You are here

Fourth Quarter 2003 Air Travel Price Index (ATPI) Airline Fares Up 1.5 Percent from 2002

Fourth Quarter 2003 Air Travel Price Index (ATPI) Airline Fares Up 1.5 Percent from 2002


BTS 13-04
Dave Smallen

Wednesday, May 25, 2004 - The Air Travel Price Index (ATPI) rose 1.5 percent in the fourth quarter of 2003 from the same period in 2002, the second consecutive year-to-year increase (Table 1), the Department of Transportation's Bureau of Transportation Statistics reported today.

The fourth quarter level was 106.33 (1Q1995=100), the highest fourth quarter since 2000.

The ATPI is a quarterly measure of changes in airfares since the first quarter of 1995 for itineraries on U.S. carriers beginning in the United States. The ATPI was released for the first time in March 2004.

ATPI shows that airline ticket prices in the fourth quarter of 2003 rose 3.4 percent from their post-2000 low in the fourth quarter of 2001. In the fourth quarter of 2000, fares reached their highest level of any fourth quarter since the base period in 1995 (Table 2).

In addition, fourth-quarter 2003 airfares rose 0.8 percent from the third quarter. Quarter-to-quarter changes may be affected by seasonal factors (Table 3).

The largest year-to-year fare increase among major cities for which BTS calculated individual indexes was 23.2 percent in Kahului, HI. The biggest drop was 6.2 percent in Milwaukee. (Table 4).

Additional information about the ATPI, including indexes for foreign-origin itineraries and the top 30 air travel markets based on passenger enplanements, can be found on the BTS website, The first quarter 2004 ATPI will be released in August.

The ATPI series are computed using price index methodology similar to that used by other federal statistical agencies. Although the ATPI is computed using a tested index methodology, the effective application of this methodology to the airlines' data is still under development and it is considered a research series at this time.

Table 1: Percent Changes for U.S.-origin flights from Previous Years to 2003

Fourth Quarter to Fourth Quarter

Excel | CSV

Percent change to 4th Quarter 2003 Since . Duration
1.5 2002 One Year
3.4 2001 Two Years
-4.7 2000 Three Years
4.5 1999 Four Years
7.3 1998 Five Years
-1.4 1997 Six Years
7.1 1996 Seven Years
7.4 1995 Eight Years


Table 2: Fourth Quarter ATPI - U.S.-origin (First Quarter 1995 = 100)

Excel | CSV

  ATPI Percent change from previous year
1995 - 4th Quarter 98.96
1996 - 4th Quarter 99.30 0.3
1997 - 4th Quarter 107.80 8.6
1998 - 4th Quarter 99.09 -8.1
1999 - 4th Quarter 101.73 2.7
2000 - 4th Quarter 111.56 9.7
2001 - 4th Quarter 102.86 -7.8
2002 - 4th Quarter 104.72 1.8
2003 - 4th Quarter 106.33 1.5


Table 3: Quarter-to-Quarter Change in ATPI for Five Quarters for U.S.-origin Itineraries

(First Quarter 1995 = 100). Quarter-to-quarter changes may be affected by seasonal factors.

Excel | CSV

  ATPI Percent change from previous quarter
2002 - 4th Quarter 104.72 1.3
2003 - 1st Quarter 107.98 3.1
2003 - 2nd Quarter 105.79 -2.0
2003 - 3rd Quarter 105.53 -0.2
2003 - 4th Quarter 106.33 0.8


Table 4: Top 30 Air Travel Markets - Top Five Fare Increases and Decreases

Fourth Quarter 2002 to Fourth Quarter 2003 (First Quarter 1995 = 100)

Excel | CSV

    Fourth Quarter 2002 Fourth Quarter 2003 Pct. Change
Largest Increase 1. Kahului, HI 109.48 134.83 23.2
  2. Honolulu, HI 125.64 143.68 14.4
  3. Charlotte, NC 116.76 128.34 9.9
  4. Columbus, OH 103.22 113.31 9.8
  5. Manchester, NH 81.22 87.86 8.2
National Average   104.73 106.33 1.5
  5. W. Palm Beach, FL 100.40 97.27 -3.1
  4. Dallas/Ft Worth,TX 104.04 100.77 -3.1
  3. Orlando, FL 96.75 93.56 -3.3
  2. Atlanta, GA 102.00 96.36 -5.5
Biggest Decrease 1. Milwaukee, WI 101.87 95.55 -6.2


Brief Explanation of the ATPI

The ATPI is based on fares paid by travelers and draws its data from the BTS Passenger Origin and Destination Survey. Through this survey, BTS collects information from the airlines on a 10-percent sample of airline tickets. Each ticket sold is assigned an identification number, and if this number ends in 0, the ticket is in the sample.

The index measures the aggregate change in the cost of itineraries originating in the United States, whether the destinations are domestic or international, but only for U.S. carriers (excluding charter air travel). The ATPI is based on the changes in the price of individual itineraries, that is, round trips or one-way trips for which no return trip is purchased, and the relative value of each itinerary, for the set of matched itineraries.

The index uses the first quarter of 1995 as the reference point (expressed as the number 100) against which all subsequent quarterly prices are measured. ATPI values below 100 represent overall "cost of flying" levels less than those in the first quarter of 1995, while values above 100 represent cost of flying levels that exceed those of the first quarter of 1995. ATPI levels can be used to compute percentage changes in overall fare costs between any two quarters in an ATPI series.

Unlike many other price index estimates, the ATPI is not based on a fixed "market basket" of air travel services. Rather, all of the data from the Passenger Origin and Destination (O&D) Survey are fed into the estimation system each quarter, and this collection of itineraries varies from one quarter to the next. New entry, including routes and carriers, will not be included in the ATPI calculations until it has been present in the O&D Survey for two consecutive quarters.

The ATPI differs from the Bureau of Labor Statistics' (BLS) airfare index, a component of the Consumer Price Index. The BLS index is based on fares advertised through SABRE, a leading computerized airline ticket reservation system, while the ATPI uses actual fares paid by travelers. Since a growing number of tickets are purchased through the internet at discounted prices not listed with SABRE, the ATPI does not show the same levels of increases as the BLS index.

For price comparison purposes, itineraries flown in each quarter are "matched up" with identical or very similar itineraries flown in other quarters. A price index formula is then used to compute aggregate index estimates such as those that appear in this release.

The fares reported in the O&D Survey include taxes, so the ATPI values reflect changes in tax rates as well as changes in fares received by the airlines. The ATPI values in this release are not adjusted for seasonality, so some movements in the series are due to seasonal variations in airfares.