Print

Comment

Email

Digg

Del.icio.us

Reddit

Investment is necessary for improving telecom sector performance. And investment risk is the primary determinant in making investment decisions.

Investors consider/face risks related to three broad environments: a) macro-level country risk, associated with factors that often affect the entire economy such as inflation, foreign exchange fluctuations, and political stability; b) market or commercial risk, associated with factors related to demand and supply, availability of substitute products, and the performance of competitors; and c) regulatory risk, emanating from government action, including but not limited to, actions of the regulatory agency with authority over the telecom sector.

LIRNEasia, a non-profit research and capacity building organization established in 2004, with operations across the Asia Pacific, conducted the Telecom Regulatory Environment (TRE) survey in eight emerging Asian economies in the second half of 2008. The countries included in the study were India, Pakistan, Bangladesh, Sri Lanka, Maldives, Thailand, Indonesia and the Philippines.

The small size of the Maldivian telecom market (population 300,000; number of access paths slightly higher) and the close-knit community in Male and surrounding islands mean that all stakeholders are known to each other, and are often reluctant to express their views about the regulatory environment candidly, even in anonymous surveys. Hence, we are not showing and comparing the result for Maldives.

Market Entry
The Market Entry TRE scores reflect stakeholder perceptions about the conditions for entering and operating in the market. Transparency of licensing, ease of obtaining a license, barriers to entry and growth are some of the factors affecting this. The clear winner is Pakistan, getting a high TRE score of 3.9 for market entry.

Pakistan has one of the fastest growing mobile industries in the world (estimated to be the 3rd fastest growing, behind India), with 58.9 access paths per 100 people compared to 26.22 in India; but with some downward corrections needed for inactive SIMs. Growth has been driven by investments, primarily very large foreign direct investments. Several factors contributed to making Pakistan's telecom attractive to investors.

• The conditions for obtaining a new mobile license or renewing and existing one are straightforward-each operator has to pay $291 mn. Though this fee is high, once it was announced it eliminated discretion on the part of the regulator, and therefore nearly eliminates regulatory risk-as long as an operator can pay the fee, access to the market (in the form of a new license or a renewal of existing/expiring one) is guaranteed.
• The unbundled licensing regime for fixed services has encouraged investors to enter the market, and offer services in the area of their choice.
• There are no limitations on foreign ownership of telecom companies and no restrictions on merger and acquisition activity.
• Mobile number portability was implemented in 2007, thereby enabling increased competition amongst players, and giving a reasonable shot at success even for new/smaller players.

The above actions have made Pakistan's telecom sector an attractive destination for regional and international investments. For example, during the time-frame of the research, China Mobile acquired 100% of Paktel; Orascom increased its ownership stake in Mobilink to 100%; SingTel purchased 30% of Warid Telecom; and OmanTel purchased 60% of World Call. The amount paid by the acquiring companies in the above deals was over $1.5 bn.

All firms have already started making significant investments in new infrastructure or upgrades. During 2007-08 Pakistan's telecom sector attracted over $1.4 bn in investments that amounted to around 27% of total FDI into the country. Sri Lanka and Thailand received the lowest TRE score for 'market entry' in the mobile sector. Sri Lanka's low score can be explained by the non-transparent nature of the licensing process used in the most recent granting of a license. Bharti Airtel was awarded a new license in April 2007. No auction mechanism was used, and the criteria for selection were never explicitly stated.

Furthermore, after obtaining the license, it was a good 21 months before Airtel was able to become operational due to setbacks and delays related to rights of way, interconnection and a host of other issues. At the time the TRE survey was carried out in Sri Lanka, most stake holders (and even the general public) were expressing concern over the difficulties Airtel was having in starting its operations.

Thailand too suffers from less-than-ideal market entry conditions in the mobile sector, giving it one of the lowest scores in the region. There are three private concessionaires operating in the Thai mobile sector. But this number is low, given the total size of the market, and the level of competition, as measured by HHI, is well above 3,500, indicating a low level of competition. No new mobile licenses have been issued since the original three concessions were granted. Furthermore, political wrangling and related legal problems have prevented (or at least significantly delayed) the rollout of 3G services.

Allocation of Scarce Resources
Though scarce resources were defined as spectrum, rights of way and numbering, spectrum is often the dominant resource that is foremost in the minds of stakeholders. This is perhaps not surprising given that subscriber growth in all three sub-sectors, fixed, mobile and broadband, in all the countries, has been driven by wireless technologies.

Once again, Pakistan is the top performer, receiving the highest scores in all three sub-sectors. Pakistan is, in fact, the only country that receives above average (above 3.0) scores. The 1996 Telecom Act requires the regulator (the Pakistan Telecom Authority) to 'receive and expeditiously dispose off applications for the use of radio-frequency spectrum'. The Frequency Allocation Board is required to process applications for spectrum within thirty days by law. In addition, real-time frequency monitoring takes place, ensuring that license conditions are enforced. Finally, in April 2008, Pakistan changed the numbering scheme for telephones from seven digits to eight digits, thereby lowering the scarcity of this resource also.

In contrast to Pakistan, India receives the lowest scores in this dimension in all three sub-sectors. India was the lowest performer in the previous (2006) TRE survey also. In fact, India's latest TRE scores for this dimension have marginally decreased since 2006. At a fundamental level, this is due to the purely administrative (as opposed to economic) allocation of spectrum that is practiced in India. Repeated recommendations to auction the spectrum have gone unheeded.

The amount of spectrum allocated to each operator is linked to the subscriber numbers, not usage (and even here, DoT and Trai were in disagreement for most of the 2007-08 year on what the appropriate subscriber-linked formula should be). There is no policy for allocating spectrum beyond the 10MHz that is already allocated, even though most GSM operators have loaded their spectrum well beyond benchmark levels and the CDMA operators are close to reaching that point.

The average frequency allotted to an Indian mobile operator is 6.2 MHz, compared to the world average of 17.18 MHz. Even if further allocation is agreed upon, there is no spectrum free to allocate: government and defense users are currently occupying valuable bands, making re-farming difficult. In addition to all of the above, before and during the time the TRE survey was being carried out in 2008, India's telecom space was abuzz with controversies related to 2G and 3G spectrum allocation. Accusations by various parties about undue advantage or preference being given to the other, and the DoT ruling out auction mechanisms for the allocation of 2G and other related issues were being publicly and widely debated. Given all this, India's low score in this dimension is not surprising.

Interconnection
TRE for interconnection assesses aspects related to interconnection rates, mechanism for setting those rates, interconnection locations, time taken to obtain interconnection, mechanisms for sharing of related revenue and related costs. Pakistan is the best performer. The rules in Pakistan make it mandatory for each operator to negotiate, and have interconnection with another operator who makes such requests. And in practice, operators indeed negotiate their interconnection rates mutually. However, the players with significant market power (SMP) are then required to produce/publish a reference interconnection offer (RIO) detailing the terms they offer to other players.

SMP is defined as any operator who has more than 25% of the revenues in a specified market. Of the countries in the study, Pakistan is one of the few countries that mandate RIOs to be published on a regular basis.

Pakistan's TRE score for Interconnection has increased by nearly a point since the 2006 TRE study (for the mobile sector it was 2.8, compared to 3.7 now). This could be due to the increased efficiency with which interconnection disputes are mediated and resolved by the regulator.

In contrast, Thailand performs poorly in all three sectors. The reasons are to be found in the conflicting rules and regulations, the lack of clear authority on the part of the regulator and the discriminatory nature of interconnection rules. The Thai Business Act (TBA) states the interconnection is mandatory; that interconnection charges be negotiated privately; and that the interconnection terms should be non-discriminatory and fair. TBA even sets out procedures for dispute resolution and mandates decisions to be given within thirty days.

All this is for naught, however, because all concession agreements are exempt from such rules. And all private mobile networks operate under concession agreements with the two state enterprises-TOT and the CAT-and are legally owned by these two entities. Since private operators are mere contractors, all interconnection charges must be negotiated and paid for by the two legal license holders only.

The terms imposed on the concessionaires are onerous and discriminatory-for example the concession mandates that all private concessionaires of CAT (namely DTAC and True Move) pay TOT a hefty flat fee of 200 baht (about $5.8) per month for each post-paid subscriber and 18% of revenue for each pre-paid subscriber, while TOT's own concessionaires do not have to pay such a fee. In protest, the two mobile operators had stopped paying interconnection charges to TOT since 2006, and in 2007 the three private concessionaires agreed to interconnect amongst themselves. TOT has filed a law suit against at least one of the private operators as a result.

Methodology

The Telecom Regulatory Environment (TRE) is a tool that enables the quantification, to the extent possible, of regulatory risk in telecom. It uses a survey to gauge the perceived effectiveness/ineffectiveness of the regulatory and policy environment that affects the telecom sector. The results, when compared historically, can indicate improvements or declines over time of regulatory risk. The differences in scores for different aspects of regulation and policy provides a very useful diagnostic tool for understanding how different policies and regulatory actions are perceived. The results when compared across countries can point to countries with lower regulatory risks, thereby giving potential investors a good indicator on where to invest (all other risks remaining equal). But cross-country comparisons must always be undertaken with care, especially across countries with different political systems and social structures. For example, the unique features of micro states such as the Maldives prevent them from being compared with larger countries. The TRE instrument asks informed stakeholders to rate (on a Likert scale of 1 to 5, 1 being highly unsatisfactory, 5 being highly satisfactory) the Telecom Regulatory and Policy Environment in a country along seven dimensions. Five of the seven dimensions are based on the GATS (General Agreement for the Trade in Services) reference paper on telecommunications. These are Market Entry, Allocation of Scarce Resources, Interconnection, Regulation of Anticompetitive Practices and Universal Service Obligation. The QoS dimension has been added, given its importance. The three sub-sectors of telecom-mobile, fixed and broadband-are evaluated separately. The respondents to the survey fall into three different categories: Category 1 Those directly involved in the sector such as operators, equipment vendors. Category 2 Those indirectly impacted by the sector or those studying/observing the sector with broader interest such as consultants and lawyers. Category 3 Those who represent the broader public interest such as media personnel, other government officials, retired regulators, civil society organizations. It is well known that the larger the number of respondents, the less biased is the survey. However, the goal here is to measure perception among senior level decision-makers, and stakeholders who have expert and in-depth knowledge about the various aspects of the regulatory and policy environment in a country (as an example, category 1 respondents are limited to CEOs or other CXOs, the heads of regulatory departments or heads of key functional units within operators). Therefore, instead of surveying junior level stakeholders or persons not directly affected by regulation and policy, reasonable targets for ideal, minimum number of respondents per country have been identified. The minimum number of respondents per category is 15, making a total of at least 45 respondents per country. In micro-states (such as the Maldives, with a population of just 300,000 citizens), it not possible to find 45 senior level stakeholders (at the CXO level or equivalent). As such, the minimum number of responses per category in micro states is 5, making a total of at least 15 per micro-state. A total of 416 responses were received from senior level stakeholders in the 2008 survey. Each category should contribute equally to the final TRE score, reducing bias. However, in these types of surveys it is not possible to control how many completed questionnaires will be returned. As such, it is not always possible to obtain an equal number of respondents from each category. Therefore, statistical weighting is used to equalize the contributions made by each category. The survey is administered through multiple modes: via the Internet, on paper, or through personal meetings. Though Afghanistan was also in the list, due to logistical difficulties in traveling to the country, the Afghan TRE survey could not be carried out simultaneously with the above countries. We are not showing the result for Maldives because of the small size of the Maldivian telecom market. The results for the seven country study are presented by dimension. Furthermore, micro states have particular characteristics that make their markets and therefore their regulatory requirements different from other states. As such, it is not often possible to pick best (or worst) practices from micro states and make them applicable to other countries. In the graphs, countries are grouped by region (South Asia and South East Asia), and ordered in ascending order of GDP per capita within region. The minimum possible score on the Likert scale is 1, and maximum is 5. Average performance is 3.0, and is marked with a red dotted line.