U.S. airports are the centerpiece of the commercial aviation industry that generates $1.45 trillion in economic activity. These airports enabled approximately 980 million trips within the United States last year. Yet, as important as airports are to our modern economy, the data quality resembles what one might expect in a developing country, a problem that requires congressional action to give the Federal Aviation Administration more stringent requirements when it comes to data reporting from airports.
The Federal Aviation Administration (FAA) Authorization Act of 1994 requires all commercial service airports to annually file financial reports with the FAA detailing:
- All amounts paid by the airport to any unit of government and the purposes of those payments, as well as all services and property provided to other units of government and any compensation for each (Sec. 111(a), codified at 49 U.S.C. § 47107(a)(19)); and
- All funds collected and spent by the airport (Sec. 111(b), codified at 49 U.S.C. § 47107(a)(15)).
Yet the actual information reported to FAA is often incomplete, inaccurate, or missing. Unfortunately, the FAA lacks the authority to require more rigorous reporting. The United States can and should do better.
One of the metrics commercial airports must report is cost per enplanement (CPE). Cost per enplanement is the average passenger airline payment per enplaned passenger at an airport. It is a key indicator of the financial costs of airlines to operate at airports. Unfortunately, the cost per enplanement data that many airports report has numerous financial problems, including no shared formula for CPE and varying reporting windows.
A comparison of two large hubs illustrates these flaws. Most private data sources show that Charlotte Douglas International Airport has a lower CPE than Atlanta Hartsfield-Jackson International Airport. Yet, the FAA has very different results.
Compare the two in Table 1 for 2024.
Table 1: Atlanta Hartsfield and Charlotte Douglas Cost Per Enplanement Data Comparison
| Charlotte Douglas International Airport 2024 CPE | Atlanta Hartsfield-Jackson International Airport 2024 CPE | |
| 2024 FAA Form 5100-127 Data | $4.74 | $3.93 |
| 2024 DWU Consulting Firm Data | $1.78 | $3.93 |
Source: FAA data from Federal Aviation Administration, “Form 5100-127,” Certification Activity Tracking System. www.cats.airports.faa.gov/Reports/Form127_Line_Item_query.cfm (accessed 20 Jan. 2026). DWU data from “Large Hub Cost per Enplaned Passenger,” DWU Consulting. www.dwuconsulting.com/airport-finance/large-hub/cpe (accessed 26 Jan. 2026).
FAA’s Form 5100-127 data, Line Item 16.5, shows Atlanta Hartsfield-Jackson International Airport’s passenger CPE is $3.93, the same as what leading consulting firm DWU Consulting reports. Charlotte Douglas International Airport, on the other hand, has a much lower CPE according to DWU Consulting than the FAA report shows. Why is that?
Per DWU’s website, FAA forms use the cost per enplanement, “calculated by using accounting data without allowing any specific modification.” This, in other words, means that airports are not allowed to tweak the formula at all and report based on whatever their latest batch of accounting data says. No, excluding one-time capital costs as they do in bond documents sometimes, or litigation settlements, or non-recurring operating costs. These discrepancies are not a one-time problem and it dates back years. In a separate report, DWU detailed the CPE variance between 2012 and 2013.
Table 2 shows the discrepancy between FAA-reported and airport-reported CPE between 2012 and 2013 based on a table compiled by Dafang Wu with DWU Consulting. The complete table found in Wu’s full article is more exhaustive.
Table 2: DWU Consulting 2012-2013 CPE Airport vs. FAA Comparison
| Airport Code | 2012 Airport CPE | 2012 FAA CPE | 2012 Difference | 2013 Airport CPE | 2013 FAA CPE | 2013 Difference |
| ATL | $2.34 | $2.36 | -$0.02 | $3.70 | $3.52 | $0.18 |
| CLT | $0.96 | $2.33 | -$1.37 | $2.89 | $2.89 | – |
| DEN | $11.56 | $11.90 | $0.34 | $12.00 | $12.52 | -$0.52 |
| DFW | $6.54 | $6.54 | – | $7.20 | $7.20 | – |
| PHX | $5.23 | $5.43 | $0.20 | $5.53 | $5.64 | -$0.11 |
Source: Table adapted from Dafang Wu, “Cost per Enplaned Passenger,” DWU Consulting, 15 Mar. 2015. www.dwuconsulting.com/airport-finance/articles/cost-per-enplaned-passenger (accessed 26 Jan. 2026).
Of the five large-hub airports selected, four had discrepancies between the FAA-reported CPE and the airport-reported CPE in 2012 and three had discrepancies in 2013. Wu found that there are three major contributors to these discrepancies:
- Which segments were reported: DWU’s example was that Honolulu Airport reported a 2013 CPE of $8.56, which was actually the average passenger airline payment per enplaned passenger in the 15-airport Hawaii Airport System rather than the CPE at Honolulu specifically. Wu’s article says this led to a “difference of nearly $2.”
- Airline revenue sharing: This greatly explains the Table 1 difference between Charlotte and Atlanta. Per DWU, “Some airports […] share net remaining revenues with signatory airlines and make payments from the account balance after a fiscal year closes.” This is one of the major problems with not allowing any specific modifications to data reported to FAA. This means that while CLT’s strong parking revenues ultimately reduce what airlines pay, the FAA’s CPE calculation reflects the gross airline charges recorded for the year rather than the net amount after revenue-sharing payments are returned.
- Other factors: Some airports include cargo airline-paid landing fees in CPE, but the FAA line item only includes passenger airline CPEs.
These airports are reporting entirely different numbers, be it net or gross CPE, or reporting systemwide averages instead of the CPE for an individual airport (like DWU’s example of Honolulu’s CPE). This means that the data FAA accepts can be inaccurate (or outright missing in some cases).
What researchers are left with is a data set that can barely be compared to itself, let alone any other airport in the country. It distorts airport-to-airport comparisons and is less transparent than an audited system—if the CPE changes dramatically depending on whether it’s a FAA filing, official statement, or an airline briefing, than transparency suffers. And, when the data is more comparable, it ensures federal grants are spent by airports more responsibly.
What best practices can the airport industry adopt for reporting data?
First, airports can adopt standardized reporting windows in which they submit their annual comprehensive financial reports (ACFR). FAA doesn’t have a mandatory reporting window for airports to submit so most airports submit ACFRs at the end of their local fiscal year. This means while some may submit in June of a year, others will submit in December, having undergone six additional months of inflation since the June reporting airport’s CPE was calculated.
Second, airports can adopt a shared calculation formula. Imagine Airport A calculates CPE by subtracting non-aeronautical revenue (producing net CPE) and Airport B does not. If both were to report a CPE of $5.00, they aren’t actually the same—Airport A is much more expensive to airlines in reality. Requiring both to report both types of CPE, clearly labeled, would allow for better apples-to-apples comparisons.
Finally, Congress should reevaluate its airport financial reporting mandate and consider giving FAA the authority to impose more consistent, accurate reporting standards, such as requiring uniform definitions and standardized formulas for each airport that reports to it. The current law is silent on the submission of audited financial data, leaving FAA to state that it simply “prefers” to receive audited financial data from airports.
Mandatory data auditing would improve data quality immensely. Implementing this kind of oversight could help FAA identify and resolve discrepancies in the data. It would also better ensure federal grants are spent by airport recipients responsibly, which was the motivation for Congress’s original public financial reporting mandate more than 30 years ago.