance business. Those in high flood risk areas got discounted premiums. In exchange, local governments had to commit to flood mitigation—levies and the like.

What went wrong?

First, the subsidized flood insurance encouraged building in high flood risk areas. Oceanfront homes and businesses that would have been prohibitively risky or expensive without NFIP, particularly the discounted premiums, got built. Some properties got repeatedly damaged and rebuilt. Their owners were never forced to learn the obvious lesson.

Second, building in flood areas causes erosion, water quality reduction and other environmental damage. Losses we all face, caused by the few living and building in high risk areas.

Third, climate change, particularly increasing storm severity, is exacerbating these problems. Areas not designated high risk have become high risk. Right now we really need incentives not to build where damage is likely.

Fourth, the program has ended up costing taxpayers—those subsidies I mentioned. Originally, the NFIP was intended to be self-supporting: owners in high risk areas would have their discounted premiums subsidized by everyone else’s premiums. And if payouts were high one year due to a bad hurricane, the NFIP would borrow from the US Treasury and pay it back later. But since the NFIP is not allowed to charge premiums that reflect actual risk, they were never able to pay it back. The fund owes over $20 billion now. And worse storms mean a rapidly growing bill.

Finally, the benefits of the program go disproportionately to the relatively well off. You know any poor people living on oceanfront property?*

All these problems, particularly the ballooning de facto subsidies, drove the Biggert-Waters Act. It mandated the large Oct. 1 premium hike—and many more hikes to come.  And that’s where Palazzo’s amendment comes in.

Why do I oppose NFIP subsidies but support Obamacare and its subsidies? Obamacare addresses serious problems that only the government can fix: providing health care to poor and working class people; and ensuring that those who get sick, including the self-employed, get care and do not face financial ruin. The market, left entirely on its own, cannot do those things. I, like many people (but not everyone), think such outcomes are really important.

Might one make a similar case for NFIP? Floods are random events, much like cancer, that we all want to protect ourselves against. If the private market can’t protect us, surely government should. And the price tag is small by health care’s standards (Medicare alone is about $550 billion per year).

As for the subsidies to the high flood risk areas, why aren’t they just like healthy people paying higher health insurance premiums to reduce premiums paid by those with pre-existing conditions?  Well, people choose where to live and build. It’s more like smoking than most causes of cancer. The more a risk is about stuff that just happens to you and less about what you choose, the more we want the cross-subsidies. And the less we worry about encouraging bad behavior, whether smoking or building coastal resorts.

But I bet Palazzo doesn’t see it that way. I bet he sees extended families that have long been living on the Gulf Coast; small businesses, handed down from generation to the next; and families and businesses that need their communities to grow. He wants to protect them where they are now and where they want to be.

Despite the unrelenting “government is bad, get government out of our lives” rhetoric from Republicans, their actions show that they too think that government can be good, on net, as well as bad, on net. It depends. It depends on the situation. And it depends on your values.

The truth is that we all want to use government to protect what is most important to us from the vagaries of the market. Some of us just won’t admit it, maybe not even to ourselves.

*Wonky footnote: Bagstad, Stapleton and D’Agostino find that the NFIP is not regressive (does not disproportionately benefit richer people, on net). But as they admit, their analysis ignores government NFIP bailout, which is likely to occur, and fails to capture any redistribution towards richer people within a county. But they have a point: all the subsidies do not go to wealthy oceanfront homeowners.