Many economic measures use a fixed base year to allow comparisons over time. However, the measures are highly sensitive to the base year chosen, and choosing a new base year can change the measure's history dramatically. In the past, when government economists changed the base year for calculating GDP, the revised growth calculations sparked numerous debates about the true state of the economy. At the same time, however, these measures become less accurate the further one moves away from the base year due to an effect known as "substitution bias." In other words, keeping the base year fixed introduces a new problem. One method to address these issues is chaining, a technique that uses values from the current year and the fixed year to calculate values. Chaining is more computationally difficult, but more accurate because it can account for substitution bias. For the Transportation Services Index, the Bureau of Transportation Services uses the Fisher Ideal Index formula to chain the data. Technical details are available at http://1.usa.gov/1PWbN8T.

SOURCE: Bureau of Transportation Statistics, 2016.