Skip to main content
"When a flower doesn't bloom, you fix the environment in which it grows, not the flower" - Alexander Den Heijer

For the fourth interview in the spectrum series, we spoke with Dr Carlos Rey-Moreno and Steve Song. Carlos is a member of APC’s staff and a telecommunications engineer with a background in development studies. Carlos has been involved in rural connectivity projects in underserved areas for more than 10 years, initially with the Enlace Hispano Americano de Salud (EHAS) Foundation in Latin America and, more recently, as one of the founders of Zenzeleni Networks, a community-owned internet service provider (ISP) providing affordable communications in rural Eastern Cape, South Africa. Steve Song is a researcher and consultant working to expand the use of wireless technologies through shared spectrum strategies and to enable greater internet access throughout Africa and other emerging markets. He is also the founder of Village Telco, a social enterprise that builds inexpensive Wi-Fi mesh voice over internet protocol (VoIP) technologies to deliver affordable voice and internet options in underserved regions.

Carlos and Steve have recently put the final touches on their research paper “Innovations on Spectrum Management” to be published jointly by the Internet Society, Mozilla and APC in the coming weeks (a draft version can be accessed here and an official link will be provided when available), where they explore the importance of enabling new actors to access electromagnetic spectrum to close the digital divide, offer recommendations for better spectrum regulation, and emphasise the need to democratise the discussion around spectrum management. 

During a conversation with APCNews, Carlos and Steve explained how the electromagnetic spectrum works, explored the history and evolution of its regulation and shared their cautious optimism about a “significant uptick in the growth and sustainability of community networks.”

APCNews: While spectrum has been on the radar in technical and policy discussions around internet access for a number of years, many of us have little understanding of what it is or why it’s important. Can you give us a brief breakdown of the electromagnetic spectrum and how it works?

Carlos Rey-Moreno: Some time ago, someone realised that by combining electric and magnetic signals (that's why we talk about electromagnetic spectrum), you could actually communicate wirelessly. The range of frequencies at which you can actually send information wirelessly is what we call spectrum. There are different frequency bands inside data spectrum of electromagnetic waves that you can use for different communication modes. We use the word “spectrum” in many ways in common vocabulary, to describe something that ranges from one extreme to another. In the case of electromagnetic spectrum, it’s related to the speed at which electromagnetic pulses oscillate. There are very low oscillation rates up to very high oscillation rates. One particular receiver and one particular transmitter use a chunk of that range of spectrum to convey information. So, the spectrum then is divided into many bands that are used for many different communication modes, and interference may happen in any of these bands.

Steve Song: The reason why spectrum is governed, managed and regulated is due to one significant factor: the issue of interference. This means that one transmitted communication using electromagnetic spectrum can interfere with the ability of another transmission to be recognised. In all electromagnetic communication, you have a sender and a receiver, and when you have more than one sender and receiver, it is possible for one transmission signal to be confused with another. This has led to a process of spectrum regulation to prevent this problem. Because there was lots of spectrum around in the early days, it was very easy to manage this simply by assigning specific frequencies to specific purposes. If you want to do wireless telegraphy, you use one frequency; if you want to radio broadcast, you use another. 

That helped a great deal, but the next problem was that if you had two radio broadcasters operating in the same area, this could lead to problems of interference. So, the next step after separating the function of the individual frequencies was to determine a way of protecting organisations using spectrum from interference. What emerged was a “real estate model” for spectrum where the use of spectrum over a geographic area was assigned to specific organisations. 

For a long time, this was effective because there was lots of spectrum and lots of “real estate”, but that has all changed in recent times because of the popularity of wireless spectrum for doing all kinds of things: for broadcasting, for mobile communication, for the internet of things (IoT). So now the demand for spectrum exceeds the availability. The original metaphor of real estate that was used, which was useful for a long time, is no longer useful because it is an inefficient way of thinking about spectrum. Spectrum is not like real estate – it is not physical in nature, it’s actually a very complex thing to understand and can be managed in a variety of ways. 

A better way to understand spectrum is to think of it more like sound waves or like ripples in a pond. When you drop a stone in a pond, those waves are easy to see as they ripple across the pond, and when you drop a stone in the other end of the pond, you can see those ones too. The interesting thing is when those waves meet each other, they interfere in the sense that they cause larger and smaller waves, but they do pass through each other. Similarly, it is possible for spectrum to coexist, as long as you have techniques for the right receiver to recognise the right transmitter. 

There are many ways of doing this. For instance, if you think about being at a cocktail party (sound waves are a great way of thinking about radio waves) and someone across the room is speaking your native language, whereas everyone else is speaking your second language, you can actually hear that speaker across the room, quite clearly. This is an example of radio encoding which makes signals easier to recognise. Similarly, if you are at a concert with a rock band playing, and you are having a conversation with the person next to you, it doesn't interfere with the rock band playing. So the actual proximity between transmitter and receiver, along with the loudness of the transmission, are both factors that affect the ability to receive communication. The point we try to get across in this paper is that there are many techniques for more effectively managing spectrum than simple real estate metaphors that give away long-term leases over large geographic areas, and we should be exploring them. 

APCNews: How is spectrum currently regulated and how has the regulatory landscape evolved in recent years?

CRM: Many governments see the regulation of spectrum as a way of earning money because of the high value that society and, therefore, operators can get out of that spectrum. In the technical kind of evolution and ownership that Steve was explaining before, we need to add the fact that the regulation of spectrum at the moment has a significant component of political economy, because most governments are having a crisis of income and they are looking at this as a way of increasing their budgets. I think there has been a big transformation on that front; the value placed on spectrum by society and by the government and operators has changed and therefore it is regulated in a different way. There is a tendency to regulate it in a way that maximises the money the government can make from it. 

SS: As demand for spectrum began to exceed the availability, there was a debate on the best means of making spectrum available. Economists such as Ronald Coase argued and ultimately convinced regulators that the best way of equitably offering spectrum to organisations was to use an auction, based on the principle that whoever valued the spectrum the most would then pay the most for it and make the most use of it. But, that has turned out to have some negative consequences. 

One is that spectrum auctions can be used to exclude others. Big operators will sometimes bid on spectrum because it is valuable enough to them to pay for it just to exclude others from the market. Also, the huge sums of money that are being paid at auction for spectrum have an impact on network roll-out. This is less true in the industrialised world where the comparative wealthiness of the economy is not affected so much by the amount of money paid for spectrum. But, in the global South, where huge sums of money are still being paid, for example, in Ghana and Nigeria recently, sums in the neighbourhood of USD 50 to 70 million, this has a real impact on the actual roll-out of the network, because a substantial amount of the money needed for the network is now tied up in a simple licence. This has a knock-on effect on pricing as well, because that money has to be recovered from somewhere, and that somewhere is the consumer. 

This presents a challenge for economically poor countries because this is a very easy source of revenue direct to the treasury of the government, yet by reducing the price of spectrum, the actual economic benefits that accrue are likely to be much greater. The spread of communication infrastructure is a kind of general purpose enabler for the economy, which is more difficult to measure in terms of direct income, but the evidence points to it having a much more significant impact. There is a “short-termism” problem with spectrum auctions at the moment, in terms of governments and their desire to generate revenue. [Licences] are typically for 10 to 15 years, and governments are elected every four years, so if you happen to be around at the time there is a spectrum auction, are you [the government] going to think about the economic impact for 15 years, or are you going to just take the USD 50 million and use it for whatever pressing need you have? For poor countries, it is a very difficult decision. 

APCNews: In your paper, you mention the importance of starting an inclusive public debate around spectrum allocation and regulation. Why do you think this conversation is so important, how could it be initiated, and what would it look like? 

CRM: I think there is a growing realisation that, if we don’t understand the infrastructure and physical elements of the internet, then there are other debates about net neutrality, internet shutdowns, privacy, security, access and affordable access that simply cannot be addressed, because this information is in the hands of very few. I think the way forward is to democratise how that infrastructure works and raise awareness about why this is important: why does all the spectrum go to the hands of three to four companies in every country? It’s a problem for the country, for civil society and for the government. More diverse competition and a more diverse set of stakeholders using these resources is beneficial for everyone. It's a win-win. 

As Steve was saying, the ways that spectrum has been allocated so far is only working for some, it isn’t working for everyone. But there are more efficient ways of utilising it if you start to understand it and propose other approaches. For me, it is about awareness. Once the awareness is there, then it’s about participation in any available public engagements, such as the public consultations that we do in South Africa. From there, we can start the conversation with politicians and regulators to actually implement and change this trend where it’s assumed that big multinational mobile operators are going to connect everyone.

SS: I think the challenge for civil society organisations (CSOs) is understanding how to get into the conversation – how do you become part of it, who do you talk to, what are the opportunities for engaging in a dialogue? An obvious opportunity is when there are national broadband plans being developed. This is a key part of that conversation and it’s important to be prepared to bring issues like spectrum into it as well as an understanding of the market and current trends. We are in a great position now to arm CSOs with the information they need for those debates to lead to constructive outcomes. But [these meetings] happen pretty rarely, every five to six years, which is frustrating to have to wait that long. 

The other area is the public consultations that communication regulators do. This is often a good opportunity to engage in discussion because, in some countries, every time a regulator wants to make a change to regulations, they are obliged to have a public consultation to seek input. Typically, they only get input from big telcos that have entire legal departments designed specifically for this purpose, but that is one thing we can change. We can track those announcements and prepare CSOs with the tools to engage in that discussion. I think most regulators would find it a breath of fresh air to hear from civil society in those spaces, because it tends to be a technocratic space that is dominated by the big operators. That's a big focus for us in the next couple of years: to think through how we can track those consultations and ensure CSOs are armed with the relevant information to engage and bring about these changes.

APCNews: In terms of that conversation, is it a matter of talking to regulators directly to intervene on behalf of smaller operators and make the playing field more equitable? And do we want big operators to be a part of that conversation? Are they receptive to these ideas or threatened by the idea of sharing the space? 

SS: There are two answers to that. One is that the existing network operators are typically happy with the status quo because the current system does not encourage competition; it limits opportunities to two, three or four operators in a given country, which mostly works out well for them, and they see any change as potentially threatening. But it is possible to have a host of new smaller operators that will only make things better for everyone. Not only will this increase access, but it will also increase the number of people dialling into big operators’ networks and using their infrastructure. So it grows the market, and part of our challenge is not only to convince the regulators that this change is a good idea, but also to convince big telcos that it will actually be good for them if we enable small-scale operators. 

CRM: I totally agree with mobile network operators liking the status quo. And, to a certain extent, why bother opening up the game? It is easier when you are three [companies] than 300. At the moment, I wouldn’t put much effort on the operators, to be honest. There are regulators in some countries that want to change things and, in certain cases, I think the problem is that there is not enough demand from civil society to change things, so [regulators] don’t see an alternative. At the same time, if we convince civil society to participate and take responsibility for the outcomes of this infrastructure deployment, then they will also be putting pressure on the regulators, and the regulators will be putting pressure on the operators. I don't know if we have the capacity to put any direct pressure on the operators. I think it needs to come from regulators and governments. One of the main drivers for them to act is public pressure, so I would do more work with civil society than I would do with operators. 

APCNews: In terms of regulation, one of the things we’ve seen recently is the creation of a community networks spectrum band in Mexico, which has been designated for use by small-scale and community network operators. Is this the kind of policy you’re advocating for in southern Africa and other regions, or are there are other measures you’d like to see implemented? How dependent are these policy solutions on specific context? 

SS: We’re not advocating for one thing. We’re advocating for a host of small changes that would enable small-scale and community network operators. So having spectrum set aside, such as in Mexico, is an amazing thing and we’re definitely lobbying hard for that in South Africa and in other countries, but there are many other changes that are identified in that paper that can enable small-scale and community network operators. It’s also about spectrum fees, access to backhaul links and TV white space spectrum. Every connectivity challenge is different and there’s no one technology that’s perfect for every scenario, so we’re saying that we need to recognise the value of small-scale and community network operators and create a suite of regulations designed specifically to enable them. 

Another point worth making is that there’s often a complex dialogue that goes on between the regulator and the government, and it’s different in every country because the relative independence of the regulator varies dramatically from country to country. In some countries, the regulator is effectively the same as government. In other countries, the regulator is quite independent. Sometimes, they’re completely in sync. Sometimes, their agendas diverge substantially. You need to understand the framework in the country that you’re operating in to know where your point of biggest leverage is going to be: is it with the regulator, is it with government, is it getting the two in the room together? They’re complementary, in terms of how you strategically engage with them. Governments are responsible for setting policy and direction, and the regulator is responsible for ensuring that the market heads in that direction.

APCNews: The Internet Governance Forum (IGF) was held 12-14 November in Paris, with a number of panels focusing on connecting the underserved through community networks. What do you think were the outcomes of this event in terms of raising awareness around spectrum issues and the potential of low-cost, community-owned connectivity projects? Where are you hoping these conversations will lead in the next year?

CRM: For me, I would say something as profound as the tide is changing. There has been a lot of talk in the other two Internet Governance Forums I’ve attended, a lot of energy build-up and a lot of people getting together that has led, among other things, to this particular project. But now, it seems to be real. [At the IGF], community networks were in the main session. There were regulators and big actors in the spectrum session. The African Union made commitments in public. At the regional African Internet Governance Forum, the African Network Information Centre made commitments in public to enable community networks. I think people are starting to take this seriously and the IGF was a reflection of that. We were not left on the sidelines, we were a main component of the event. There is still a hell of a lot of work to be done. This is only the beginning, but I think the tide is changing. We have some wind on our side that we didn’t have before. Before, we were against the wind. Now, the wind is changing.

SS: I’m a little bit more cautious than Carlos. I think that what is missing still is a critical mass of examples. We have some great examples of what can be done from Rhizomatica to Zenzeleni to Broadband for the Rural North (B4RN). These are amazing examples, but they are still small in number. And I think Carlos is right, we’re on the cusp of a change, of a significant uptick in the growth and sustainability of community networks, but we haven’t experienced that uptick yet. It’s coming. That will be the next big thing – to see an order of magnitude step change in the number of community networks around the world.

Coming next in the “What's new on the spectrum?” series: A conversation with Karla Velasco Ramos and Erick Huerta from REDES A.C. and APC member Rhizomatica.

Read the previous interviews in the series with Carlos Afonso of Nupef, Peter Bloom of Rhizomatica, and internet access specialist Mike Jensen.