Photo by: Greg Demarque/Executive

During five days in April, Lebanese internet users were given proof that their notoriously terrible connections have been kept that way on purpose. Lebanon had become a laughing stock. Back when Ookla – the US-based operator of the ubiquitous speedtest.net – offered public rankings of average internet download speeds by country, Lebanon was always near the very bottom of the list. With the rankings no longer public, it is hard to reliably say where we stand today, but we do not really need an official ranking to know typical Lebanese download speeds of 1 to 2 megabits per second (mbps) are pathetic.

Last month, the publicly owned gatekeeper of Lebanon’s internet – Ogero – conducted five days of speed tests in different parts of the country. The results were jaw-dropping. The company’s new director general (appointed by the cabinet in January) claimed on Twitter that some users reached download speeds of 27 mbps, a 1,250 percent increase from the 2 mbps standard of the past few years. The majority of speedtest.net results tweeted at Ogero showed speeds around 16 mbps. And this without any intensive infrastructure projects. It was programmable.
“We’ve been saying for years that a simple administrative decision would improve the internet by at least 30 percent in terms of bandwidth, in terms of speed,” explains Maroun Chammas, CEO of IDM, a local internet service provider (ISP).

The bad old days

The Ministry of Telecommunications (MoT) is the sector’s steward. In addition to devising and implementing policy, it owns nearly all the infrastructure (fiber, copper and mobile-phone base stations). On the MoT’s recommendation, the government decrees the price of every phone call (mobile and landline), and the ministry’s hand is heavy on the internet. Competition in this segment is permitted, though in recent years it arguably has not been encouraged. In 2007, the steward oversaw a transformation in the market. At long last, the dial-up connection (complete with screeching modem) was replaced with a digital subscriber line (DSL) service. The future. Lebanon was a last adopter of this technology. Local rollout was slow, and download speeds are, a decade later, still abysmal.

To make a long and technical story short, the MoT and Ogero, a state-owned enterprise working as a ministry contractor, have considerable power over the internet market. With control of most of the infrastructure that allows someone in Lebanon to check their email, competition in the market can be easily hobbled. Private companies are allowed to use some of their own equipment to serve customers, but Ogero is in charge of the links that move customer traffic from one piece of equipment to another, before actually moving that traffic on to the information superhighway (via cables Ogero manages all access to). In other words, Ogero is the one closing lanes on access roads and opening only two tollbooths onto the highway during rush hour. Except it is at all hours. Every day. Whenever an unhappy customer – or a curious reporter – would ask a private provider why the internet was so bad, there was always one answer, however diplomatically delivered: Ogero, specifically in the person of its former director general. This has been true for 10 full years. It almost seemed like a convenient excuse (“Abdel Moneim Youssef ate my homework”) until the internet magically got better. Much better. Just like that.

The grey market

There used to be informal advertisements all around Beirut (which, when Executive decided to photograph one in April, were surprisingly very hard to find). An A4 piece of paper adorned with a few words in black ink, slapped up on a wall: “Wireless Internet Free for one month [local mobile number].” Some included monthly prices, some did not. This is the grey market. You pay a guy. He brings a wire to your house, sells you a router, and you have internet as fast as the two lowest-priced DSL packages, which an estimated 90 percent of legal users opt for, according to interviews with private players and an Ogero official that Executive conducted in 2014. Until recently, these illegal providers would get international bandwidth (actual internet access) in one of two ways: 1) by redistributing several legal connections to multiple users (small fry, and as best Executive can ascertain, the smaller part of the grey market segment) or 2) by bringing it in from abroad. As noted, Ogero controls the cables connecting Lebanon to the actual internet, but international capacity (as that on-ramp to the highway is called) can be secured via satellite or microwave. Many grey market providers were avoiding Ogero (and associated costs) all together.

Anywhere in the world, an ISP is that link to the highway. They buy access to the internet (bandwidth) and sell it to users at a higher price. The price at which Ogero buys bandwidth and from which provider has not been made public, but market estimates of around $4 per E1 line (2 mbps of bandwidth) seem reasonable based on internet research into the topic. ISPs in Lebanon currently pay Ogero $250 per E1 line (reduced from over $1,000 in 2014). Grey market providers buying bandwidth abroad no doubt pay more than Ogero (having at least a satellite link provider between themselves and an actual internet seller), but almost certainly less than local ISPs pay Ogero. And the advantage is not only on international bandwidth. ISPs must deposit  a $1,000 letter of guarantee on every E1 line they get from Ogero, according to both IDM’s Chammas and an ISP owner who spoke on condition of anonymity. Also, ISPs can only offer DSL on an Ogero cable. The grey market providers just add another wire to the tapestry slowly being weaved from building to building across Lebanon and hook users up the next day.

Between 2014 and 2015, the grey market got a bit whiter. Chammas as well as one private-sector and one publice-sector source told Executive that around 120 grey market ISPs were semi-legalized during that time period. Details are unclear, but what Executive can confirm is that companies formerly buying international bandwidth from abroad were sold bandwidth by Ogero – at a time when previously licensed ISPs were asking for bandwidth and not receiving it – but continued to be exempt from providing letters of guarantee and respecting distribution rules. Shortly after this apparent attempt at better regulating the market – which disrupted a supply chain at least 10 years old – corruption charges against then-Ogero Director General Abdel Moneim Youssef began to fly.

A new era?

Ogero’s director general is one of three directors general at the MoT. As such, the company’s head is an integral part of devising the ministry’s strategy for developing the country’s telecom sector. Ogero implements whatever strategy it helps write. The three-seat power structure gives the MoT’s Director General of Maintenance and Operation power to oversee Ogero’s work, and act as a check and balance, as Imad Kreidieh, the new head of Ogero, explains it. Until January, Abdel Moneim Youssef was both Ogero’s DG and the DG of maintenance and operation, “judge and jury,” Kriedieh says. That situation has now been resolved. Youssef is currently being scrutinized by the judiciary and is out of, well, two jobs. Kreidieh replaced him at Ogero, and Bassil Ayoubi is now in Youssef’s other former leadership spot at the MoT.

Market reaction to the change – only around 12 weeks old when Executive made the rounds – was cautiously optimistic. Communication between the private sector and both Ogero and the MoT is significantly improved, and Kreidieh speaks to Executive like a penitent. He admits that, prior to assuming his post in January, for Ogero, “the private sector used to be seen as the enemy.” He insists, however, “my role is not to consolidate a monopolistic position, but to offer the infrastructure for anyone who has a license, the technology, and the content to deliver it.”

Offering the infrastructure he reportedly is. On April 11, 13, 15, 27 and 29, Ogero quite simply unleashed the speed, to borrow the company’s hashtag. The tests lasted only a few hours, and were conducted in a handful of areas on each day. However, according to Habib Torbey and Patrick Farajian, heads of the data service providers Globalcom Data Service and Sodetel, respectively, the private sector has been able to keep the speed unleashed in the spots Ogero tested, as the lines between privately owned equipment are no longer congested.

Kriedieh has repeatedly said in public (both in March at ArabNet and during April on Twitter) that a new internet pricing decree will be presented by the MoT to the cabinet sometime soon, never committing to a precise deadline. While in the mobile phone segment, such a decree sets the price of all phone calls in the country (and out of it, which is why voice over IP services, like Skype, are still technically banned). In the internet segment, such a decree only sets internet package prices for Ogero. Private sector ISPs can charge customers whatever they choose (although most stick close to the decree to remain competitive), but the price of internet access (an E1 line) they pay Ogero is set by the decree. In the past, the price of internet packages for end users has been based on two factors: the speed of the connection and the monthly cap on download capacity. The decree will remove speed as a factor, Kriedieh explained, meaning users will be given the fastest speed their connections can deliver. Word on the street is that Ogero’s speed test days will become market realities after the decree. Executive interviewed Kreidieh before the tests, and he was unavailable for comment after.

Remaining constraints and the road ahead

This is not to say it is internet Christmas eve for every last one of us. Congestion is only one part of a series of problems. Individual internet users in Lebanon connect to central offices (COs), which then connect them to the internet, with copper cables. This is an old technology, with speed transfer limits and a serious problem transferring data quickly over a distance more than one kilometer. Fiber optic cables are now industry standard, and Lebanon has fiber in many places where it is needed (the country has 6,000 kilometers deployed, Kriedieh says). With a new government taking office in January came a new policy for the sector, Kreidieh explains. He describes the new strategy as “not politicized, user-centric and time-bound,” adding that “by the end of 2018, things will be much, much better.” Four projects are currently in motion, he says, the centerpiece of which is a long-discussed fiber-to-the-cabinet project (a cabinet being a piece of telecom equipment placed between users and a central office, not to be confused with the Council of Ministers). The project involves connecting individual users via copper wires to nearby cabinets –  which would themselves be connected by fiber to COs, meaning only a few cabinets and fiber cables would need to be installed to provide a multitude of users fast connections instead of fiber from the CO to each individual home. The project should begin in September and take 18 months to complete, Kriedieh says. He promises download speeds of “over 100 mbps without fiber” to the home (a technological breakthrough circa 2010).  [Editor’s note: Fiber to the cabinet was phase two of the now-fully-abandoned MoT national strategy launched in 2015. It was supposed to be nearly completed by now.]

The elephants in the room

Kreidieh claims not to see himself at the helm of a purely profit-driven commercial enterprise. At one point during the interview, he describes Ogero as a regulator. What he promises to deliver is a nationwide telecom network capable of providing all users with voice, data and streaming video services on the same cable. Triple-play, as it is called in the industry. Yes, a better network will benefit Ogero and help expand its market share (which he pegs at 290,000 of 700,000 legal subscribers, or 41 percent, admitting, however, that there are “leakages” that make the true number of subscribers unknown). Kreidieh, however, insists strengthening Ogero’s market position is not his goal. With a network capable of modern offerings, he says that he wants the private sector to flourish. The private sector, meanwhile, seems to want the access it has long been denied and a level playing field (meaning either fully licensing the 120 or so providers semi-legalized two years ago or forcing them out of business). Many estimate these semi-licensed providers have a significant customer base, although the numbers are very fuzzy.

Lebanon’s reported total number of fixed-line broadband (DSL) connections stands at slightly over 1.2 million, according to MoT data supplied to the International Telecommunications Union. If both that number and Ogero’s figure of 700,000 fully legal connections are correct, the formerly grey market would service over 500,000 customers. Ignoring the fact that 1.2 million fixed connections would mean Lebanon’s estimated 900,000 households are all wired, with some households actually having two or more DSL lines, it is safe to assume that there are at least some potential market share gains for fully-licensed ISPs if some providers leave the market when the rules apply equally to everyone.

Kreidieh offers no detail on how to deal with the 120 or so semi-legalized ISPs aside from suggesting that when the service offered by Ogero and fully licensed private ISPs is significantly improved the market will correct itself. If these providers are allowed to continue buying bandwidth from Ogero while being allowed to ignore the rest of the rules, however, it is hard to see how they will be pushed out of the market completely. Fully licensed ISPs may have more services to offer in the not-too-distant future (i.e., streaming, high-definition video and TV over the internet), but the market overwhelmingly demands a cheap connection, and is conditioned to accept vastly inferior services.