I have been beating the drum for some time regarding the state of intermodal service and questioning why we haven’t seen more improvement. That’s why it feels good to be the bearer of good news: Things are starting to get meaningfully better.

To gauge intermodal service quality, I track the average weekly intermodal train speeds as reported by the railroads themselves (all but Canadian Pacific, which doesn’t report). Admittedly, this is a highly imperfect measure because intermodal shippers are much more concerned with consistency and reliability than with raw speed. But I also know from experience that there is quite a bit of correlation between speed and service, particularly when speeds are dropping.

I normally look at the year-over-year comparisons, but this has gotten a bit misleading lately. Service has certainly improved from last year but perhaps is still not up to what used to be considered “normal.” So let’s take a look at the current level of intermodal service, but put it into a long-term context.

Chart 1

Chart 1 presents the results of this analysis, showing a time series of the four-week moving average intermodal train speeds back to the year 2000. Each line represents the average speed for a given year or period by week from the first week of the year to the last. The heavy dotted line portrays the average speed for the 10 years from 2000 through 2009, and then each of the five years since then are shown individually, with the current year being shown in heavy red. Chart 2 presents the same data, but is a snapshot just comparing the most recent four-week period to the same time period in previous years.

Chart 2

During the first decade of the 2000s, intermodal trains were typically averaging approximately 30.5 to 31 mph. This slightly exaggerates the typical performance because it includes 2009, when volume dropped into the tank but speeds were consequently very high. It was when we came out of the Great Recession that railroads achieved a new level of service. Train speeds were high for the better part of four years after the 2009 downturn. Only in late 2013, when the BNSF Northern Tier bogged down and savage weather hit Chicago did performance fall below the previous norm. Things clearly worsened in 2014.

This year got off to a shaky start, but railroads have made slow, steady progress since midyear, and it’s starting to add up. Most recently, average intermodal train speeds have returned to the average level seen prior to the recession.

The network averages, however, provide only a part of the picture. It becomes clear when looking at individual railroads that some are much further along on the road to recovery than others. The western railroads have made substantial progress and are most of the way, if not entirely, back, depending on how one sets the goalposts. But the two eastern rails remain bogged down, with only small year-over-year improvement and substantial deficits versus even pre-recession operating levels. Canadian National Railway’s performance lies in the middle of the road, while Kansas City Southern is running slower than last year.

The analysis provides some reason for optimism and perhaps also a lesson. Although not the sole reason, it’s no coincidence that the most substantial growth in domestic intermodal occurred when service was at its peak, and that recent slowing in intermodal growth has followed a period of service difficulty. As the railroads begin to get their service house back in order, an impediment to intermodal growth is being removed. What remains to be seen is how badly the intermodal reputation has been damaged and how long that damage will take to repair.