Intelligent benchmark regulation: Forbearance within benchmark limits

Posted on April 11, 2007  /  3 Comments

This colloquium will be on a new paper that is being developed on tools for intelligent benchmark regulation, based on Harsha de Silva and Tahani Iqbal’s presentation on Price & Affordability Indicators at the WDR Expert Forum in Singapore. The tools under consideration are price baskets and price elasticity of demand.

Type of regulation:

Rate-based Return on Regulation
This is no longer appropriate in today’s context of fast changing mobile markets because this kind of regulation does not give operators the flexibility to change prices.

Price Cap Regulation
This is not applicable in settings (like Sri Lanka) where inflation is high.

Price forbearance
This type of regulation is different from de-regulation. Forbearance is where regulators take a decision to desist from regulation unless otherwise specified. This has worked in India, as per TRE scores. Although it may not be effective in countries like Maldives, where there are a limited number of operators.

What is being suggested is forbearance within limits or bounded forbearance. For example, if a price limit has been set, the regulator can only intervene if the operators charge above or below the price.

As such, two tools that will be useful for regulators to look at:
Price comparisons – for accurate price/cost comparisons to set appropriate benchmarks
Price elasticities of demand – for predicting effects of regulation on industry revenues

Price Comparisons –
OECD uses t-basket (Teligen basket) which is updated regularly. The basket takes into comparison the account airtime, SMS use, rental, connection charges and free airtime and SMSs. The monthly tariffs cheapest tariff package of the largest operator (in terms of market share) is used to calculate the cost of service for three levels of use: low, medium and high.

How will three different baskets be used to obtain one price benchmark? Using the weighting based on low, medium and high users.

In order to compare regional prices accurately, there is a need for regional weighted averages. Baskets need to be calculated for post and pre-paid tariffs, RPP has to be taken into account where necessary. Differences in the cost of living should also be taken into consideration by converting prices to PPP.

For the purpose of regulation, raw basket data gives best comparison. For others, PPP is a better comparison.

When baskets for India vs Sri Lanka were calculated, Sri Lankan tariffs were found to be more expensive. But because India has special characteristics, the weightings etc are all skewed towards India while the Sri Lankan weight are almost negligible.
So why regional benchmarks? Small island and landlocked countries (according to UN definition) can be grouped together, etc.

Setting the benchmark – the lowest price in the region should not be the basis for setting the price range for the region because this will disadvantageous for smaller countries in the same region (for example, case of India and Bhutan).

Price elasticities of demand –
This tells us how responsive demand is to small price changes. Once regulators have identified the benchmark band, and it operators maintain prices within the benchmark limits, then no actions are necessary on the part of the regulator.

However, if the prices of operators are outside of the benchmark limits, instead of clamping down on operators, regulator must first consider the impact of regulating prices downward, on industry revenues. The necessary action depends on the value of price elasticity of demand.


  1. For the purpose of devising an instrument for banded forbearance (BF), there is no need to work with low-, medium-, and high-user baskets (those baskets are needed for different purposes). For BF all that is needed is a simple, single benchmark figure. This means that the minutes of use reported by the Sri Lankan operator and the Indian regulator are enough; there is no need to apply the OECD ratios (which can be criticized as inferior to real ratios developed for use studies in the region itself).

    Simplicity is best in regulation.

  2. Your brief concerning benchmark regulation helps me get a handle on benchmark criteria to consider for regulating electricity tariffs. The issues identified are relevant. Can you point me in the direction to material that may develop the issues in more detail and/or identifies and develops other issues? I am preparing for a presentation about benchmarks for utility regulators in Eastern Europe and Russia. Thank you.

  3. Thank you for your interest and comment. However, I am afraid, that particular line of work has kinda stalled (due to pressures of other commitments) so all that we have is what you have seen. We hope to get it restarted soon, so would appreciate any material you can share with us.