The Indian Economy Blog

June 13, 2007

Auto Rickshaw Pricing

Filed under: Infrastructure,Uncategorized — Karthik @ 11:22 pm

Acute shortage of parking space means I usually take an auto rickshaw whenever I have to go to the M G Road area. The usual fare is Rs. 35, so when the meter shows Rs. 45 (as it did last week), I demand an explanation from the driver. And the last time round, the driver’s explanation was quite interesting.  “You can pay what you think is appropriate, sir. But what to do? There are so many traffic jams nowadays we are forced to tamper with the meter”.

What the driver pointed out is an interesting concept when it comes to pricing something like an autorickshaw ride. The basic concept is that there is not one (fuel, as is generally assumed) resource that gets consumed in ferrying a passenger, but actually two – fuel and the driver’s time! And it is important that both these are taken into consideration while setting fares.

However, what is currently being done (at least in Bangalore – not the case in Mumbai) is that fares are loosely linked to fuel prices, and whenever there is a major hike in fuel prices the auto drivers’ unions demand and get a fare hike. The last increase I remember was maybe a couple of years back, when the fare was set at Rs. 6 per km. To their credit, the regulators (the government) also fixed rates for “waiting time” (I usually don’t make autos wait, so I don’t really know what it is).

Coming back to the problem of two resources, suppose an auto driver works fixed hours. Say 10 hours a day. On an average he travels 20 km every hour, and if his margin is Rs. 2 per km, he makes Rs. 40 an hour, or Rs. 400 a day. Now, suppose the speed of traffic suddenly slows down and our man can do no more than 10 km per hour. His operating profit for the day is effectively cut by half!

This problem, of a “two-dimensional resource” (dimensions here being distance (or fare) and time) is similar to the one that we face in Cargo Revenue Management (where weight and volume of the goods are the dimensions). In the cargo scenario, we work around the problem by taking into account the “density” of the shipment, and having separate fare structures for different ranges of densities. Could we think of a similar solution in the context of auto-rickshaw?

Just to elaborate, the way we deal with this problem in cargo is by fixing a “standard density” (us cargo guys are funny people – we define density as volume/weight! and in the airline context, standard density is usually 6 m^3/kg). Now if the density of the shipment in question is greater than the standard density, it means that it is in a way “heavy on volume”. And the way we deal with them is to charge them based on volume and not on weight. Shipments with density lesser than the standard density are charged based on weight (they are clearly “weight heavy”).

If we want to incorporate this concept in the auto-rickshaw world, the concept analogous to density would be “average speed” – the average number of billable kilometers an autorickshaw runs during an hour. We could pick a “standard” value for this average speed. If the average speed during the time of travel is greater than this standard speed, charge the customer by distance. Else charge by time.

Of course, implementation of such a complex system is not at all feasible so we need to look at simpler options. Mumbai has its unique way of solving this problem – in that the meter runs even when the auto is stationary (provided the engine is running). So in case of slow moving traffic/ traffic jams, the driver is adequately compensated. There are a couple of issues with this system too  – that the driver could deliberately go slow, and that the engine needs to be kept on during jams.

Are there any other simple solutions to this problem? One option could be peak hour rates – at times of the day when traffic moves slower than average (average is quite slow nowadays), auto-rickshaws should be allowed a premium. A premium in order to compensate for the greater strain on the resource that is driver’s time. It would be something like if you get into an auto between 8 and 11 am or 5 and 8 pm you have to pay a 20% premium on the meter fare. Implementation of this could be through fare charts like those they have in Mumbai.

Of course, note that this peak hour premium is very different from night fares. The latter is in place because the supply is low at that time. The former needs to be there to compensate for strain on additional resources in a regulated market!

I’m not saying peak hour premiums is the perfect solutions to this problem (please leave comments with what you think would be a good solution). All I’m trying to say is that the constraints on auto-drivers’ resources need to be carefully noted, and adequately compensated (i’m reminded of Naveen’s point in one of his posts regarding directional demand for auto-rickshaw traffic). A lot of people might say they are high, but I believe the current auto-rickshaw fares in Bangalore are unrealistically low – creating unrealistic expectations from the consumers on one hand, and encouraging cheating by the auto drivers on the other.

Remember that fair deals are far easier to implement than unfair ones – which take an extraordinary degree of policing. It is hence an expectation that cheating on the part of auto drivers will reduce in case of fair setting of prices. Of course, there will still be nothing to stop the driver from cheating, but in case of a fair deal I believe that policing should become easier which will further reduce possibility of cheating.

Oh, and for the sake of not leaving a story hanging, I paid that auto driver Rs. 35 – what I believed was the “fair” or “normal” price (and btw, it was an off-peak hour ). And he happily accepted it and moved on.


  1. Test

    Comment by MadMan — June 14, 2007 @ 10:41 pm

  2. In Bangkok (where traffic moves very slowly at the lights) taxi meters keep running at a slower rate while the car is stationary. This solves your compensating-driver-for-traffic problem. I’m not sure how much a new meter would cost though.

    As far as “peak-hour rates” are concerned, they are already implemented in a de-facto manner. The auto driver demands “10 rupees extra”, “1.5 meter”, “80 rupees saar” etc depending various parameters such as destination, time of day, rain, mood etc

    “…. A lot of people might say they are high, but I believe the current auto-rickshaw fares in Bangalore are unrealistically low ….”

    I find that hard to believe, especially when you admit that a lot of people feel otherwise. Maybe a comparison of fares/fuel prices across the metros would help.

    Comment by Prasun — June 15, 2007 @ 1:16 am

  3. Sir, the truly free market solution escapes you entirely.
    More at my blog.
    Also I thank you and this amusing post for kick starting my blog.

    Comment by Gurmeet — June 15, 2007 @ 1:16 am

  4. The article misses one important element of Bangalore’s Auto rickshaws. The tendency of Auto rickshaw to go for “Out-of-Meter” settlement of fares. This practice doesn’t have anything to do with missing “Average Speed” element of the ‘fair’ fare pricing puzzle. The basic motivation stems from the fact that Bangalore’s public transport system (buses, I mean) is unable to remove or eliminate high commutable dependency on auto rickshaws. This factor alone may become a stumbling block on any fair pricing model approach that we come up with.

    So until and unless somebody works towards addressing this issue and commuters become more disciplined to not give in to the “Out-of-Meter” demand, we will not be able to come up with a solution we are looking for.

    Comment by Mohit — June 15, 2007 @ 11:05 am

  5. Interesting. Please excuse my introduction of a comparative perspective on the Indian economy blog. A number of these concepts, to compensate for the ‘opportunity cost’ and to cater to instances where one benchmark does not adequately reflect the resources spent, seem to have been covered by taxi operators in Singapore. You should be able to see a laundry list of charges that may be levied on commuters in Singapore on the taxi sections of the SMRT website. They even have special charges for street pick-ups at the Airport to adequately recompense the drivers for the long wait-times and serpentine queues to pick-up passengers at Changi. Somehow, in comparison to most other ‘developed’ countries, the taxi charges in Singapore remain relatively low.
    However, the margins earned by the taxi drivers when stalling/waiting are far lower than those earned by them on the move (assuming that the average speed through Singapore’s roads is roughly about 60 kmph). Obviously, they don’t take very kindly if they are asked to wait or halt! The cab drivers in Singapore would not like to be denied the chance to clock earnings like the Sri Lankan cricket team (Sidhu once commented, quite regrettably, that the Sri Lankan score was running an Indian taxi meter!)!

    Comment by Skeptical Blogger — June 15, 2007 @ 12:41 pm

  6. This is the millionth time im trying to post.. But talking about Sri Lanka..

    As outlandish as it may seem, an obvious solution is to do away with the pricing system.

    i have been in Delhi a few times and i was appalled with how the ‘auto rickshaws’ work there. sometimes the rickshaw wallah refuse to take you somewhere because its not feasible with the rates set, then there were times i walked on long in the the middle of the night when no autos were available.

    You should take a leaf from Sri Lanka’s book, which imparts almost 0 regulation on auto rickshaws (what we call 3-wheelers) the prices are negotiated on the spot, and because there is competition (no licenses) the prices remain as low as feasible. you can get three-wheelers at any hour of the night provided you can negotiate the ride for a good amount.. and they rarely say they cant go anywhere. and for some reason the drivers also seem nicer..

    now it may be that i’ve been stuck at the wrong place at the wrong time, and i did find the system working for the most part. but generally i find the situation here in SL much much better. sure you get screwed over occasionally, but that’s just like any other business.

    I’m also told Delhi has a cap on the maximum number of autos it allows which inevitable leads to corruption and bullying by the authorities…it seems the license-raj in India is still very much alive.

    Comment by Deane — June 16, 2007 @ 12:07 pm

  7. Deane, you make an excellent point here. the “ideal” solution would be the SL system – no meter and negotiated fares. it’ll be perfect markets at work.

    and the reason for the cap i think is more to do with the available space on the road (cities like bangalore are already way too congested). i mean for once i see some kind of a rational argument to put the quotas. need to think whetehr removing caps on no. of ARs will actually work.

    mohit and prasun, thing is as deane has explained, “out of meter” is perhaps actually the best way of pricing, but supply limitations mean we need to regulate the industry, unfortunately.

    and people say autos are exorbitatn because they usually upt “out of meter settlement” nad charge teh passengers more than what they believe is the “fair” value. if the official rates are higher, the feelign of being cheated is lower, and it’ll “appear” more reasonable. the actual fare won’t change. just hte perception

    Comment by Karthik — June 16, 2007 @ 11:12 pm

  8. And in case you have all missed a rather unique explanation of meter tampering in Delhi, see this:

    Comment by Shefaly — June 17, 2007 @ 11:15 am

  9. Economics says perfect price discrimination (charging everyone a different price, according to their willingness to pay, exactly as Skeptical Blogger suggests occurs in Sri Lanks) in the is an optimal solution. But this only works if the stipulations for a perfect market are there: high number of small sellers, high number of small users, low costs of entry, prices are comparable between sellers easily.

    Clearly, these are all true for rickshaw drivers. I use the same analysis to see how call girls should (and are) priced in Amsterdam (

    Comment by pratiksrandomwalk — June 17, 2007 @ 11:21 pm

  10. Recommending no meters and negotiated fares is a step backwards to the previous generation where our parents dealt with horse carts.

    But coming to the point of the discussion – I think in cities like Bangalore distance is redundant. Meters should simply be revised by time. Fix rates by the hour. Half hour would be half that etc. I don’t think it’s possible to really go much faster or slower than the rest of the flow.

    Comment by Nikhil — June 20, 2007 @ 11:24 pm

  11. Bombay is my native and these days DC and NY are two places I juggle between. I must say NY taxi metering is the best. I have traveled Bombay lately and have negotiated fares with Rickshaw as well as Taxi wallahs. I do take in consideration the time factor but that is definitely not a niche segment for anybody to claim considering the competition available (Humans to do the job). Countries like US and some EU nations are scarce with humans to help people commute whereas that is not the case in India.

    I had a wonderful explanation during one of my sales pitch in Bombay – there is lot of talent available in this country and lot of talented people with lot of time on their hands. Justify your talent and pricing as opposed to your competition?

    I was swept away by this question and have changed my perception about selling anything in India. Just put the time/price situation of a taxi/rick driver in the above hypothesis.

    Comment by Dhaval — June 22, 2007 @ 7:51 pm

  12. The real problem with the auto rickshaw system is not in the fares, but in the regulation. The number of auto rickshaws wery very limited for a long time. Even today plans to increase the number of rickshaw licenses are met with opposition. People have come to see these licences as a means to a lifelong livelihood.

    I am enthused to find an example of zero regulation working in Sri Lanka. If it will work there it will work here too, but being the realist that I am, I know that there is no chance for that to happen. (at least for the next 20 years)

    Comment by rishi — June 27, 2007 @ 9:27 am

  13. speaking of SL auto systems… this system pretty much runs in the outskirts of cities like mumbai, pune, etc. the trend here seems to be something of this sort… the auto-rickshaw drivers form a union and set fares (reasonable or not, is open to debate) which are certainly notches higher than the normal Rs. 6/km trend in the metros… well… the reason they usually state is that they charge ‘half-return’ as they ‘may not’ get a fare back… (which i believe is rarely the case)…

    so leaving the market directly to the suppliers is like invigorating the ‘invisible hand’ and letting the market equilibriumize (pardon the verb) on its own…! i fear the repurcussions… in places like b’lore where i guess the rest of the public transport isn’t too much to talk about, auto-wallahs might just get the better of the general public…

    Comment by Abinav Kumar — July 14, 2007 @ 2:19 am

  14. Deale, Karthik

    sorry but your reasoning is not correct, and if I might say so, a tad naive. For the “perfect market” to exist, perfect anti-cartel rules need to be in place. Thus, a perfect market is never an unregulated one, though the mistake is often made. An unregulated market (“free pricing”) would only lead to arbitrary price-hiking depending on the knowledge of the customer and transparency is lost.

    Like someone else said, free pricing would be a huge step backwards.

    The problem is quite easily solved. A combined fare of price/hour and price/km. For example: 10 RS * km + 30 RS * hour.


    Comment by Martin — January 15, 2008 @ 5:39 am

  15. Hi Deane,

    The system of negotiable pricing does not seem practical, for the following reasons:

    1. A person who is well-versed with all the routes and distances might be able to strike a good deal, but an outsider/tourist would be royally fleeced.

    2. In a negotiable price scenario without any regulation, instead of competing with each other, the auto-rickshaw drivers will smartly gang up among themselves, mutually deciding upon a high fare to charge the passengers for specific routes. There would be a monopoly of groups of autorickshaws in various areas, leaving the passenger no choice but to pay the expensive fares.

    3. Auto-drivers using petrol-driven vehicles would naturally have to charge higher than those with diesel-driven vehicles, thus leading to loss of business for them. A passenger would prefer to go in a diesel auto, as it would invariably be cheaper for him.

    Lastly, if negotiable rates are allowed, one would need to pay through his nose to travel at odd hours (late night, early morning).

    There surely has to be a strict regulation when it comes to auto-rickshaw prices. The above-mentioned system, where a combination of time and distance is used to calculate the fare seems most logical, though difficult to implement as well.

    My solution – design a meter which internally stores the “expected time” for each kilometer. e.g. Assuming an average speed of an autorickshaw as 30 km/hr, it should take 1 hour to travel 2 kilometers. As soon as the meter starts up, a stopwatch also starts internally, which calculates the actual time required for travelling. Depending on how much the actual time differs from the expected time, add or subtract from the conventional fuel charge. What do you feel?

    Comment by Vivek S — February 13, 2008 @ 1:29 pm

  16. Just gave you a link on my blog.
    Greetings from Ireland.

    Comment by John McCartney — May 1, 2008 @ 11:49 am

  17. Having frequently used autos in Bangalore for about an year in 2007, I have the following observations:

    1. Seven out of ten times, autos in Bangalore will refuse to go wherever you want to go specially if its a short distance travel.

    2. Have seen them sitting idle for hours together implying they know 3-4 long rides a day on their LPG-auto (LPG cost is half of petrol) will make their day.

    3. Eight times out of ten , the meters show higher fares implying they cheat out of habit than an economic necessity.

    4. None of the autos have protection-curtains from rain.. Passenger service is not their botheration.

    A similar situation with amby-diesel taxis was present in Kolkata about 7-8 years back. The problem was found to be in the supply side. Within a year, taxi licenses were doubled and today Kolkata taxis ply at Rs.10/ a Km and go wherever you want them to go. Meters are still tampered but passengers experience convenience of travel.

    Significantly increase the auto licenses in Bangalore till the mass transit system (metro) comes into play and the problem will be solved.

    Comment by Salty — May 5, 2008 @ 2:31 pm

  18. Fuel Consumption and Environmental Impact of Rickshaw Bans in Dhaka

    Most trips in Dhaka are short in distance, usually one to five kilometers. These trips are perfect of Rickshaws. Rickshaws are cheap and popular mode of transport over short distances. Rickshaws are safe, environmentally friendly and do not rely on fossil fuels. Rickshaws support a significant portion of the population, not only the pullers, but also their families in the villages, the mechanics who fix the rickshaws, as well as street hawkers who sell them food. From the raw materials to the finished product the Rickshaw employs some 38 different professions.
    Action needs to be taken to support the Rickshaw instead of further banning it in Dhaka. The combined profits of all Rickshaws out earn all other passenger transport modes (bus, rail, boats and airlines) combined. In Dhaka alone, Rickshaw pullers combine to earn 20 million taka a month.

    We think that over the coming holiday of Eid du Ajah, new Rickshaw bans will be put into action on roads in Dhaka. Eid was used in the past to place new bans on roads in Dhaka. Last Eid many roads were declared Rickshaw free without public support or approval. By banning Rickshaws roads are clogged with increased private car use as well as increased parking by cars.

    Banning of Rickshaws on major roads increases the transportation costs for commuters. Not only due to longer trips to avoid roads with bans in effect, but also due to actually having to take more expensive forms of transport such as CNG or Taxi, where in the past a Rickshaw would suffice. The environmental impact of banning Rickshaws is obvious because it exchanges a non-motorized form of transport for a motorized form of transport, thus increasing the pollution and harming the environment. Rickshaw bans harm the most vulnerable in society, mainly the sick, poor, women, children and the elderly; generally those who can not afford or do not feel comfortable on other forms of public transport. To ban Rickshaws also hurts small businesses that rely on them as a cheap and reliable form of transporting their goods.

    Rickshaws are ideal for urban settings because they can transport a relatively large number of passengers while taking up a small portion of the road. In 1998 the data showed that Rickshaws took up 38% of road space while transporting 54% of passengers in Dhaka . The private cars on the other hand, took up 34% of road space while only transporting 9% of the population (1998 DUTP). This data does not include the parking space on roads that cars take up in Dhaka . If included this would further raise the amount of space taken up by private cars. Every year the Rickshaw saves Bangladesh 100 billion taka in environmental damage.

    The government makes many efforts to reduce traffic congestion in Dhaka but with no success. Blaming Rickshaws for traffic congestion and subsequently banning them from major roads has not had the desired affect. Traffic is still as bad now as it was before the Rickshaws were banned on major roads. Rickshaws thus can not be seen as the major cause of traffic congestion. Instead one should look towards private cars and private car parking on roads as the major cause of traffic congestion. The space gained by banning Rickshaws is often used for private car parking. The current trend in transport planning reduces the mobility of the majority for the convenience of the minority. The next time a ban on Rickshaws on another road is discussed please take into consideration who is being hurt and who is being helped. For a better transport system in Dhaka we need to create a city wide network of Rickshaw lanes. If this is done Dhaka can reduce its fuel usage dramatically as well its pollution. We ask your help in our fight to keep Dhaka a Rickshaw city. Any information or help is very much appreciated and sought after. I write you this letter to describe the difficulties we are facing and some solutions but they are by no means exhaustive and we look forward to your help and input.

    Syed Saiful Alam shovan
    Volunteer of Save The Environment Movement?

    Comment by Syed Saiful Alam — May 15, 2009 @ 12:16 pm

  19. Well Karthik, its difficult to concur with the point that the Auto Rickshaws effectively under charge. It could very well be because of the reason that you haven’t been in Bangalore since long time, or you might not have had necessity to use auto rickshaws .In Bangalore , I strongly feel that the auto-rickshaws effectively charge sky-rocket proices. I think that the amount of money that they charge creates a fear psychosis in people who would like to use autos. God save you if you are caught in the rain in a relatively desolate area.

    However, I do agree with the system proposed by you. Nevertheless, in order for that system to be implemented, the current “cartel” of auto rickshaws needs to be broken. It has to be regulated strictly by law , wherein overcharging or refusing would be a punishable offense , not only on paper , but in actuality. However, I understand that its not politically feasible for the govt. Metro trains hold a great promise and would be a great leveler of the current auto-monopoly. Once that happens , the alternatives suggested by you could be considered seriously and enforced. Till then , auto wallahs don’t deserve any such compensations for time.

    Comment by Siddharth — May 26, 2009 @ 3:22 pm

RSS feed for comments on this post. TrackBack URL

Leave a comment

Powered by WP Hashcash

Powered by WordPress