Taxicab fares: A theory

After being in one too many rushing taxi cabs I am beginning to think the way cab drivers’s rates are set up is a bit skewed. Don’t get me wrong – it’s a very simple system, and simplicity is a very important value. I think right now it’s 1.40$ per kilometre and 40 cents per minute of waiting time. A cab, while driving, makes more money by driving than he does by waiting, per minute, considering he can drive more than 1 km in a minute (notwithstanding the price of gas). Not only this, but the faster he completes that km, the faster he gets his money and picks up another fare to recklessly speed his way to his destination.

So you have a system in place where cab drivers speed past red lights and drive as fast as possible, occasionally despite the safety of their customers. There have got to be tests to be a cabbie, but let’s be honest – most of them involve knowledge of the city or basic driving reqs, neither of which are that stringent. What if we revamp the system to allow cab drivers who drive more carefully to make more money?

Formula (alpha stage): fare = x / y, where x is the time, in minutes, for a driver to get to his destination, and y is the speed at which he attempts to get there (average km/h, let’s say). All this because a) the longer the distance, the greater the cost, and b) the lower (greater level of caution) the speed is, the more profitable the cab ride is for the driver. The cab driver wants x to be high and y to be low, so he’ll drive slower rather than faster (in principle). If the rider wants to get somewhere faster (thus reducing the profit of the cab driver), the driver may refuse and stay at a careful speed, or the rider may attempt to tip (read: bribe) the driver to compensate for the lower base fare at which the fare will then be calculated. The result of a system like this is that cabbies who go out of their way to provide safe trips get monetarily compensated.

Ideally, we wouldn’t want y to be a simple whole number, like km/h or somesuch, but some sort of calculated denominator based on max/min speeds in an area. You get the point. Whatever, I came up with this in a speeding cab on the way home, gimme a break.





9 responses to “Taxicab fares: A theory”

  1. Clarence Avatar

    I’ve often wonder exactly how cab rates are calculated — my math is suspect, however, it seems to me that it would also prove profitable if you were to enforce a flat rate, based on distance.

    1-3 miles, $5
    4-10 miles, $10

    …etc. I prefer a flat rate, because I can bank on it (this also makes it easier for me to tip based on my experience — which is usually how I tip).

    You should pitch this piece to GOOD magazine or another along the same lines, bruh. This is pretty solid.

  2. Kat Avatar

    One thing I’ve noticed while sitting in many a cab: during the wait at a red light, the driver will creep forward every few seconds. My theory is that this practice keeps his meter from switching over to the waiting fare, thus charging the customer full fare for all stops.

    Considering how many circuitous rides I’ve been subjected to (despite my protests, even), and the number of times I’ve been nearly killed by taxis while cycling, I feel my cynical opinion of cab drivers is somewhat justified. At least in this city.

  3. At Home with Kim Vallee Avatar

    The taxi standards have lowered over the last few years. In MTL, I had more than my shares of bad rides so far this year. The other night, my husband and I tweeted on the cab about his poor driving and his awful itinerary choice.
    Whether it is the formula or their training that is at fault I cannot say. But something has to be done to promote a better taxi service.

  4. Dave Avatar

    So… the longer a cabbie takes to get somewhere, the better. The best thing for a cabbie to do (just in terms of the equation) would be to take lots of side streets, and drive as slowly as possible.

    What this equation works out to is:
    (time)/(distance per time)
    = (time squared)/(distance)

    This means that the money accrued for a fare increases with the square of the time taken, and decreases linearly with the distance. I’ll give a few examples, using fare=1000x/y — the same principle applies regardless of what multiplier you insert into the equation:

    Trip 1: 10km
    Average speed: 40kph
    Fare: 1000$*0.25h/40kph = 6.25$
    Average speed: 30kph
    Fare: 1000$*0.33h/30kph = 11.11$
    Average speed: 20kph
    Fare: 1000$*0.5h/20kph = 25.00$
    Average speed: 10kph
    Fare: 1000$*1h/10kph = 100.00$
    Average speed: 1kph
    Fare: 100$*10h/1kph = 1000$

    Now let’s say we have the same destination, but we take a bunch of side-streets:
    Distance: 20km
    Average speed: 10kph
    Fare: 1000$*2h/10kph = 200.00$
    Average speed: 1kph
    Fare: 100$*20h/1kph = 2000$

    Clearly you’re not going to see cabbies taking 10 hours to drive someone somewhere. But I just wanted to point out that what you’re proposing is a quadratic, not linear scale.

    I think what you want is a scale that optimizes fares at a certain average speed — say, 30kph. Then the further away you get from this speed, the more stiffly your fare is penalized. Or, to avoid penalizing cabbies for being in traffic, the speed penalty only really picks up past 30kph (or whatever). So, for instance:

    distance – (speed/30)^4

    would give a graph like:


  5. Martin Avatar

    Interesting theory, but as others pointed out it wouldn’t be fair as rates wouldn’t be linear.

    As far as bad drivers goes, choosing the right company seems to help a lot. In general I tend to avoid unaffiliated taxis, as those are the worst.

    Diamond VIP is generally very good, but unless you’re going to the airport you can not get one easily unless you specify it and ask to pay with credit card or debit card.

    Pontiac Hemlock is also very good in general.

    I hate Taxi Co-Op and Atlas.

    Something I can’t wait to see implemented in Montreal is GPS-based tracking & dispatching service, where all the metrics are monitored by the headoffice.

    They have that in Brampton, and I must say it was one of the best taxi service I’ve seen. Very effective dispatch as they can send you the closest available cab, and by asking your destination they know how much time the ride should take, so they can proactively re-dispatch a new call to the driver. And by having everything tracked, if you make a complaint they can check the records, so it’s in the driver’s best interests to respect the road code… 😉

  6. Dave Avatar

    That link didn’t show up. Hopefully this will:


  7. Whitney Avatar

    Some cities, like Washington DC, actually calculate fares based on what zone of the city you are traveling in, and how many zones you cross- and cabs in more rural areas often charge flat rates. Interesting how geographic norms dictate differences in charges.

  8. John Avatar

    I think I can help here,:)

    I own a cab company here in Carlsbad, CA

    1. Once you get in the cab, the meter starts, $3.00 to start.

    2. $0.30 for every 1/10 of a mile you travel.($3.00 every mile travled.)

    3. When the cab stops at a red light, the meter stops. After 3 minutes the meter will switch over to wait time. My wait time is $50.00 per hour./$0.84 per minute.

    4. Each cab company sets there wait time a little differant from each other. Same as there rates.

    5. As for the driver at a red light slowly rolling forward, all that does is keep the wait time from turning on. He is most likly tring to get the light to change. 🙂

    Hope this has been helpfull:)

  9. Malgorzata Avatar

    Tend to agree with Clarence. Flat rates would be nice. perhaps not so much for the ones taking the taxis quite often. Or am I wrong? Anyway, they do seem to be out of control and taking a cab just isn’t what it used to be… Times have changed… So what are we to do.. Other means of public ransportation does not seem to deliver either…

Leave a Reply

Your email address will not be published. Required fields are marked *