What I want to clarify is that it's not about whether the cost is justified or not. Under the listing of requirements or studies that I've suggested, are seat belts on buses, under our conditions with kids in snowsuits, proven to decrease a very low probability of risk now to the extent that there is not an increased risk from the implementation of belts?
We see a lot of other issues that arise from using belts, other than the loss of efficiency. There is the possibility that kids are not clipped in properly. That causes more distress to us than the possibility of a potential accident at some point in the future, because we have 24 million riders a year in our system. To have to monitor the belting of each one of those riders to ensure that they are properly belted in, in the event of even a minor accident—because they do happen—would be a challenge. Then we'd also be increasing the potential for injury in minor accidents because they're wearing belts, whereas today they wouldn't have an injury.
So that's the testing, really, that we're looking for. Does the implementation of the belts lead to greater risks? If it doesn't, and if it actually is proven to lower risk, then great. We would say, “It definitely makes the bus safer. It's worth it to implement belts.” However, that data is not available for us right now.