Government liability for flooding

Following up on last week's posts on Christine Klein and Sandra Zellmer's new book, here's another post from them, originally posted on CPRBlog: 

Ft. Calhoun Nuclear Reactor , Nebraksa
in the midst of flooding Missouri River
(Corps of Engineers)

Landowners flooded by the Missouri River in 2011 have sued the Corps of Engineers for a Fifth Amendment “taking” under the U.S. Constitution. Their attorneys hope to rake in over $250 million in claims for their clients and at least $1 million in expenses and fees for themselves.  They’re likely to be disappointed.

Lawsuits seeking recovery of flood damages from the federal government almost always fail. First, the United States is immune from suit for negligent construction or handling of flood control structures under the sovereign immunity shield of the 1928 Flood Control Act, as plaintiffs whose lives were destroyed when levees failed during Hurricane Katrina quickly discovered.  My co-author Christine Klein and I have called for a repeal of this provision in our article and book on Unnatural Disasters, but it hasn’t happened.

In hopes of avoiding the immunity problem, the Missouri River plaintiffs have brought a claim under the Fifth Amendment, which is not barred by the Flood Control Act.  However, this claim is just as unlikely to stick, for good reason.  As we document in our previous work, courts find that floodplain management constitutes a regulatory taking in only the rarest of cases, whether the impact to private property occurs through land use restrictions on construction or through flood control structures like dams and levees.  This is because the impact is neither a “permanent physical occupation” of the property by the government, nor is it an excessive regulation that deprives property of “all economically beneficial use” or has otherwise gone “too far” in adversely affecting reasonable investment-backed expectations of the floodplain owners (in the words of the U.S. Supreme Court). It is simply not reasonable to settle in the floodplain and expect that the property will never flood.

These plaintiffs are attempting to bring their claims within the purview of a 2012 Supreme Court case, Arkansas FGC v. U.S, where a landowner (the State Fish & Game Commission) prevailed on its claim that the Corps had physically taken a flowage easement over its land.  The case raised a unique set of facts and the decision is a remarkably narrow one, and it is completely inapposite to what happened on the Missouri River.  Here’s why.

In Arkansas, the Corps opted to depart from its Water Control Plan for the dam in question by releasing water over longer periods each year during a seven-year period, not because of any physical imperative (e.g., unusual amounts of rain or snow) but because farmers urged it to do so to keep their croplands dry for longer periods during harvests.  The deviation caused a dramatic increase in flooding in a wildlife management area owned by the State, causing widespread and permanent damage to its trees.  The flooding was significant enough, for long enough periods, to change the character of the area and to substantially interfere with the State’s ability to use its land.  The Corps had effectively taken title to the land without going through the appropriate processes for exercising the government’s power of eminent domain.  

In stark contrast to the 2011 Missouri River flood, the Corps’ intentional flooding of Arkansas’s land was the direct and proximate cause of the foreseeable destruction of the State’s property.  The Corps deviated from its Arkansas Plan in order to benefit the farmers, when it knew (or should have known) that the deviation would inevitably destroy the State’s land.  The Corps created winners and losers, and the Supreme Court forced it to pay the loser.

On the Missouri, the 2011 flood made losers out of just about everyone.  The Corps’ flood control structures were taxed to their limits by unprecedented amounts of snowmelt and rain over a long period of time in the spring and summer of 2011.  In April, Rocky Mountain snowpack was 140% of normal; later in the summer, rainfall was three to six times normal in the upper Missouri River Basin. At Sioux City, Iowa (the demarcation between the upper and lower river), runoff measured 13.8 million acre feet (MAF), smashing the old 1952 record of 13.2 MAF.  The third wettest month ever documented on the Missouri River happened to be May 2011 (10.5 MAF) and the fifth wettest was July 2011 (10 MAF).  See National Weather Service, The Historic Missouri River Flood of 2011; Senate Report 112-075 - Energy and Water Development Appropriations Bill, 2012.  That water had to go somewhere, and once the dams were filled to capacity, it went downstream and into the floodplain, as rivers naturally do (especially the Missouri, which is widely known for its flood-prone tendencies).

The plaintiffs argue that the Corps has abandoned its flood control mission in favor of other priorities on the Missouri River.  Specifically, they claim that the Corps kept the reservoirs full in the spring to benefit recreation and endangered species, and that fuller reservoirs means less storage for flood waters. The factual record doesn’t back them up, and the law is more nuanced than they allege.  In truth, Congress directed the Corps to build the dams and manage the system for seven purposes in addition to flood control: navigation; hydropower; water quality; water supply; irrigation; recreation; and fish and wildlife.  Flood control and navigation may be toward the top of the list, but they are far from the only concerns that drive river management.  More to the point, none of the other purposes were prioritized at the expense of flood control in 2011.  The Missouri River system was operated in accordance with the Master River Manual in response to abnormal snowmelt and rainfall that just kept coming for months on end.  The operations were dictated by conditions, not by other priorities.  Sometimes, the river simply reclaims its floodplain, despite human efforts to hold it back.

The tired refrain that the government elevated the concerns of fish over people is a red herring. The real problem is that people wanted to settle in the floodplain, so the federal government undertook flood control, which prompted more people to move into harm’s way. It’s ironic that the landowners who cry “foul” today have received a bounty of flood control-related benefits from the government through the years.  No doubt the flood damage to their properties in 2011 would have been worse if the federal government hadn’t built dams and other structures on the Missouri River.  Consider the 1993 flood, which set the record for the highest water level in Kansas City, but resulted in much lower discharges (flooding) than pre-dam floods in the 1800s and early to mid-1900s. 

Meanwhile, individuals and communities who chose to reside in the floodplain demanded additional protection through the construction of levees, dikes, and revetments on the river and its tributaries, along with subsidized flood and crop insurance.  Once they put themselves in harm’s way (aided and abetted by government), it’s only natural for sympathetic officials to provide federally funded disaster relief when the inevitable happens.  These are policy choices that the government and floodplain communities have made throughout the many years of floodplain occupation, and we can argue the pros and cons of these choices until we’re out of breath without ever reaching a consensus.  Don’t get us wrong—no one wants to see human suffering in the wake of a flood. But adding a constitutional takings claim to the list of government payouts demanded by property owners is a wholly unwarranted sort of “double take” from the government (and the taxpayers) (see Unnatural Disasters Chapter 10).  Instead, we should be talking about how to make people safer, how to make buildings more flood resilient, and about cases where it is more prudent to retreat from the floodplain and out of harm’s way.