• Home
• ICT
• Telecommunications
• Benchmarking
• Mid-2006 Report
• Benchmarking Methodology ...

• Telecommunications

• Benchmarking

• Mid-2006 Report
• Comment on NZIER Report
• Media Statements
• Archive

Methodology Issues

68. The use of representative "baskets" of telecommunications services is a widely accepted method of benchmarking the price of key telecommunication services between countries. An internationally recognised method of benchmarking a number of telephone services has been developed by the OECD that is based on an internationally agreed demand pattern as opposed to any one country's pattern.

69. The relative accuracy of benchmarking methodologies, such as the OECD tariff basket comparison methodologies, is limited by the following factors:

• the basket of services used to benchmark relative performance of each OECD country may not be representative of the typical average user demand within a country;
• they adjust imperfectly for the differing prices across countries of inputs to local services by using a suitable exchange rate, such as a purchasing power parity exchange rate;
• they do not readily take into account the effects of network density including scale and scope economies;
• they do not readily take into account productivity differences between countries that are outside of the control of telecommunications industry managers and regulators;
• some services may not be directly comparable (such as different speed broadband Internet access services);
• the methodologies do not provide uncertainty bounds for assessing the reliability of the resultant rankings.

70. In comparing prices between two countries, the two primary methods that are typically used are Purchasing Power Parity (PPP) and some form of longer term averaged Monetary Exchange Rates (to average out the influence of non-trade related factors). This report uses the OECD PPP rates as they focus on the relative purchasing power of non-traded telecommunications services and avoid problems involved with real exchange rates which can be driven by a range of factors unrelated to the price of goods and services.

Service Provider Input

71. Subsequent to finalising this benchmarking report the Ministry sought comment, from key stakeholders, Telecom, TelstraClear, Vodafone and TUANZ.

72. Telecom expressed a number of concerns, its main concerns in summary were:

The relative performance of residential and business PSTN services as reported in the draft excludes international calling from the PSTN baskets. The omission of international calling distorts the comparison of New Zealand's industry with other countries, as it does not consider the total basket of PSTN services that users purchase. In order to ensure that there is a fair and reasonable comparison of PSTN services, the Ministry should consider all PSTN services by not excluding international calling from the business and residential PSTN baskets.

The Ministry has chosen the OECD purchasing price parity (PPP) to convert prices in a local currency to United States dollars (USD). Telecom submits that the appropriate conversion rate is the long term average of the nominal exchange rate. However, in the absence of a long term average data set, it is proposed that the PPP and nominal exchange rates used in the Teligen T-Basket used to define a range for the comparison.

The OECD mobile service usage baskets differ significantly from the usage patterns of Telecom's mobile customers. In particular, the average Telecom mobile customer sends many times more text messages than is assumed in the OECD mobile services basket. This draws into question the validity of the OECD Basket in this case and thus the benchmark for drawing any robust conclusion.

73. Vodafone expressed a number of concerns, its main concerns in summary were:

The figures you report lack credibility. The Teligen figures are error-prone and subject to significant limitations as a measure of what is actually going on in the marketplace. We have argued that these figures are unreliable for the last two years. It seems highly questionable that only now, when the figures are improving, you decide to exclude some of the results. It is particularly questionable when you do not have any information on the plans for the other countries in the Teligen data.

You are not reporting the correct figures. You dismiss the use of any figures for Vodafone because you have decided that consumers do not benefit from three year terms. But the Base plans only make a significant difference to the ranking for low users. In fact our You Choose plans are responsible for the price falls for medium and high users, and your figures should reflect this.

The picture you present is misleading. There have been dramatic reductions in the prices of mobile services over the past year that are not covered in your report. Consumers continue to benefit from increasingly intense mobile competition, and your report does not refer to any of these developments.

The table below shows August figures from Teligen over the last three years compared with the OECD as a whole. The price reductions for medium and high users are particularly large, with New Zealand prices falling more than half. Price reductions for low users are three times the OECD reduction over the same period.

These falling prices are a critical part of the story and of our strategy. This is missing from your report at present.

Table 1: Teligen Results over Time (USD PPP)

	August 2004	August 2005	August 2006	Change 2004 to 2006*
Low user
Rank	28	23	21+
NZ	233	204	197	-15%
OECD	172	160	164	-5%
NZ/OECD	135%	128%	120%
Medium user
Rank	30	30	24
NZ	794	695	393	-51%
OECD	429	407	312	-27%
NZ/OECD	185%	170%	126%
High user
Rank	30	30	21
NZ	1,412	1,235	582	-59%
OECD	737	690	527	-28%
NZ/OECD	192%	179%	110%

* This is the August 2006 result divided by the August 2004 result minus 1.

⁺ I calculate New Zealand's ranking at 21 for low users (rather than 20). There are two other very small differences, in the % of OECD average for medium users and the annual cost for high users, which I assume are a result of rounding.

74. On 25 January 2007 Vodafone expressed concern that the lack of inclusion of capped call and text pricing in the OECD/Teligen pricing comparisons means that the figures do not accurately reflect the prices available to New Zealand consumers, and provided the following further comment in support of this point of view.

The Teligen data for August suggests that on Vodafone NZ prepaid plans across all three user baskets average revenue per minute was approximately 72 NZ cents.

In fact, in August 2006 Vodafone's revenue per minute on prepaid plans was […] cents. In August 2006 we offered capped 10 minute calls to anyone at any time in any of 15 countries for no more than $2 dollars. In addition we were offering free texts during weekends.

For illustrative purposes a prepaid plan based on a flat rate of […] NZ cents per minute would rank us […] in the OECD in the low user basket, […] in the medium user basket and […] in the high user basket.

Note, the above […] Vodafone revenue data is commercially sensitive and was provided on a confidential basis.

75. TelstraClear said that its key comment relates to reliance on pricing data for mobile services that have a three year contractual commitment. It considered that these plans, where a customer accepts a three year contract in exchange for lower calling prices, are unlikely to reflect the commonly occurring retail mobile charges in the New Zealand market. Accordingly, there is a significant risk that the benchmarking report may under-represent the true cost of mobile services in New Zealand.

76. TelstraClear supported the proposal to seek clarification from the OECD Secretariat as to the appropriateness of including plans requiring a three year term in the benchmarking analysis. TelstraClear considers that such plans are unlikely to reflect the commonly occurring retail mobile charges in the New Zealand market.

77. Copies of Telecom's and Vodafone's comments on the benchmark report are attached in Annex 1 [Note: the web publication does not include copies of the Annex 1 correspondence].

Ministry Response to Service Provider Input

78. In reply to Telecom and Vodafone the Ministry notes that:

• the use of purchasing power exchange rates (PPP) is widely accepted as a reliable method of comparing the prices of non-tradable goods or services between countries. In general official OECD publications usually use this approach to compare telecommunications services prices between countries;
• the inclusion of international calls in the fixed telephone service basket is optional and on balance the Ministry considers that they should not be included. The main reasons that underlie this are: the comparison of international call prices between OECD countries by a basket methodology is problematic as countries international call communities of interest vary widely; and the international calls market is very competitive with a wide range of suppliers and levels of call quality (from PSTN grade to best efforts VoIP grade);
• the cellular benchmark results are based on the OECD cellular comparison basket methodology, a comparison methodology widely used by OECD countries;
• the new OECD cellular baskets definitions came into force early in 2006 following revision in mid 2005 by a process into which both Telecom and Vodafone New Zealand had the opportunity to input their views;
• Teligen⁴ applies the rules when selecting tariffs for OECD basket comparison results, not the Ministry;
• the basket/call tariff rules in general preclude one off or short term specials;
• it is up to service providers to ensure that Teligen has up-to-date data on current service plans;
• Vodafone New Zealand has been regularly in contact with Teligen;
• the OECD/Teligen cellular basket methodology tends to underestimate the price of cellular plans with included minutes per month where the service provider rounds call duration up to the nearest minute, because the Teligen cellular basket methodology does not take this into account (i.e. Teligen prices on the call duration specified in the basket definition). This can be significant, for example including rounding up to the nearest minute increases the total basket minutes by ~13% for the high and medium user baskets and ~36% for the low user basket;
• the OECD basket methodology is not set up to track relative price over time and should not in general be used for this purpose;
• there were problems with the February and May 2006 Teligen cellular basket results which were caused by a faulty macro. The problem was fixed by Teligen in the August 2006 results release;
• the non-inclusion of capped call and text pricing in the cellular OECD/Teligen call pricing comparisons does not mean that the results do not accurately reflect the prices available to New Zealand consumers. The OECD tariff comparison baskets are periodically reset to reflect call usage in mainstream OECD countries. Where a capped price call plan option is the cheapest available plan the OECD/Teligen methodology will select that plan;
• many OECD countries also offer cellular capped price call options or an equivalent the benefits of which are not captured by the OECD/Teligen tariff comparison basket. For example, Optus in Australia offers turbo recharge options that provide either $120 of calling credit for $30 or $250 of calling credit for $50. The Cingular "Pay As You Go" plan allows users to make unlimited mobile to mobile calls to more than 55 million Cingular customers across the United States for $1 per day and 10 cents per minute for all other calls ($1.00 per day charged only on days phone is used). Such options typically require a certain minimum monthly payment in order to access the capped or reduced price calls.