|
Publication Date: February 2007
Over the past ten years, regulators around the globe have issued spectrum for broadband wireless access (BWA). With the recent standardization of both fixed and mobile WiMAX, more regulators are auctioning WiMAX spectrum. Still, the patterns are not always uniform, and spectrum pricing varies from market to market. This report takes an in-depth look at the drivers of WiMAX spectrum pricing. We also ran a benchmark analysis of more than 1,000 licenses awarded in more than 20 markets across the world, as well as a correlation analysis of WiMAX costs against a number of key indicators, including GDP per capita, mobile and broadband penetration, urban vs. rural areas, etc. We analyze the results of those correlations in this report, and provide an assessment of the trending of WiMAX spectrum costs at a time when such key markets as India, the UK, and Malaysia are readying to award more spectrum.
The data used in our analysis has the following characteristics and constraints:
-
We looked at the licensing of 2.3 GHz, 2.5 GHz, and 3.5 GHz spectrum in more than 20 countries located on six continents.
-
Most countries award spectrum through auctions. A minority goes for beauty contests, where the frequency is allocated to the operator with the best weighted sum of credentials. Therefore (and although we did include a few markets with beauty contests), our data is largely influenced by licenses sold in 17 countries through an auction process.
-
Variations exist in the data used to calculate the price per MHz per population (PMP). For example, the estimate for Australia is based on a single license. Where multiple licenses have been awarded, we either use individual licenses or a country average of all the licenses. South America data includes Brazil only.
-
The US auctions of 2.3 GHz and 2.5 GHz occurred in 1997 and 1996, respectively. Mobile penetration, broadband penetration, GDP per capita, and exchange rates used represent values at the time of each auction in a particular country.
-
It should be noted that the bidding behavior of auction participants is often random, spontaneous, and subjective, leading to winning bids which may not always be correlated with factors which should affect the market price of spectrum under normal conditions.
Our benchmark analysis is based on comparisons of price per MHz per population (referred to as “PMP” throughout this report). PMP equals the price paid for the license divided by the amount of spectrum in MHz in a given frequency band divided by the population covered by the license. PMP enables an apples-to-apples comparison of license costs. License fees are the result of an assessment of a number of factors, including the size of the spectrum allocated (often referred to as blocks, lots, or channel pairs) and the addressable market, which is restricted geographically. Without PMP, the information from two countries cannot be compared reliably. For example, in Iraq, a license was sold which covers about 24 times as many people as a license in Denmark, but the size of the actual spectrum allocated was almost four times smaller. Such dynamics have an impact on Capex, and ultimately on bid levels, as will be discussed later in this report.
Exhibit: PMP by Country, US$*
You may purchase this report using one of the following options:
Online: Click the ‘Buy’ button, view contents of cart and then checkout.
Phone: For Europe, Africa, Middle East and Americas, please call +1.617.494.1515. If you are in the Asia-Pacific region, please call +852.2525.8455.
Fax: Download our order form complete and fax to +1.617.494.8898.
If you have any questions, please contact us directly via phone or email jmcelroy@pyr.com
Author: Dan Locke, Analyst, CMT
|
