Baker administration revises T absence rate

THE BAKER ADMINISTRATION is now acknowledging that the MBTA absenteeism information cited in a report by the governor’s special advisory panel probably overstated the problem and that the reported comparison to other peer transit systems had little relevance.

In its April 8 report, the governor’s advisory panel said the absence rate across all positions at the T was 11 to 12 percent, about twice the level reported by peer agencies. State transportation officials are now using a more conservative calculation that puts the rate at 8.6 percent, and they make no claims about how the T’s absence rate compares to those of peer agencies.

CommonWealth raised questions about how the panel’s initial numbers were derived in a story published on April 30. Transportation Secretary Stephanie Pollack, in a talk at Suffolk University on May 5, insisted the panel’s numbers were accurate. “I categorically reject the charges that the panel’s numbers were not true. The panel’s numbers are accurate,” she said.

The new absence rate numbers were posted on the T’s website on May 10, in a section entitled “Further analysis prepared after completion of panel report.”

In a statement issued on Thursday, Pollack said the governor’s advisory panel tried to highlight areas affecting the T’s performance. “One of the issues highlighted was unscheduled absenteeism,” she said.
“The panel’s initial effort was to highlight unscheduled absenteeism as both a management and operations problem and, in the short duration of the panel’s work, provide context to reinforce the point that it needed to be addressed and warranted further study. There is no definition for scheduled and unscheduled absences, nor is absenteeism data collected based on uniform standards applied industrywide; as such, efforts to provide context to this problem were meant to place emphasis on the high number of unscheduled absences based on the best available information at the time and within a short window of time.  Regardless of interpretation of available data from the T and other transit agencies, what we do know is that unscheduled absenteeism impacts the ability of the MBTA to provide reliable service (shown, for instance, through the number of dropped bus trips because of a lack of an operator); drives up overtime costs; reduces the cover list and the ability to react to emergencies; decreases customer satisfaction; and creates an administrative burden.”

The T’s new absence rate is calculated by dividing the average number of unscheduled days-off taken by T employees by the number of working days in a year. The governor’s special panel estimated the absence rate was 11-12 percent, but there have been different explanations on how that number was calculated. State transportation officials initially told CommonWealth the number was calculated by dividing the average number of unscheduled days-off (22.5) by their estimate of total average days a T employee works (204). After CommonWealth reported on the absence rate, the members of the advisory panel wrote the magazine a letter saying their number was calculated lumping workplace training, military duty, and jury duty in with other categories of unscheduled days-off, including family medical leave and sick days, to reach a total of 26 days. That number was divided by 231, which was derived by taking 261, the total number of working days in a year, and subtracting the average 30 vacation days and holidays taken by T employees.

The method that produces the 8.6 percent number uses different numbers. State transportation officials say they divided 22.5 (the average number of unscheduled days-off, not including workplace training, military duty, and jury duty) by 261 (the total number of working days in a year). Officials say the new number is a very conservative calculation that minimizes controversy over what to include or exclude.

Although the new numbers on the T website differ from those in the panel’s report, they still paint a serious problem. Overall, the average T employee is out of work 57 days a year – 30 days for vacations and holidays; 22.5 days for such unscheduled absences as family medical leave and sick days; and 4.6 days for such activities as workplace training, military duty, and jury duty. The average T employee shows up for work 204 days a year, 12 days short of the T’s internal target of 216 days a year.

The unscheduled days-off present the greatest problem for T managers, who have to scramble to find replacements for workers who take a day off with no notice. While the overall absence rate was 8.6 percent, the rate varied by individual worker categories. On an average day in fiscal 2014, according to the data on the T’s website, 10.9 percent of operators and 8.1 percent of mechanics were out unexpectedly, compared to 6.8 percent of transportation supervisors and 5.5 percent of maintenance supervisors.

The unscheduled days off can lead to service disruptions. In fiscal 2014, a total of 22,102 bus and rail weekday trips were dropped, or 1.64 percent of the 1.35 million trips that year.  Of the total dropped trips, 86 percent were bus trips and 14 percent were rail trips, primarily because the system runs far more bus trips than rail trips. Slightly more than half of the dropped rail trips were due to no operator being available, while 76 percent of the dropped bus trips were caused by the absence of an operator.

Of the average 22.5 days of unscheduled absences taken by T employees each year, 7.56 days are taken under the Family and Medical Leave Act, 6.93 are sick days, and 3.68 days are injury or worker’s comp.

Steve Koczela is a data analyst for CommonWealth and the president of the MassINC Polling Group, which is owned by MassINC, the publisher of CommonWealth.