Why Internet in the Philippines is Slow and Expensive

There was a time when having an unlimited dial-up account at home is already a luxury. Having access to the internet, even at maximum speeds of only 54Kbps, was the most liberating and addicting as you get instant access to vast information and online services. It was also very expensive — about Php5,500 a month, and unless your job requires and can justify for it, your only other viable option is a Php50-an-hour session at a nearby internet cafe.

All that has changed. We get access to the internet from almost anywhere — the traditional wired DSL at home, fixed wireless for last mile, and 3G or LTE for nomadic use. For really remote areas where no cellular coverage is possible, there’s the option to hook up to satellite internet.

Internet speeds have also gradually improved. From the very first 128Kbps DSL line from 10 years ago, we now can choose anywhere from a basic 1Mbps line to a more generous 10Mbps connection. If you want faster, there are options that go from 20Mbps to 200Mbps but it’s still limited to very few areas in Metro Manila.
Prices have also significantly gone down. The Php5k unlimited dial-up connection is now an equivalent of 8 or 10Mbps DSL or fiber connection.

Of course, as the connectivity service has progressed, the amount of information and content the consumers access have also greatly increased. Back then, there was no Facebook or YouTube. Online games are very limited and HD streaming was unheard of. Websites and portals were pretty basic and light.

All things considered — speeds are up, costs are down — the state of internet in the Philippines have significantly improved. If it didn’t, we won’t be labeled as the social media capital of the world with almost 92% Facebook penetration, one of the biggest YouTube consumers and producers in the region and become the selfie capital of the world.
Yet, there is this unanimous sentiment that internet access in the Philippines is the slowest and most expensive in the region, and perhaps even the rest of the world. Even independent studies have shown this many times before.

Everything is relative and the few who can afford a Php20,000 monthly service fee for a 100Mbps fiber connection at home would tell you otherwise.

But when we see hard numbers across Asia and globally comparing the average internet speed of the Philippines with other countries, we always lag behind in numbers.

We’re not that ambitious to compare ourselves to the blazing fast connections of Singapore, Hong Kong, Japan and Korea but at least our numbers should be within the range of contemporaries such as Malaysia, Vietnam and Indonesia.

Two sources frequently publish reports about internet speeds. One is Ookla and its very popular SpeedTest website and the other is Akamai that has analytics on everyone that accesses their Content Delivery Network (CDN) mirroring popular sites like Google, Facebook, Yahoo and YouTube.

If we are to interpret the results of their published report, we have to look at and consider two factors:

What is the subscribed internet speed of the users that are being measured?

Most ISPs have varying speed offerings but a significant majority of subscribers are on the lowest tier (1-3Mbps). Hence, this tier will determine the average speed of the whole country if we are to base that on Speedtest.net’s crowdsourced data.
In any case, the number is still reflective of most households in the country.

What is the average purchasing power of subscribers?

How much a typical household allocate for internet service will be reflective of their purchasing power rather than the ISPs capacity to deliver.

If ISPs offer a higher speed for the starting subscription plan, then the general average would also be higher. When providers scrap the 1 to 3Mbps starting plan and just start their services at 5Mbps, our general average would double. The result, however, is that fewer people or households will be able to subscribe to the service.

In other words, the current subscription plans that are being offered by the current set of players are a result of fierce sales and marketing push for bigger market share. There is more opportunity and revenue to be made by converting the remaining 60-70% of the population to subscribe rather thanthan up-selling and providing faster service to the existing subscribers.

However, once market saturation reaches 80-100%, the dynamics will change from subscriber acquisition to subscriber retention. That means service providers will focus more on selling higher speed plans to existing subscribers than selling starter plans to new subscribers.

Overselling, Underserving.

Let’s put this into a simple analogy. An ISP is selling tickets to a movie screening that runs 24 hours. Each ticket costs Php999 and allows you to come in and go out of the theater at any time. It even allows you to hand over the ticket to a friend so that when you are out, they can use the ticket and watch the movie while you are away.

The seating capacity of the theater is 1,000 seats. The ISP sells up to 2,500 tickets in the hope that not all 2,500 will enter and watch the movie at the same time. Besides, each one would probably only watch for 2 hours and not the entire 24 hours. Technically, every 2 hours they can accomodate another set of 1,000 new viewers which puts the ceiling capacity at 12,000 in a 24-hour window.

For a time this was okay — only around 800 to 1,200 people watch the round-the-clock movie at once. During those times where there are 1,200 people, the theater is over-crowded and it is standing-room (congested). It was still bearable as 1,000 gets to sit and only 200 are standing.

As years go by, the types of movies shown become more interesting and lasts longer than 2 hours (FB, YouTube, blogs, etc.). Of the 2,500 tickets sold, almost 2,000 people enter the theater and stay for 8 hours instead of the usual 2 hours. That’s 1,000 people seating and another 1,000 people standing. It’s too crowded and while people can still watch, it is not comfortable and others are blocking people’s view. Some would even stay thw entire 24 hours and hog the seats (ISPs refer to them as the 3% abusive users).

The ISP expands floor space or builds a second floor and adds 50% of its original seat count. There are now 1,500 seats. Theater owner says we now have bigger capacity, hence we can sell more tickets. The booths sell 1,000 more tickets. The seating capacity is increased to 1,500 seats but the tickets are now 3,500. The ISP waits a few years until it recovers the cost of adding the new 500 seats.

That is where we are at — a never ending cycle of adding more seats and selling more tickets.

This is exactly the same as in the airlines industry where carriers are actually allowed by law to overbook tickets by up to 10% (selling 99 tickets to a 90-seater plane). That’s why some people get bumped off of their flights once and a while.

In the ISP industry, there is no such thing as a limit on overbooking. The service providers are allowed to overbook in multiples. If an ISP has a capacity of 1Gbps (1,000Mbps) and it sells that capacity to subscribers at 2Mbps per plan they don’t just divide 1,000Mbps by 2, giving them a maximum of 500 customers paying Php999 each. That 1Gbps pipe might cost them $100,000 a month but they will only make Php499,500 from the 500 customers. A Php4.5 million pipe only making Php500k a month in subscriber fees — that’s not the way they will sell it.

What they will do is sell that 1,000Mbps pipe to as many as 20,000 customers which would net them around Php19,980,000. Of course, giving 20,000 customers 2Mbps each would actually require 40,000Mbps in total despite the fact that 2Mbps needs to be guaranteed. That is where the congestion starts. We go back to the movie theater analogy.

Just one of many reasons.

Now that we have a better picture of how the industry players behave, we can better understand the state of internet in the Philippines.

There are a lot of factors to take into consideration, some more obvious than the others.

1) Topography. The Philippine population is scattered in 7,107 islands so the type of connectivity solutions to cater to all of them will vary. Compared to Singapore or Hong Kong which is just a small land mass, the Philippines’s topography poses a more expensive roll-out of infrastructure. Rolling out an underground fiber line from Vigan to Naga requires huge investment. Add to that additional roll-out of submarine cables just to connect the numerous islands of Visayas (Negros Island, Panay Island, Cebu, Bohol, Samar and Leyte).

It is possible that any single internet provider will have to spend 2,000 – 3,000% more on infrastructure in the Philippines compared to the how much it costs to roll out in Singapore. Yet they both have almost the same subscriber base (Something like 2.5 million home broadband subscribers in the Philippines vs. 3 million home broadband users in Singapore. This does not include nomadic 3G/LTE users.)

This is one of the reasons why many barangays and islands in the Philippines only have weak signal or just GPRS/Edge connection.

No ISP is willing to invest and put up the 3 or more towers it needed to give the appropriate coverage. Hence, they will just put up one tower to serve the 100 to 500 or so potential subscribers in those areas. From a business stand-point, it makes sense — why invest millions on a small island of 1,000 people when only 500 of them will subscriber to you and pay you Php300 a month (ARPU or Average Revenue per User for prepaid SMS and voice).

2) Investment Cycle and ROI. The existing players that have already invested in service infrastructure in the last 5 to 10 years are still waiting for their return on investments from the last upgrade. Until then, they need to increase their market share and average revenue per user in order to hit their annual targets. Since most of the players are public companies, they have to balance their growth and expenditures against their revenue and performance in order to please the shareholders. In short, service providers need to recoup their investments and make a profit before spending some more to improve the service quality and coverage.

3) Lack of Competition. Did you know that in the early 2000, there were more than 300 independent ISPs in the Philippines? Though the market then was mainly just for dial-up, any service provider can easily hook up to the backbone and buy more dial-up numbers to increase their capacity. Hence, competition was fierce to gain and retain subscribers.

Today, with mostly DSL and fixed wireless services, the players have shrunk to just the bigger ones.

Believe it or not, we still have so many Internet Service Providers in the country. They range from cable internet, DSL, wireless (WiMax, LTE) and satellite (Read: All the 7 Philippine ISPs to Choose From) but the problem is they all have to deal with PLDT or Globe for the outgoing pipes. So yes, there is some sort of duopoly in play.

4) Peering and Local Exchange. There is some level of inefficiency when ISPs do not locally peer. Content has to go out of the country and back in order to reach a user instead of making the shortcut and travel locally. It also reduces lag time which is essential for certain services like online gaming and VOIP.

One ISP claims that they built their own backbone over the years and paid for it. It is an advantage that will give them a huge lead over competitors. If they interconnect, that would remove their advantage.

5) Appropriate Legislation. The current set of laws on competition, consumer protection and regulations are not enough or outdated. There’s the Competition Act of the Philippines that has been already enacted into law.

The NTC needs to work faster to resolve issues and penalties to erring ISPs must be updated to give it more teeth. Did you know that NTC is only allowed by law to penalize ISPs up to Php200 per day for failure to deliver service? It’s an old law that needs to be amended.

6) Government Support. The National Broadband Network which was proposed and scrapped many years ago would have been a first step had it not been riddled with controversy. A national budget to provide a bigger backbone or wider network would help alleviate the network coverage.

There is already a bill that is being prepared to come up with a Universal Charge which will collect from ISPs and the funds will be used to build more infrastructure. This is already being done in the power sector (check your Meralco bill for Universal Charge) and you can even see this in your water bills.

We’re seeing some progress but for most consumers, that progress is very slow especially when compared to our neighboring countries.

Just like electricity where our rates are among the highest in the region, internet connectivity is even a more complex problem.

For the mobile internet users, read more about “Why Your Telco Needs To Cap Your Mobile Internet“.