Capacity Utilization of U.S. Airports: A Data Envelopment Analysis

Air travel demand is forecasted to grow substantially following a long standstill period. However, it is unclear whether the air transportation industry is adequately equipped to handle this surge. In particular, concerns about the airport capacity have resurfaced. While previous studies have extensively examined airport capacity from an engineering perspective, an economic capacity assessment could shed further light on this issue. We examine the capacity structure of 59 large and medium hub airports in the U.S. in the years from 2015 to 2019 with a linear programming model that includes a modified DEA based on Johansen (1968)’s capacity definition and a standard DEA. In addition, a probit model is employed to investigate if external factors, such as airport size, legacy airline hub status, multi-airport system, population, and per capita income in metropolitan statistical areas account for excess capacity. The major findings in the first stage of analysis are: (i) Based on 2019 estimates, 76% of sample airports would not experience a capacity problem in the next seven years after air traffic returns to the pre-pandemic levels if the variable inputs like the number of employees and operating expenditures are utilized; (ii) the average efficiency of U.S. airports in the sample period is 0.79, indicating that the airports could increase their output by 21%; (iii) the variable inputs are underutilized, suggesting that there is a shortage of these inputs, and (iv) capacity utilization rates of large hub airports are 9% higher than the medium hubs. The second stage of our analysis provides deeper insights into the findings from the first stage, revealing that larger airports and airports in a multi-airport system are less likely to experience excess capacity. Besides, the results identify population as a significant factor, with a larger population leading to a lower probability of excess capacity. Overall, this study suggests that the US airport industry could benefit from improved efficiency and better utilization of variable inputs to meet the expected surge in air traffic demand.