Our nation faces multibillion-dollar disasters, a changing climate, a rising sea, fires that can’t seem to be put out, and many other varieties of mayhem. From the flatlands of Kentucky to the seaport district of Boston, we are not built to withstand the harms we are likely to face in the years ahead. Public-policy experts and politicians talk of building a more resilient nation, the de rigueur notion of our time. Resiliency is understood as the capacity to create systems that can regroup, bounce back, adapt, and return stronger, and this notion of resiliency has taken on a life of its own as leaders in industries as far-ranging as business, health care, law, and psychology all have jumped on the bandwagon.

But in all these discussions of how to build resiliency into our domestic fabric, we may be getting far ahead of ourselves. There is one fundamental flaw in how we organize, plan, and adapt for the disasters ahead. In short, our disaster management is a disaster. It is time to rescind or fundamentally revise the Stafford Act.

The Stafford Act was passed in 1988 as a mechanism to formalize a rather hodgepodge system of grants and investments to help cities and states that suffer a major disaster. The act establishes a cost-sharing arrangement among federal, state, and local governments after a disaster. Administered by FEMA, it sets in motion a rather technical process, requiring proof of preliminary damage assessments and resource needs, and has different standards for personal and public-sector losses. It requires that the federal government provide at least 75 percent of the cost of the damage of any disaster, so that communities and individuals can get back to where they started. And that’s exactly the problem.

The notion of a federal disaster relief fund is absolutely necessary; it is the obligation of government not to abandon citizens in times of distress. But the Stafford Act suffers from a terrible and obsolete design flaw: It essentially requires that the disaster funds be used to bring the damaged building, road, or home back to the way it was before the disaster came. That’s a problem, because recipients, knowing the government will be there to compensate, have no capacity or inclination to alter their behavior. There is simply no requirement that those rebuilt structures be built better, stronger, or more resilient. In fact, anyone who wishes to build differently faces considerable administrative burdens. Rarely flexible and tied to its statutory language, the act contains distribution mechanisms that create little to no incentive to build better; nor does the act have any mechanism to compel recipients to do so. Sean Reilly, a board member of the Louisiana Recovery Authority, expressed the frustrations of working with the act while trying to re-envision New Orleans after Hurricane Katrina when he told PBS, “Okay, that was a 40-year-old building; let’s rebuild a 40-year-old building.”

In addition to this, there’s another problem. The Stafford Act was passed at a time when disasters were viewed as both rare and random. But it can no longer be said that disasters are rare or random. Extreme weather events are increasing in frequency and causing more damage. From 1980 to 1990, the United States experienced 30 different billion-dollar disasters, which inflicted about $156 billion (adjusted for inflation) in damages. From 2004 to 2014, there were 88 billion-dollar disasters, which caused about $623 billion (again, adjusted for inflation) in damages. Our country is changing, and the Stafford Act creates a terrible incentive by not requiring that recipients actually use the money to build differently. While there have been some changes to allow funds to be used for resilience planning, they are minimal and rarely mandatory.

The consequences of Hurricane Sandy, and the congressional fights over how many billions should be authorized, started an important conversation that resulted in some amendments to the Stafford Act allowing for more resources to go to hazard mitigation. Yet the crux of the act remains the same—people just want their money, to go back to where they were—so much so that many Republicans who opposed Sandy aid were back begging for disaster relief when their jurisdictions were hit with fires, tornadoes, or hurricanes.

Most debates today about the Stafford Act are either about the burdensome administration of the process or the perceived stinginess of FEMA. In 2010, the United States Conference of Mayors recommended changes to the Stafford Act and almost every provision addressed the cumbersome bureaucracy, not the overall vision. There will be more flooding, more sea-level rise, more hurricanes, more tornados, more fires. Getting back to where we were is not a smart option. When it comes to disaster-management payments, we’ve created a system that undermines the potential to build smarter, applauds amnesia, and relegates citizens to a constant loop of misery and vulnerability, as predictable as the oceans rising.

Imagine instead a disaster relief system based on the belief that disasters are likely and not exactly random. Waters rise in coastal areas. Tornados kill in flatlands. Earthquakes shake along fault lines. True, there is an exception to every rule, but our policy mechanism ought to create requirements for future likelihoods. Funds would require cities to create zoning laws to prohibit building in certain areas, states to support beach nourishment, citizens to build sea walls or abandon their homes. The Stafford Act, or a replacement law, must have strings attached to require resiliency and mitigation measures by those who receive federal aid, no matter the tragedy they have withstood. In the long run, it will serve everyone better.