Dynamic Pricing for a Dynamic World

By Yonatan

We live in a congested world and the trend of urbanization is only making it worse. Given limited real estate and financial resources, governments and municipalities use various allocation strategies to combat congestion and accommodate growing populations and vehicle use. Dedicated public transportation lanes, toll roads and HOV lanes are all means to encourage use of public transportation, increase vehicle utilization (i.e. carpooling) and tax vehicle use, while increasing productivity and reducing GHG emissions. In that sense, and as discussed in class, toll roads are an efficient means to capture time benefits via tariffs.

But what better way to force drivers to internalize the externalities associated with vehicle use than through dynamic pricing mechanisms? With advancements in technology and increased information collection, toll lanes with truly dynamic pricing – determined real-time based on current traffic in order to maintain a desired minimum speed – are now a viable, and quite attractive, option.

Dynamic pricing is not a new concept, yet still needs to be used with care. Consumers tend to be dissatisfied with what may be viewed as a cynical abuse of one’s needs. Coca-Cola, for example, learned this firsthand when it tried to introduce temperature-based pricing in its vending machines. Immediate consumer and media backlash led them to quickly stop the initiative [1]. Nevertheless, in some industries consumers have become adapt to such pricing schemes. Airlines, hotels and car rental companies all price based on real-time supply and demand, as do more and more online services and retailers[2]. So when is dynamic pricing appropriate? INSEAD prof. W.J.Reinartz identified 5 conditions required for price customization to work[3]:

  • Customers must be heterogeneous in their willingness-to-pay
  • The market must be segmentable
  • There is limited potential for arbitrage
  • The cost of segmenting and policing must not exceed revenue increases
  • It must not breed violations of perceived fairness

Though his research is focused on online services, it seems appropriate for other industries as well and transit in particular. Clearly, the main contention point is that of perceived fairness. In the context of toll roads, there is obvious concern that the imposed fee for using a road excludes the service from certain consumers who are not willing, or just unable, to pay for it. The counter-argument is that such arrangements can fund expansion of existing roads and transit infrastructure in a manner that alleviates some of the congestion burden in other parts of the system. Moreover, such routes don’t have to be restrictive for all vehicles. Such is the case in the Israeli “Fast Lane” project – an expansion of the road from Jerusalem to Tel-Aviv – in which public transportation and HOVs can ride free of charge[4]. The dynamic pricing mechanism can then be viewed as revenue-maximizing which in turn allows for a greater cross-subsidy.

Whether you find it fair or not, it is clear that the increasing convergence of information technology and infrastructure services (e.g. water and electricity) will lead to increased use of dynamic pricing as a means to manage scarce resources in our daily life.

[1] NY Times, “Why Variable Pricing Fails at the Vending Machine”, June 27th 2005: http://www.nytimes.com/2005/06/27/business/27consuming.html

[2] Information Age, “Liquid Commerce”, May 12th 2007:


[3] “Setting Prices in an Online World: When Price Customization Works (… And When it Doesn’t)”, Werner J. Reinartz, April 2001,


[4] Tel-Aviv Fast Lane, accessed 2/10/2013: http://www.shapir.co.il/en/content/tel-aviv-fast-lane


2 thoughts on “Dynamic Pricing for a Dynamic World

  1. Yonatan,

    Like Stacie, I like the idea of dynamic pricing for tolls.

    My only question is on the phasing of the investments. To me, for this system to be fair, there must be robust, viable alternatives to driving, most importantly a low-cost public transportation system able to handle a significant increase in passengers at peak times. Without this system, tolls seem very unfair. If the idea is for public transit to be financed by these tolls, that is the wrong approach; the transit system must be in place BEFORE the tolls.

    On that note of timing, then, it seems like one option is to implement tolls very gradually, on limited areas and in very low amounts, use that revenue to fund increases in capacity, and then continue to build out the system. Still, some upfront funding will be necessary.

  2. I grew up in the San Francisco Bay Area and used the bridges all the time growing up. Whether it was to get to a Golden State Warriors game in Oakland or to visit my grandmother in Fremont, crossing the bridges used to always cost a dollar. Over the years, however, prices have risen – partially due increased maintenance costs and partially due to the need for increased state revenue.

    The Bay Area Toll Authority tends to agree with the theory Yonatan has proposed by moving from a strictly flat toll to a variable toll that takes into consideration 1) carpool status, 2) time of day, 3) FasTrak electronic toll collection, 4) clear air vehicles and 5) number of axles. For example, on the San Francisco-Oakland Bay Bridge:
    • 2 axle vehicles between 12:01 a.m. and 5 a.m., between 10 a.m. and 3 p.m., and from 7 p.m. to midnight, Monday through Friday—$4.00
    • 2 axle vehicles between 5 a.m. and 10 a.m. and between 3 p.m. and 7 p.m., Monday through Friday—$6.00
    • 2 axle vehicles between 12:01 a.m. Saturday and midnight Sunday— regular toll: $5.00
    • 3 axles regular toll of $5.00 times 3 axles – $15.00, etc.
    • HOV or low-emission vehicles with DMV sticker with FasTrak receive reduced rate of $2.50
    • FasTrak additionally gets a discounted toll on the Golden Gate Bridge

    So, I tend to agree that transportation is only the beginning for variable pricing models in the public infrastructure space. However, I am concerned that many of the variables that we take for granted in the US or Israel are completely uncontrollable in developing countries. So, while I do see the future for dynamic pricing in improving efficiency in developed countries, the upside in developing countries seems more limited. In the example of Mexico City, the amount of corruption and water access inequality makes a system of differential pricing as unfair and unlikely.

    Source: http://bata.mtc.ca.gov/index.htm

Comments are closed.