Building an open access fiber network to make Internet for all a reality

NetEquity’s mission is to make high speed Internet access affordable for every human within a decade. We believe the best way to accomplish our mission is by building an open access fiber network in every country.

A fiber optic cable has virtually unlimited capacity which implies that it has negligible marginal cost of serving additional users. When a fiber network is financed under a fixed or formula driven investment return structure, its incentive structure becomes aligned with the ideals of open access. Specifically, the price paid to access such a network declines as usage grows. How low this price falls depends on the cost of deploying and financing a fiber network.

To minimize the price paid to access a fiber network, we will:

1. Deploy fiber along the electrical grid using a technology developed by Facebook and their partners that can deploy fiber at a cost of a few dollars per meter.
2. Finance the fiber network using a combination of Electric Company and telecom operator cash flows under a fixed or formula driven investment return structure.

Today, the majority of Internet infrastructure is not financed under a fixed or formula driven investment return structure because it was financed by private capital that assumed demand risk to begin with. A notable exception is the East Palapa Ring Project in Indonesia, which is a submarine fiber optic network circling the rural parts of eastern Indonesia.

To finance the East Palapa Ring Project, the Universal Service Fund (USF) of Indonesia assumed demand risk and offered private investors a minimum guaranteed return independent of network usage[1]. This financing structure made it feasible to raise $375 million (5.1 trillion Indonesian Rupiah) and deploy 8,450 kms of fiber for a rural connectivity project.

We have developed a globally scalable business model that can build an open access fiber network in every country within a decade. To develop this model we conducted several hundred interviews with industry experts. These interviews helped us map out relevant stakeholder incentives. In this article we will first present a synthesis of our incentive mapping exercise and then describe how our model works.

Stakeholder incentive mapping

We recognize that each country will have different industry dynamics. However, each country will also have a number of similarities. We have chosen to focus on similarities of stakeholder incentives. This section describes incentives of the following stakeholders in a given country:

1. Electric Company
2. Mobile Network Operator
3. Internet Service Provide
4. Universal Service Fund
5. Debt provider
6. Equity provider
7. Management team

Electric Company

An Electric Company grows revenue by adding assets to its Regulatory Asset Base or Rate Base. Rate Base is the value of property on which an Electric Company is permitted to earn a specified rate of return, in accordance with rules set by a regulatory agency. Rate Base is also used to calculate rates paid by an Electric Company’s customers. To ensure electric companies remain efficient and good stewards of their customers, regulators sometimes include incentives that are tied to operating performance.

In developing markets, electric companies are usually government owned and subsidy dependent. Eliminating subsidies provided to electric companies is a priority of most governments. Enhanced network monitoring capabilities can identify technical losses in the electric grid and diversions such as theft. Fixing technical losses and diversions can lower expenditures, eliminate subsidies and lower end consumer tariff.

Access to a fiber network enhances an Electric Company’s grid monitoring capabilities better than any other communication medium because of fiber’s virtually infinite capacity. Existing fiber deployments along the electrical grid are limited to Overhead High Voltage (OHHV) transmission lines where fiber is most often deployed using Optical Ground Wire (OPGW). Our research shows that Overhead Medium Voltage (OHMV) distribution lines are 8x greater in length than OHHV lines [2]. Secondly, OHHV lines are many tens of kilometers away from households, but OHMV lines are typically less than 250 meters away[3].

Electric companies would benefit from a fiber network along OHMV lines but have struggled to justify the high cost of construction to regulators. Regulators are also hesitant to approve fiber investments because fiber is viewed as a telecom asset and most electric companies lack the know-how to lease fiber capacity to telecom providers. While leasing fiber capacity is a challenge for Electric Company staff, they are well equipped to deploy and maintain a fiber network.

Mobile Network Operators

A Mobile Network Operator (MNO) grows revenue by selling voice and data services. The Average Revenue per User (ARPU) for an MNO is declining whereas data consumed per customer is increasing.

~10% of an MNO’s towers (mostly urban and fiberized) contribute more than 50% of revenues. ~50% of an MNO’s towers (mostly rural and not fiberized) contribute less than 10% of an MNO’s revenues. Fiberizing rural towers is technically required by MNOs to keep up with data demand but is not desirable by shareholders because rural areas do not have enough paying customers.

Source: Generalization based on analysis of 10 telecom markets

Telecom regulators were aware of an MNO’s economic incentive to focus on urban areas which is why rural coverage requirements were imposed. MNOs achieved coverage targets but those have not translated into connectivity.

As of December 31, 2018, 3.3 billion people live within a mobile Internet coverage area but are not online[4]. In addition to coverage targets, regulators often impose taxes and/or surcharges to be collected from telecom operators and deposited in a Universal Service Fund (USF). The USF remains underutilized in most countries[5].

While only 10% of MNO towers are fiberized, over ~80% are connected to the electrical grid[6]. This implies that a fiber network deployed by following the electrical grid (referred to hereafter as “NetEquity Fiber Network”) can be shared with MNOs with negligible incremental cost. Thus, a NetEquity Fiber Network will lower an MNO’s expenditures while increasing its network capacity.

Internet Service Providers

An Internet Service Provider (ISP) grows revenues by selling data services. Unlike MNOs, ISPs do not provide mobility and generally have a balance sheet size ~10x smaller than an MNO. The small size of ISPs makes them unattractive customers for most fiber networks.

When a fiber network is financed under a fixed return structure similar to the East Palapa Ring Project, an ISP’s lack of a scale becomes less relevant for the fiber network owner. Thus, a change in incentive structure makes accessing a fiber network easier for smaller ISPs thereby making it open access in letter and spiri. More importantly, when ISPs avoid capital expenditures related to building a fiber network, service levels for end users improve.

The best example of improved service levels driven by the presence of an open access fiber network is from Ammon, Idaho — a rural town of 15,000 residents. The Ammon fiber network was financed partially by the City of Ammon and partially by Ammon residents.

This Ammon network allows its residents to select an ISP through a virtual marketplace and lets residents switch an ISP within 20 seconds if service level is not satisfactory. For ISPs in Ammon, the shared fiber network is desirable because it virtually eliminates their capital expenditure requirements and let’s them focus on serving customers.

ISPs in Ammon can access this shared fiber network for a fixed fee of $50 / month[7]. This low price point is the result of cost-based pricing followed by the City of Ammon. There are 3 ISPs currently operating in Ammon whereas most rural towns like Ammon have only one ISP. It is worth noting that since the Ammon fiber network was launched, the price of a Gigabit connection fell 10x within 3 years.

Source: EntryPoint Networks, Inc.

Note: these prices do not include a fixed $17/month payment made the City of Ammon. This payment lets Ammon residents become owners of the fiber cable coming into their homes over a 20 year period. Ammon residents also have an option of buying the fiber cable coming into their homes for a one time payment of $3,000.

Universal Service Fund

In many countries, the USF’s mandate is to make Internet access available to all citizens. The USF receives contributions from taxes and/or fees imposed on the telecom industry.

A typical USF board is comprised of government officials and telecom industry representatives. This implies that a USF is intimately aware of connectivity demand and is well positioned to assume demand risk of a fiber network. More importantly, a typical USF is cash rich, which means its promise to fund a revenue shortfall will be considered credible by capital providers.

Although over 80% of cell towers are connected to the electrical grid, fiberizing all of them will take some time. Therefore, in the first year of operation a NetEquity Fiber Network’s revenue may be insufficient to cover all costs. Our calculations confirm that the USF of a given country should comfortably be able to cover the resulting revenue shortfall. However, we will ensure that this shortfall is eliminated as quickly as possible.

To eliminate revenue shortfall quickly we will incentivise all stakeholders to maximise the number of cell towers (or endpoints) connected to the NetEquity Fiber Network in the shortest possible time. Specifically, after the USF shortfall is eliminated, the financial benefit arising from more endpoints connected will get shared with other stakeholders as follows:

1. The Electric Company will get a lease fee discount
2. MNOs and ISPs will get an endpoint price discount
3. Equity investors will get a higher financial return
4. Management team will get a higher performance bonus

The specific apportioning of financial benefit will depend on risks assumed by each stakeholder. As a matter of principle, stakeholders that assume the highest risk will be entitled to a greater share of benefit realized.

Debt provider

A debt provider grows its revenue by maximizing interest rate charged to a borrower relative to collateral available. Subsidy dependent electric companies in developing markets often delay payments to manage cash flows. Therefore, debt providers will be reluctant to lend against a subsidy dependent Electric Company’s lease contract alone. MNOs are considered highly credit worthy in most markets. Therefore, a bank will want to collateralize an MNO’s cash flows and USF payments to secure its investment. Providing better security to a bank will help lower interest rates. For context, when debt is 80% of capital, every 1.25% increase in interest rate results in a 1% increase in cost of capital.

A debt provider will give high importance to equity investor quality. Banks will prefer to see well capitalized local business groups as equity investors because such business groups have prior relationships with most local banks. More importantly, involving local business groups reduces project execution risk. For this reason, incentivizing local partners to assist NetEquity during the project development phase will be essential.

Equity provider

Equity providers grow revenue by maximizing returns relative to risk assumed. An equity investor in a typical fiber project is exposed to a number of risks including demand risk. If an equity investor assumes demand risk, their pricing strategy will focus on maximizing equity returns. However, if demand risk is eliminated, then equity investors can become willing to accept a fixed or formula driven investment return structure.

A fixed or formula driven investment return structure is desirable because it ensures cost-based pricing. A cost-based pricing structure in a fiber network is desirable from our perspective because it ensures that price per endpoint gets reduced as more endpoints get connected.

When a financially credible entity like the USF assumes demand risk, equity investors become more willing to accept a lower Internal Rate of Return (IRR). For example, emerging market investors in tower sharing companies assume demand risk and therefore target an IRR greater than 25%. However, the same emerging market investors will accept a mid-teens IRR in a project without demand risk.

The earliest equity capital to get involved will expect the highest returns because it will assume maximum risk, in particular project development risk. As a NetEquity Fiber Network approaches “financial close” i.e. a time when all binding contracts are signed, returns expectations of new investors will fall because project development risk will get eliminated.

Management

The Management team of a NetEquity Fiber Network in a given country will be seeking purpose, career growth, equity upside and salary. The earliest team members to join will assume the most risk and expect the greatest share of equity upside.

The major challenge facing a Management team will be during the project development period, which will likely be 3–5 years. Given the likely restrictions imposed by debt and equity investors, Management team members may not realize equity upside for at least 10 years under base case assumptions.

Negotiating majority ownership for Management locally is an ideal but will be challenging because Management may not have capital to invest. Irrespective, ensuring substantial “skin-in-the-game” for a Management team will be essential because of long project development cycles.

Post deployment, Management’s performance bonus will be linked to connectivity indicators such as number of endpoints connected and percentage of fiber deployed in rural areas. As a matter of principle, we will not link Management performance bonus to profitability growth.

How our model works:

To build a NetEquity Fiber Network in a given country, NetEquity will set up a company domiciled in that country (OpCo). This OpCo will be the owner and manager of a NetEquity Fiber Network. Through this OpCo, NetEquity will:

1. Sign a long-term fiber capacity lease agreement with an Electric Company.
2. Raise long-term debt secured against Electric Company’s lease contract and fiber asset deployed.
3. Lease excess fiber capacity as wholesale lit-service to telecom providers.
4. Raise equity capital by offering equity investors a fixed or formula driven rate of return.
5. Work to convince the Universal Service Fund (USF) to assume demand risk and fund revenue shortfall, if any.
6. Work to convince USF to subsidize the Electric Company’s lease fee if rural fiber deployments are maximized.
7. Deploy optical fiber using Electric Company’s labor force and the technology developed by Facebook along Overhead Medium Voltage power lines.
8. Use Electric Company’s labor force to maintain (passive maintenance only) fiber network.
9. Reduce Electric Company’s lease fee and endpoint price as more endpoints are connected. Beyond an annual endpoint target, incremental returns will accrue to equity investors and Management.
10. Transfer the fiber network to Electric Company after a 25 year concession period.

At present we are in discussions with six electric companies in two countries. During 2020 we intend to expand our dialogue in multiple markets. To learn more please visit our website at www.netequity.net or write to us at info@netequity.net

Acknowledgement

We want to express our sincerest gratitude to the team Facebook Connectivity (FBC) for giving us early stage access to the fiber deployment technology mentioned in this article. Specifically we want to acknowledge the immense contributions made by Karthik Yogeeswaran in showing us how to reason from first principles.

Sources

1. Financial Close of Palapa Ring East Project — March 29 2017; https://ptpii.co.id/index.php/en/news-publication/news/financial-close-palapa-ring-timur/86
2. This figure is an average across 13 markets where we obtained grid data from public filings.
3. Interviews with electric companies` in the United States, South Africa and Pakistan.
4. 2018 State of Connectivity by GSMA; https://www.gsma.com/mobilefordevelopment/wp-content/uploads/2018/09/State-of-Mobile-Internet-Connectivity-2018.pdf
5. 2013 Universal Service Fund study by GSMA; https://www.gsma.com/publicpolicy/wp-content/uploads/2016/09/GSMA2013_Report_SurveyOfUniversalServiceFunds.pdf
6. Average based on analysis across 10 markets
7. Unique Model Makes Citizens a Funding Partner in Broadband Network; https://www.dailyyonder.com/unique-funding-model-makes-citizens-funding-partner-network/2017/10/05/