Monday, November 21, 2005

Dude, where is my bus?

What is the probability of this happening?

I leave the office towing my huge bag to the Big Blue Bus stop.
Midway: Shucks, I remember my laptop and rush back for it.
Oh good! Now I have my laptop and proceed to the Big Blue Bus stop.
Midway: Damn! I forget my electricity cable for the laptop and go back to pick it up.
Now I am on my way to the Big Blue Bus stop.
And when I arrive, I find the Big Blue Bus rushing past me. Gawd!
The day after...
I am blogging about this!
My laptop hangs!
I curse and make way to the Big Blue Bus stop.
And again, I find the Big Blue Bus rushing past me.

I laughed till I couldn't help crying!

When you have waited for buses in Kolkata, Delhi, Bangalore and finally, Santa Monica you pretty much get used to the agonies of waiting for buses.

We all have had that familiar feeling "I wait for one hour at the bus-stop for one damn bus and then after one hour, I find three of them arriving." It is called the "bus bunching" phenomenon and it happens even when buses are dispatched at regular intervals, primarily due to variability of passengers at the bus-stops and cumulative effect of traffic signal delays.

What is interesting from a public policy perspective is how two agenices: government and the private sector would respond to this problem. I run a government bus service where the drivers are paid fixed wages. You, well, smart you, run a private bus service and you pay the drivers per passenger transported. What would be the possible outcomes? Remember that your incentive would be to have as many passengers as possible in a scenario where other private buses may be competing for the same.

For that, turn to the study The war for the fare: how driver compensation affects bus system peformance done in Santiago, Chile. As it turns out, there are a group of information intermediaries called "sapos" who earn their living by standing at bus-stops, recording arrival times, and selling the information to subsequent drivers who drive past. These drivers change their driving patterns to compete for passengers. That leads to an approximate 10% longer wait on a fixed-wage route compared to the per-passenger route. The fixed-wage contract also leads to more bunching of buses, and hence longer average passenger wait times.

But there is a downside. The per-passenger drivers cause approximately 67% more accidents per kilometer driven. The authors mention "Initially, we conjectured that drivers would improve their spacing by slowing down if they got too close to the bus immediately ahead, but this turned out to be incorrect. In fact, once they get sufficiently close, they attempt to pass the bus in front...this technique often involves aggressive driving by the driver attempting to pass, which can result in an uncomfortable passenger ride or an increased probability of accidents... The other problem also includes stopping at non-designated stops."

So reminiscent of bus travelling in Delhi, where one finds that private buses wait for long at a bus-stop (waiting to maximise the time elapsed) and then rush to pick passengers waiting at the other bus-stops.

The authors suggest policy design like incentivising drivers to maintain even intervals between buses or even GPS technology. However, I wonder if there are any regulations that are debarring brands from arising, especially in Delhi. Ideally, we would want less bus bunching, less passenger waiting time and better quality of service. Designing such a market may involve not having an uniform colour for buses; having the logo of the owner; auction of routes; no limit of buses to be owned. Any other ideas?

Back to the Top

posted by Naveen Mandava at 4:43 PM - Permalink

Back to the Top