In view of media interest on the compensation methodology, IDA would like to further explain the methodological approach. Our consultants had recommended the compensation methodology and the compensation amounts payable to the two companies. This process was conducted with due diligence...Singapore, 12 September 2000 | For Immediate Release
In view of media interest on the compensation methodology, IDA would like to further explain the methodological approach. Our consultants had recommended the compensation methodology and the compensation amounts payable to the two companies. This process was conducted with due diligence and was presented to the Government. The Government is satisfied that the process was robust, and the numbers are fair and reasonable. IDA is releasing the details on the methodology, and will make available its consultant for a media briefing on Thursday morning. A separate invitation will be issued shortly. The following is a summary of the compensation approach.
The compensation covers services subject to Public Basic Telecommunication Services (PBTS) licensing, and includes local and international telephony, local and international leased circuits, and public switched data services. The compensation calculation excludes non-PBTS services, such as mobile communications and Internet access services, as these service markets are already open to competition.
Specifically, a discounted cash flow (DCF) approach was used to estimate the difference in the net present values, for SingTel and StarHub respectively, between two scenarios:
I. Status Quo - duopoly from 1 April 2000 until 1 April 2002 followed by full competition; and II. Accelerated Liberalisation - full competition from 1 April 2000.
The approach first projects the market size for PBTS services in Singapore. It then estimates the market shares for SingTel and StarHub, respectively, and finally estimates the differences in earnings and net present values under the two scenarios.
The net present value for a service provider is the sum of discounted free cash flows. The discount rate is the after tax weighted average cost of capital (WACC) which reflects the required rate of return for debt and equity holders. The WACC incorporates the impact of debt financing in the discount rate rather than in the cash flows. Accordingly, free cash flow in a given year is cash remaining from revenues once expenditures (both capital and ioperating), changes in working capital and tax are subtracted.
A ten-year modelling duration has been employed, as the impact of Accelerated Liberalisation is expected to be most pronounced through this period. Projections longer than 10 years would be less reliable.
A terminal value, based on discount rates and growth rate assumptions, is employed to represent the present value of the free cash flows for each of the two service providers in the period beyond that covered by the financial model.
Revenues for the service providers are modelled as a function of the total market size, the share of market that the provider achieves and the price that it charges. Market size forecasts have been developed based upon historical growth in Singapore and consideration of information from sources such as investment analysts, regulators, and research organisations. Generally, growth rates are forecast as combinations of "natural" rates with premium rates reflecting stimulation due to competition. In the Accelerated Liberalisation scenario, the stimulation due to competition is adjusted to reflect earlier competition and to approximate parity in market size at the end of the modelling period for all PBTS services with the exception of international data. Market shares have been forecast by considering experiences in other countries, investment analyst forecasts and Singapore market conditions. Similarly, forecast price evolution has been based on that observed in other countries post competition, overlaid with a consideration of Singapore's current price levels, historical patterns and unique features.
Operational expenditure (opex) and capital expenditure (capex) have been modelled at an aggregate level. Direct variable costs have been estimated using direct traffic cost per minute and estimated outgoing international traffic. Other opex has been estimated using an aggregate ratio of other opex to revenues and an assumption regarding variability. Capex has been estimated as a steady state percentage of revenues.
Variability due to uncertainty in the inputs has been addressed using an established statistical methodology, known as Monte Carlo simulation. This is an approach whereby the output is calculated many times using different randomly selected inputs based on an estimate of the distribution of those inputs. This is tantamount to running many "what-if" analyses using random inputs within their valid ranges. This yields a distribution of possible outputs which provides insight into the range of outputs and the likelihood associated with a given output.
Further explanation of the methodology is contained in the Annex, which is an extract from the report produced by IDA's consultant, Deloitte Consulting.
ISSUED BY CORPORATE COMMUNICATION DIVISION
INFOCOMM DEVELOPMENT AUTHORITY OF SINGAPORE
Notes to Editor
The IDA engaged Deloitte Consulting to assist in working out the compensation amounts. The press release issued by IDA on the compensation to SingTel and StarHub, on 11 Sep 00, can be found in the Media Room of the IDA website at www.imda.gov.sg.
About Infocomm Development Authority of Singapore
The Infocomm Development Authority of Singapore (IDA) is a dynamic organisation with an integrated perspective to developing, promoting and regulating info-communications in Singapore. In the fast-changing and converging spheres of telecommunications, information and media technologies, IDA will be the catalyst for change and growth in Singapore's evolution into a vibrant global info-communications technology centre. For more information, visit www.imda.gov.sg.
For media clarifications, please contact:
Manager, Corporate Communication
Infocomm Development Authority of Singapore
Tel: (65) 6211-1999
Fax: (65) 6211-2227
The approach to arriving at the appropriate compensation is to use an earnings based model to estimate the difference in the net present value (NPV) of SingTel's and StarHub's projected earnings with and without the Accelerated Liberalisation. The model projects the growth of the Singapore market and the share of the market achieved by SingTel and StarHub. Projected revenues and costs are then used to arrive at earnings differentials for SingTel and StarHub for the public basic telecommunication services (PBTS) of interest.
To estimate compensation payable, the model compares two scenarios:
1. "Status Quo": a duopoly moratorium for two years from 1 April 2000, followed by full competition introduced from 1 April 2002; and
2. "Accelerated Liberalisation": the accelerated introduction of full competition from 1 April 2000.
For each scenario:
First, the Singapore market for PBTS is projected based on expectations regarding market evolution;
Second, the market shares achieved by SingTel and StarHub are projected for those services; and
Third, earnings for SingTel and StarHub are estimated based on expectations of prices and costs, and are used to calculate the respective net present values (NPVs).
The difference in the net present values of earnings under these two scenarios provides an estimate of the level of compensation payable.
Net Present Value
The net present value for a service provider p (SingTel - ST, or StarHub - SH) is the sum of the discounted free cash flows:
where rp is the discount rate for service provider p. The modelling duration is ten years, with a terminal value to perpetuity applied. Data available from international experience regarding market share evolution are most robust for the first few years following the introduction of competition. A ten-year forecasting duration is used as the impact of Accelerated Liberalisation is expected to be most pronounced through this period and the reliability of telecommunications industry forecasts decreases with longer timeframes.
Free Cash Flow
Free cash flows CF have been modelled as a function of revenues R, operational expenditures O, capital expenditures C and taxes T:
where t = 2001,2002...2010
The estimation model does not address the impact of changes in working capital, as they are considered negligible and expected to contribute little to the compensation in comparison to other key variables such as market growth rates. The free cash flows also exclude the impact of debt financing - i.e., they are calculated as if there were no debt. In this context, the discount rate required is the after-tax weighted average cost of capital (WACC), which recognises that any investment will be financed by both debt and equity - i.e., the impact of debt financing is reflected in the discount rate rather than the cash flows.
Conceptually, the revenues for each service provider Rp,t, for each service of interest svc, in year t, will depend upon the total market size (usage) Msvc,t, the share Sp,svc,t each service provider achieves, and the average price Pp,svc,t, that each service provider
where t = 2001,2002...2010
The model focuses on services subject to PBTS licensing. The model also does not consider other services, such as mobile or Internet access, that are outside the scope of the PBTS license on the basis that these services were not protected by the moratorium.
In general, the market size for each service Msvc,t depends upon starting market size Msvc,1999 for that service and expected annual growth Gsvc,t for that market over time:
, where t = 2000,2001...2010
The starting market size Msvc,1999, has been set from data provided by SingTel, IDA, and third party sources.
The growth in services has typically been modelled as the "natural" growth of the market GN,svc,t plus a competitive stimulus factor GC,svc,t:
where t = 2001,2002...2010
GN is the natural growth in the absence of competition, and GC is stimulation (incremental growth above natural growth) due to competition driven decreases in price as well as non-price factors such as increased choice, improved customer service and increased advertising. For both scenarios, natural growth will be the same over time, whereas growth stimulation due to competition will differ due to the timing of entry of competitors. Under Accelerated Liberalisation, the competitive stimulus is adjusted to reflect earlier competition and to achieve parity in market size at the end of the modelling period.
SingTel's share for a service in a given year SST,svc,t is a function of SingTel's share in the previous year SST,svc,t-1, and the decline in that share in that year SDST,svc,t:
where t = 2001,2002...2010
SST,svc,2000 is SingTel's share in 2000.
The decline in SingTel's share is calculated by estimating the gains made by other service providers including StarHub.
The evolution of prices for each service is based on that observed in other countries post competition, with consideration given to Singapore's initial price levels and historical price change patterns.
Conceptually, SingTel's price for a service in a given year PST,svc,t is related to SingTel's price for the service in the previous year PST,svc,t-1, and the change in that price during the year, PCST,svc,t:
where t = 2001,2002...2010
StarHub's price for a service PSH,svc,t, where assumed different from SingTel's, is calculated as a discount relative to SingTel's price PST,svc,t:
where t = 2001,2002...2010
Operating and Capital Expenditures
Generally, the level of operational expenditure (opex) is expected to be responsive to changes in the level of business activity. As the level of business turns down from the Status Quo case to the Accelerated Liberalisation case, opex is expected to also turn down.
Capital expenditure (capex) for SingTel and StarHub is expected to be lower under Accelerated Liberalisation than under Status Quo on account of lower market shares.
The appropriate discount rate is the weighted average cost of capital (WACC). The after-tax weighted average cost of capital is equal to the weighted average of the after-tax cost of debt and the cost of equity, where the weights are based on the debt to equity ratio in the capital structure.
Capital Structure: In calculating the WACC, a capital structure appropriate for a basic services company operating in Singapore has been used; this may differ from the actual capital structure of each of the two companies. The actual capital structure may reflect factors other than those applicable to the basic services business in Singapore.
The appropriate capital structure that minimises WACC is a function of market requirements and is best estimated by what capital markets require for similar types of companies. It is not possible to directly determine what capital markets require; however, it is likely that actual capital structures will tend to reflect those requirements. As a result, an average of actual capital structures for selected telecommunications companies was used to estimate the appropriate capital structure with appropriate adjustments for each company to reflect their circumstances.
Cash flows have been modelled out to FY10. A terminal value has been used beyond FY10 to estimate the value from later years, as the reliability of the projections decreases with length of time. The same approach is used for both scenarios.
The terminal value is intended to represent the present value of the net cash flows for each of the two companies in the period beyond that covered by the financial model. It is the estimated value as of the end of the period covered by the model. As a result, it is discounted for the period from the end of the model to current period.
If it can be assumed that free cash flow (i.e., cash from operations less capital expenditures and taxes) will grow at a constant annual rate for an extended period into the future, the terminal value (i.e., the present value of the future cash flows) equals the free cash flow in the last period of the model multiplied by one plus the growth rate and divided by the discount rate less the growth rate, i.e.
To apply the above formula in estimating the terminal value, it is expected that future free cash flow will grow at the assumed growth rate each year.
The above approach is consistent with the discounted cash flow approach. It is the same as carrying the model forward with free cash flow growing at the assumed growth rate.