Inefficient operation of some ATC towers the Federal Aviation Administration manages cost the U.S. government $853 million in additional controller hours and equipment over a five-year period, according to the Department of Transportation inspector general. The FAA agreed with its parent organization that tower efficiency can be improved, but said the IG’s methodology in assessing efficiency is flawed.

In an audit report it released to the public on August 24, the IG found that inefficient towers on average required $142 million in additional costs each year relative to efficient towers from fiscal years 2008 through 2013—or $853 million in total. The office found a wide disparity it the towers it assessed as either efficient or non-efficient. “While we found a large share of towers to be relatively efficient in each year that we examined, the gap between their performance and that of the least efficient towers was substantial,” the report states.

The report landed at a sensitive time for the FAA. In pending legislation to reauthorize the agency’s spending programs, the U.S. Congress is considering a fundamental restructuring that would separate its ATC and regulatory functions. One possible model would see the Air Traffic Organization spun off as a not-for-profit entity, comparable to some air navigation service providers.

The IG determined tower efficiency by assigning weighted values to “inputs,” including labor hours and equipment, and “outputs,” including the number of air traffic operations handled. It then calculated the ratio of the weighted sum of outputs to the weighted sum of inputs. The analysis was separated into two parts; it compared busier airport hub towers to each other and non-hub towers to non-hub towers. Nevertheless, the “environmental difficulty” a tower faces, taking into account factors such as the percentage of local traffic, runway configuration and the number of runways “is not the primary determinant of whether a tower is relatively efficient or inefficient,” the office found.

Large hub airport towers the IG ranked as “consistently relatively efficient” are: Hartsfield-Jackson Atlanta, Denver, Dallas-Fort Worth, Newark Liberty, Houston George Bush Intercontinental, Las Vegas McCarran, La Guardia, Chicago O’Hare and San Diego airports. Those identified as “frequently least efficient” are: Boston Logan, Ronald Reagan Washington National, Washington Dulles International, Orlando, Chicago Midway, Seattle-Tacoma and Salt Lake City airports.

The difference between relatively efficient and inefficient hub airport towers “is not necessarily a consequence of the inefficient towers using less productive combinations of labor hours and equipment to do their work,” the IG stated. “Instead, the difference results at least in part from the fact that the inefficient towers are actually using more of each input to handle their operations and prepare trainees.”

In a response to the findings attached to the audit report, the FAA agreed that tower efficiency can be improved. But the IG’s comparative analysis and ranking methodology for tower efficiency “is flawed for several reasons,” the agency said. Among them, the office did not consider that the FAA closes some ATC towers at night for more cost-effective handling of traffic; did not consider the relative use of contractor resources by towers; and used the “book value” of equipment as an input, making towers with newer equipment look relatively more expensive. “The agency’s position is that any meaningful facility-to-facility comparison should reflect the variables noted above, as well as other relevant facility-specific variables,” the FAA said.

Asked for its reaction to the IG report, the National Air Traffic Controllers Association cited the agency’s rebuttal, adding that it would leave any response to the FAA.