Inside Sri Lanka’s incompetent, inefficient, and increasingly overloaded transport systems.

I: Why should you care?

We all have problems with transport in Sri Lanka. The buses are packed, the trains are late, there’s traffic on the roads, and you just escaped a near-death experience involving a tuk and an SUV, these are practically national experiences. We might be divided by age, religion, ethnicity, and politics, but bitching about transport unites us.

In this analysis, we’re going to take a look at how transport works in Sri Lanka, how much it costs, who it serves, and where it can be improved. We’re going to start with the basics and go up from there, so that at every step you have the tools to understand the data. We’ve dug up a hell of a lot of research and also spoken to local experts to understand the context around why things are the way they are.

This is going to be a long journey. Annoying, perhaps, difficult, perhaps, but when things break you should know what to point to.

We’re also doing this because we’re tired of stupid conversations - endless half-baked anecdotes about how tuks are the Great Evil, or how the government can just snap its fingers and make railways better, and stuff of that nature.

We’ll do what we can to entertain, answer, and hopefully even inspire. Strap in, and let’s go.

II: Axioms

Let’s start right with the basics. There are three fundamentals here. Firstly, transport infrastructure - roads, rail and routes - is a critical part of the development of businesses, nations, and even the whole world.

A key part of civilization, after all, is being able to move people and goods around. Ancient empires thrived on transport hubs near roads and rivers. The Silk Road of ancient times connected China to India to the Middle East, and along these flowed everything from frankincense to literature. In the late 18th and early 19th centuries, the Industrial Revolution was linked to more sophisticated canal systems in Europe and North America. In the 19th century, it was railroads that drove the boom - bringing about more complex economies, more mobility, and thus more social opportunities.

Map by National Geographic:

In the 20th century, the game shifted to private transport - in the form of better roads for cars - and the rise of global air travel and cheaper shipping. It’s the reason we have mobile phones designed in California and chew toys made in China. Trade and transport logistics activities are mutually reinforcing entities.

Generally, transport infrastructure - roads, rail, the physical things [1] that make transport easier - affect the economy like so:

The links between transport investments as a means to economic growth. The graph was taken from the Geography of Transport Systems, by Jean-Paul Rodrigue, and redrawn for ease of reading.

Now, we can all agree that better transport is always a nicer thing to have. And by better, we typically mean more comfort, more reliability (we want to get from point A to B on time!), more options, and more affordability.

This brings us rather neatly to the second point we want to make: better transport infrastructure = better transport = a better economy = better access to goods and services = more convenient lives for us all.

However, there’s a third point to think of: if good transport helps, bad transport hurts. Better transport isn’t just something we all want; it might be something we all need.

We’ve all felt the effects of traffic and shitty buses and trains being late, but it goes much deeper than that.

In the ‘90s and 2000’s, researchers and policymakers in the UK started paying attention to the links between transport disadvantages and social exclusion - poverty, poor access to services, and so on. Thus, was born the UK Social Exclusion Unit (SEU) and its hugely influential 2003 report on the subject.

They systematically mapped out something we all feel instinctively: the lack of access to transport can have major ripple effects on your social and commercial success.

We like to think that skilled people will be recognized no matter where they are. This is a fallacy. It turns out that for most people, not being able to move around severely hurts their chances at almost everything.

Diagram to illustrate the relationship between transport disadvantage, social disadvantage and social exclusion. The graph was taken from the paper Transport and social exclusion: Where are we now?, by Karen Lucas, and redrawn for ease of reading.

Having poor transport doesn’t just mean you have to think twice before meeting your friends at Randoli. It means you have less access to people and networks, jobs, money, services. It means you have fewer chances to build skills, find a better job, and make your voice heard. These things feed right back into themselves, creating a vicious cycle. Already poor and socially disadvantaged people end up worse.

These effects also show up in the kind of top-level economic analysis that’s typically done. In the Geography of Transport Systems, Jean-Paul Rodrigue posits that at an aggregate level, efficient transport systems reduce costs in other economic sectors, while inefficient systems increase these costs. However you slice it, good transport is essential.

III: Modes of Sri Lankan transport

Now let’s see if we can figure out how Sri Lankan transport works. We need to look at a few things:

  1. The mode of transport (essentially, a choice in types of transport: buses, cars, tuks, walking, etc.)
  2. The infrastructure and its geography
  3. The costs of maintenance, which is where all of this links to the economy.

That being said, there are a few quick notes we want you to keep in the back of your mind. These points are essential to understanding the data you’re going to see:

  1. Most people in Sri Lanka need some form of motorized transport to get to work. A 2015 survey showed that only 10.9% of the population can get by without.
  2. Public transport accounted for 60% of all transport in Sri Lanka, but evidence shows an increasing shift to private transport. So, in order to present a fair summary of the whole system, we have to look at both public and private aspects.

To understand choice, we need to map out what exists and how and why people choose them. There are two ways of looking at the problem: we can look at the vehicle populations, which are fairly accurate - or we can use passenger kilometers, which is an estimate of the number of kilometers travelled in each type of vehicle class. For obvious reasons, this latter type of data is a bit more inaccurate because of how difficult it is to actually estimate.

We’ll be using data from the National Transport Commission, cross-referenced against Central Bank data to double-check numbers and categories [2].

Vehicle populations

As of May 2022, we had an estimated total vehicle population of 8,341,861. Of these, 7,479,265 are vehicles used for transport. If we look at the populations on the road over the years, this is what we get:

In 2015, it was determined that the country had a fleet of “2.7 million operating a vehicle ownership rate (VOR) of 129 vehicles per 1000 people”. Or, in simple terms, 1 vehicle for every 7.75 people.

In 2016, it was reported that Sri Lanka imports 1,800 vehicles on a daily basis. During the first nine months of 2015, a total of 489,000 vehicles had been imported into the country (although possibly not unleashed upon the roads instantly, so they weren’t ‘on the road’ yet). The same goes for motorcycles. The line just keeps going up.

New vehicle registrations

And then there’s this chart of passenger kilometers - which is based on data from the National Transport Commission; this is where they try to estimate the actual number of kilometers spend in different modes of transport.

Passenger kilometers

These three charts tell us an interesting story. The first is that there’s been an explosion in the number of people in cars - almost double since 2014. Ditto for three wheelers. The second is that despite us having more buses in stock, both CTB and private buses now move less people around. The third that we have a lot more motorcycles being bought, but the actual amount travelled on those motorcycles might be quite a lot lower than on cars or three wheelers.

The bus story deserves a closer looking at. Because it’s easy to say that COVID might have crippled bus transport - but this shift towards private transport seems to be part of a more long-term phenomenon. Our current transport modality (that is, general choice of mode of transport), in terms of actual passenger-kilometers travelled, at least as of 2020, looks like this.

As you can see here, the humble tuk-tuk, has not only increased in number, but those numbers are carrying a lot more people around - it’s become a critical middle tier of our transport system, between a bus and a car.

Our math for 2022 (total human population in Sri Lanka as of now - 21,623,536) suggests the numbers are now at 1 vehicle for every 2.5 people, or approximately 1 vehicle every 3 people if we’re looking at purely transport vehicles [3]. Theoretically, we should have enough transport for everyone.

But we don’t, because these changes are not equally distributed. Most of this change is in metropolitan areas, according to the research we cited earlier. The Western Province, in particular, has the highest rate of vehicle ownership in the country.

How do people choose?

Unfortunately, there wasn’t a lot of research in Sri Lanka on why people chose particular types of transport. We think it’s anecdotally accepted that you take what you can afford, but it’s always best to find more data. And so, we turned to a 2015 paper that looked at 422 people from across Colombo and tried to rank the most common factors behind their transport choices.

  • Income, occupation, cost, and vehicle ownership. As expected, income affects how people chose transport. Duh.
  • Gender, age, education
  • Reliability, safety, comfort, and distance
  • Weather, and also whether they travel alone or with others

Now, let’s get a little more specific.

Motorbikes - Motorbikes riders tend to be young, working males with a middle to low level income. They’re seen as ideal means of working around traffic congestions.

Three wheelers - a wide range of people use three wheelers. Students, private sector employees, the elderly, people dropping children off at school, people going on shopping trips or recreational activities . . . the list goes on. Three wheelers are quite popular among those who rent housing.

Public transport - More than half the trips between home and school or home and work happen on public transport. It’s considered a cost-effective system while however, the trains are seen as less reliable in terms of time.

Cars - Privately owned cars tend to be largely preferred by men with high income. There’s a huge correlation with education, which makes sense (education and income are typically tightly correlated). Comfort and time efficiency tend to be the biggest criteria in this selection.

Walking - there’s a large, gendered aspect to walking: unemployed women, who register this as one of their most important ways of getting around. Most people who choose to walk do so in order to fulfil everyday activities including shopping, dropping and picking up children from school and classes - again, these tend to be gendered.

This paper isn’t perfect [4]. Our other options were papers on the transport choices of schoolchildren in Colombo and Kandy - not too useful, since schoolchildren don’t get much choice in the transport they take.

However, this does give us a rough idea of choices, as well as the factors that shape them. It follows that if a country is optimizing for cars, they’re implicitly optimizing for wealthy men; if they’re optimizing for walking, they’re aiding more unemployed women; if they’re improving public transport, they’re actually improving that home-office or home-school commute.

IV: The shape of Sri Lankan transport

Now that we have some idea of what vehicles and their population looks like, we need to look at what they travel on.

The majority of Sri Lankan transport happens on roads, specifically 12,537.247 km of them. The beating heart of this road network is essentially in Colombo. We have a rail network - largely a rust bucket leftover of British colonial rule - that moves people and goods around, but it’s not the biggest fraction of our transport.

Since roads constitute the greater fraction of this country’s transportation network, let’s dissect that puppy! [5]

Sri Lanka broadly has six ‘grades’ of roads

  1. Expressway (E)
  2. Main Road (A)
  3. Main Road (B)
  4. Secondary/Minor Road
  5. Jeep/Car Track
  6. Footpath

Of these, grades E, A, and B are what the Road Development Authority consider to be ‘National Highways’ under their purview. E (312.586 km) grade roads are highways; A grade roads (4,217.420 km) connect provinces together; B grade roads (8007.241 km) connect towns, urban areas, and so on.

All told, that’s about 12,537.247 km of national roads in the country, with highways making up the smallest percentage. To visualize this, we went and pulled the data from OpenStreetMap, which is used by everyone from Microsoft to Amazon Logistics. Here’s what our national road system looks like:

Secondary roads, tracks, and footpaths come under provincial, municipal, and local authorities.

Okay, it doesn’t make much sense without a comparison. Let’s think of it this way: size matters ahem.

Sri Lanka has an area of 65,610 km². There are a few other countries that match us in size: Latvia (64,559 km²), Lithuania (65,300 km²), and Georgia (69,700 km²).

We have 12,537.247 km of national roads. Latvia has 20,066 km of roads; Lithuania says they have 21,249 km of national roads; Georgia has 21,100 or so. It’s not exactly an apples-to-apples comparison, but in purely km terms, we’re a bit under [6].

To add a little more context to this work, we went and interviewed Professor Amal Kumarage, Head of the Department of Transport and Logistics at the University of Moratuwa. Kumarage’s work is oft-cited in the field, and he’s notably been part of many large transport projects over the years.

He pointed out something curious: Sri Lanka has rookie numbers when it comes to national roads, but we have one of the highest densities of rural roads compared to countries of similar size. That’s those secondary / minor roads and paths.

Now, this whole road network is, increasingly, being used by private transport.

A lot of this is focused in the Western Province. Access to Colombo, as the economic hub of the country, is seen as so important that much of the planning done so far (which we’ll get into) is based on Colombo. In 2016, it was determined that an average of 100,000 vehicles enter Colombo each day.

There are seven major transport corridors that see huge volumes of commuter traffic:

  1. The Negombo corridor (Negombo to Ja Ela to Kelaniya to Colombo)
  2. The Kandy corridor (Kandy to Kadawatha to Colombo)
  3. The Low Level Road Corridor (Kaduwela to Colombo)
  4. The Malabe Corridor (Malabe to Battaramulla to Colombo)
  5. The High Level Road Corridor (Kottawa to Maharagama to Nugegoda to Colombo)
  6. The Horana Corridor (Horana to Piliyandala to Colombo)
  7. The Galle Corridor (Galle to Moratuwa to Dehiwala to Colombo)

Think of Colombo as a heart and these as the veins and arteries leading in and out of it.

Now let’s look at railways. Our rail network is 1,508 km. Here we’re much more competitive. Latvia’s state-owned railway company runs 2,364 km; Lithuania has 1,868.8 km; Georgia has 1,323.9 km. You wouldn’t know it, because the map on the Railway Department website looks like someone scribbled over it with Microsoft Paint, but the old network is pretty decent in length.

However, we need to remember that this system is pretty old. When we say we have a rail system, we have a truly ancient one - most of the infrastructure is literally from British times. Secondly, it’s pretty lopsided. The Southeast side of the country is practically untouched, and very few lines emerge from the main lines - in short, unless you’re living close to a rail line already, the actual access is pretty poor. You’ll notice that the rail network also closely follows the same Colombo-centric corridor system. We’ll get into this later.

It also tells us a lot that the authorities don’t even count pavements, so many of the walking population aren’t even recognized as data. This is stupid, for obvious reasons. It’s an old adage that what gets measured gets improved on, so the next time you wonder why your stretch of road doesn’t have a viable pavement, understand that it isn’t even measured.

The overall takeaway is:

a) We have a lot of vehicles on the road, and the distribution is skewing even more towards private transport

b) Three-wheelers are an increasingly important middle ground transport system

c) We have no optimization for walking because we don’t even seem to count pavements

d) Our transport system is very Colombo-centric, but we probably need more roads; not just highways, but B-grade roads that serve the everyday commutes of more people

b) We have an old railway system with a fair length but relatively little coverage

V: Transport and economy

We looked at the Central Bank’s annual report data to try and understand how much of private expenditure goes towards ferrying us from point A to point B.

Composition of Private Consumption Expenditure - 2021

Private spending categorized by the type of expenditure. Source : CBSL Annual Report, 2021, Chapter 2

As you can see, we spend about 20% of our money as private citizens on transport, on average. This makes it the second largest chunk of private expenditure. Central Bank data shows that this has been a near constant for the past decade.

Consider Germany, a country known for its transport and logistics infrastructure. In the home of Mercedes-Benz, Porsche, and BMW, transport only accounts for 14% of household expenditure on average. Germany ranks first in the World Bank's Logistics Performance Index (LPI). Sri Lanka ranks 94th.

One of the easiest ways to explain this is to think about the costs of private transport (as you saw above). When we compute averages like these, the costs of expensive cars and cheap bus fares all get rolled into one figure; so we spend more than the Germans because we’re more reliant on private transport. Another factor could be that we earn a lot less, so the impact of transport cost on the wallet is much higher.

What does transport do for our GDP?

We began by pointing out that transport has a lot to do with an economy. We did this because there’s now a lot more awareness about Sri Lanka’s economy and how it functions. [7] Now that we understand a bit about how Sri Lankan transport works, let’s look at its impact on our economy. [8]

This data is taken from the Department of Census and Statistics and shown as a line graph for ease of understanding.

As you can see, transport is consistently over 10% of our GDP. Compare this to sectors you’ve probably heard a lot about - the garments industry and hotels and restaurants. Transport plays a far bigger picture, and this GDP contribution metric doesn’t even take into account how much of every other sector is only possible with transport being in place.

Don’t just take our word for it. In the Geography of Transport Systems, Jean-Paul Rodrigue and Theo Notteboom - both economists who delve into transport geography - pointed out that efficient transport systems reduce costs in other economic sectors, while inefficient systems increase these costs.

Rodrigue and Notteboom also note that in most developed countries, transport accounts for between 6 and 12 percent of GDP. Logistics costs account for a further 6 to 25 percent: meaning that the value of all transportation assets, including infrastructure and vehicles, can easily account for almost half the GDP of an advanced economy. They calculate that transport also accounts for around 4% of the costs of each unit of output in manufacturing in advanced economies, with the caveat that this figure varies greatly according to sub-sectors.

What this means is that if we put more money into transport, it would also boost other sectors, including forex-earning sectors like the ones we’ve shown above.

This tracks with a lot of what we’ve seen in other countries. For example, a study of Indian road and rail investments between 1970 and 2010 showed that road and rail correlate with capital and economic growth - or, as they put it, “bidirectional causality between road transportation and capital formation, bidirectional causality between gross domestic capital formation and economic growth, unidirectional causality from rail transportation to economic growth and unidirectional causality from rail transportation to gross capital formation.” In other words, road and rail transport made people richer, which contributed to overall economic growth.

In a study looking at China between 2000 and 2015, they found a similar phenomenon with freight transport. As of 2021, Chinese investment in inland transport infrastructure alone accounted for 5.5 percent of GDP; most of that is road-building, but they’ve also made significant investment over the decades in rail networks, which together have contributed to China’s average annual growth rate of 9.05 percent between 1989 and 2022.

V: How much does this all cost?

This data is taken from the National Transport Commission and shown as a line graph for ease of understanding.

There’s a whole lot we can say about costs - we went digging through National Transport Commission data; here are the tables of expenditure for buses and trains; we’ve got everything from maintenance costs to how much the SLTB spends on water and postage.

But the diagram above should tell us a lot. At a high level, SLTB was profitable for a while between 2016 and 2020. Railway revenue has been consistently lower than railway expenditure.

There are plenty of arguments for why this lossmaking is bad - everyone from Pavithra Wanniarachchi to Advocata has said it. Pro-market economists are one thing, but when the lady who thinks sacrificing herself to the sea might have stopped COVID says it, you know the writing is very much on the wall.

There are also plenty of arguments for why a lossmaking railway is fine. As we’ve seen over and over again, a public transport system isn’t an isolated corporation. It boosts everything.

Estimating costs of private transport is much more difficult. What’s the total cost of all the Altos and the Honda Vezels running around?

In our conversation with Professor Kumarage, we touched on the fuel crisis. He told us this:

Around 63% of petroleum imports amounting to US$ 2,065 million has been used for road transport annually (Sri Lanka Sustainable Energy Authority, 2019). This was when oil was US$ 50-60 per barrel. At current prices of over US$ 100 per barrel, this will increase to 20% of our usual import bill of US$ 20 billion per year. It is possible that oil prices will stay at this level for several years, as in 2008 when it crossed US$ 100 mark, it held for four years.

. . . Private vehicles consume around 80% of this fuel (Weerawardena, 2019), leaving the entire goods and public transport services consuming just 20%.”

That’s insane. And that’s maintenance cost. But we still need to consider the base cost of vehicles. Instead of trying to break our brains trying to calculate it, we decided to build an app.

This is the Watchdog Private Transport Ownership Tool.

Select your province, select your dream vehicle (we have a few) and enter your salary. It’ll tell you how many years it’ll take for you to afford said vehicle, depending on how much of your income you want to throw at it.

Access it here:

This way, you can see the costs of private transport for yourself.

We’ve also incorporated data from the Household Income and Expenditure Survey to show you how you and your dream vehicle stack up against the average household income for your province. Next time you see a vehicle you like on the road; you’ll know how long it’ll take you - and your fellow citizens - to get to that level [9].

VI: What’s wrong with this system?

In our subjectively objective opinion, Sri Lankan transport is garbage. Granted, it’s not as terrible as, say, Nepal, where two of the authors of this piece witnessed multiple deaths travelling from one city to another. But everything about this system is stupid. The buses are packed and uncomfortable, the trains never arrive on time, the roads are choked waaaay beyond capacity. This system is such a meme that we recall an apartment in Borella running a giant advertisement saying IF YOU LIVED HERE, YOU’D BE HOME ALREADY.

But why? We can ask naive questions - for example, we pointed out earlier that Sri Lanka has approximately 1 vehicle for every 3 people if we’re looking at purely transport vehicles; theoretically, we should have enough transport for everyone. So why don’t we?

Well, that’s easy to argue away. As science fiction writer William Gibson pointed out, the future is already here, it’s just not evenly distributed.

In the process of compiling this gigachad report, we began to dig at the forces that keep things broken. This is where Professor Amal Kumarage’s paper, “Why transport in Sri Lanka continues to go from bad to worse” comes in.

Kumarage lists twelve factors that contribute to the madness, among them:

  1. A focus on rural road building at the expense of funds for putting roads where people are
  2. Middle income groups shifting to private vehicles over public transport
  3. A lack of structural change in the institutions responsible
  4. Trade unions dominating these institutions
  5. Three wheelers
  6. Lack of road safety
  7. A lack of investment in railways and other public transport

We’re not entirely sure whether we agree with Kumarage on all these things. As we highlighted in the data, we think three wheelers serve a vital role in the transport ecosystem. However, we agree with most of them, and we have some stuff to toss onto the table ourselves:

  1. A complete lack of thinking around walking and unpaid work We pointed out earlier how research shows that walking is a preferred mode of transport for unemployed women. Let’s unpack that statement: “unemployed” should probably read “unpaid work” here. We’re seeing the same patterns that have been noted in Spain, for example: transport networks are built on a linear model that essentially calculates for a home-office commute and doesn’t take into account access to neighbouring areas with day care centres, schools, and supermarkets. When we looked at research on links between transportation and quality of life, we found that having an environment that allows walking has benefits for the quality of life for older people. And remember, everyone ages; everybody dies.

  2. Almost total disregard towards women-based transport and safety Women are almost never accounted for in transport decision making. The harrassment and abuse women face in daily transport commutes is like a well-known secret. Well talked about, and nothing done. It is important that we take into account women’s travel practices, safety, and comfort along with any other distinctive characteristics of women’s mobilty needs. Why? because data from the Department of Census and Statistics shows that women are little over 50% of the population; 34.3% of women are part of the Sri Lankan labour force. Safe transport for women is a huge concern; we’re essentially ignoring the needs of 50% of the population.

  3. Congestion. In 2015, the University of Moratuwa’s Transport and Logistics Department predicted that average speeds in the Colombo metropolitan area could drop to as low as 6 kilometres per hour by 2020. Of course, COVID lockdowns may have skewed this number, but it’s a simple equation. We’re essentially getting to where Singapore was in the 1970s.

  4. A general lack of intelligent design in how public transport (particularly buses) are arranged. We went and scraped hundreds of bus routes to understand:

  5. Inter Provincial Bus Routes – Normal Buses

  6. Inter Provincial Bus Routes - Semiluxury Buses

  7. Inter Provincial Bus Routes - Luxury Buses

  8. Inter Provincial Bus Routes - Super Luxury And even 5) Western Province Bus Routes

To attempt to visualize these routes is to venture into an abyss from which there is no return. No, seriously. Look at the data directly. It’s like a giant spider took a metric ton of drugs and wove the most drunken spiderweb known to man, god, or beast. If it was any crazier it would have qualified as modern art.

How did this all come to be? This is where we have to understand a bit of history. Kumarage breaks down transport development in post-independence Sri Lanka into three periods of big change in transport:

  1. The Interventionist period (1948-1977) Immediately after independence, priority was given to rural transport over developing connectivity between ports. There was a plan to build a network of expressways in the 1960’s; this was overruled in favour of increased spending on rural roads and more rural bus services. Rapid expansion of rural roads and bus services led to a 80,000 km network of rural roads that became expensive to maintain. Kumarage notes that the main road network did not improve, and very few new major roads were built during this period. The Colombo tram network, which had been handed over to the Municipal Council in 1929, was decommissioned in 1960.
  2. Post-liberalisation period (1977-1990) Opening up the economy in 1977 led to an increase in import and export activity. Increased trade meant the Colombo Port was equipped with roads to handle container trucks. However, trunk road connectivity remained poor, and development, was limited to areas close to Colombo. Public transport infrastructure investment was on the back burner and as a result, the open economy saw surging demand for private vehicles which put pressure on roadbuilding, especially in urban areas. Private investment was allowed for bus transport and this sector gradually became the largest public transport supplier. However, the industry was poorly structured. Regulation made it so that buses could only be owned by individuals, and not by companies; Kumarage argues this regulation brought forth a legion of individuals that were impossible to regulate. This, he says, made the sector resistant to change and improvement.
  3. The war and post-war years (1990-present) - With internal conflicts at their height, moderate economic growth saw congestion in and around Colombo, while the rest of Sri Lanka were largely ignored. Immediately after the end of the war in 2009, private vehicle ownership began to grow exponentially. Congestion also increased in this time and spread to urban centres including Kandy, Kurunegala and the suburbs of Colombo. Still, public transport was not prioritised as a viable alternative to private means of transport. Road building took precedence as different forms of intermediate transport such as office vans, school vans, and lately, three-wheelers grew rapidly. Kumarage also claims that the high taxes on private vehicles saw a proliferation of motorcycles and cheap low-quality motor vehicles. This has caused increasing pollution, road accidents, and fatalities. We’ve written before of how the humble three wheeler has become the go to transport option for many people for a variety of considerations including versatility and cost.

We have to note here that there was a point in Sri Lanka’s history where there was serious attention on improving bus services in Sri Lanka. For example, the Wikipedia article lists uncited anecdotes like this:

”In 1970, the UF United Front government of Sirimavo Bandaranaike was returned to power, with Leslie Goonewardene of the LSSP as Minister of Transport. The CTB workers immediately took matters into their own hands and reconstituted the workers' councils. On their demand, Anil Moonesinghe (now no longer an MP) was brought back as Chairman and General Manager.

Anil Moonesinghe would dress in a bush shirt and trousers and operate as a one-man flying squad to catch errant bus crews in the act, lying in wait in his metallic blue Volkswagen Variant or his Citroën 2CV at places like Dematagoda Junction to prowl on them. Stories would abound (all untrue) of him being spotted disguised with a beard. In order to prove that eight buses could be serviced in a day, he once carried out eight vehicle services at the Central Workshops, Werahera.”

We don’t know where this anecdote is from, but it’s clear that public transport has fallen a long way. In fact, there’s so much fascinating history to dig into that we ended up writing a whole accompanying article about it.

Kumarage believes that one of the main causes for the fall is political considerations. He cites a desire to keep public transport fares low and to continue to provide for the ever-expanding network of rural roads; the latter functions as a handy political tool for provincial politics. He blames the dominance of trade unions in public transport agencies for keeping these institutions (Sri Lanka Railways, Sri Lanka Transport Board) from evolving to better serve the needs of the public.

Ultimately, here’s where we are right now:

Courtesy: Kirk Anderson, Pioneer Press

VI: Who’s responsible for all this, and what are they doing now?

You’d think that there’s one body or authority that you can write to with your suggestions. Not really. The Sri Lankan transport authorities are a strangely byzantine mess.

This is an org chart. To assemble this, we had to go through many, many websites, Acts of Incorporation, and figure out what parts of the machinery actually come under what [10]. Sometimes these bodies don’t even seem to acknowledge each other. And often, depending on the context, various bodies have increasing or diminishing powers.

For example, let’s talk roads. The RDA is responsible for E, A, and B roads, as we’ve said before. Smaller roads (C and D class) fall under Provincial Road Development Authorities and/or the relevant Provincial Council; even smaller roads fall under even smaller authorities.

Let’s then talk buses. NTC provides advice to the government, handles some permit work, and seems to ‘ensure service’; SLTB provides bus services, shelters, stations, and terminals; RDA provides bus stops; but so do local municipal authorities; operator federations bring issues to the machinery; route associations provide route oversight. It’s a deeply fragmented and convoluted system with (sometimes) overlapping responsibilities and jurisdictions.

And here’s an even bigger one: Wikipedia says “Ministry of Transport and Highways”. And that’s also how it’s been run for some time now. Yes, but in 2022, Highways and Transport functioned as separate Ministries.

The takeaway here is that for any given problem, the Ministry of Transport / and / or / Highways is theoretically responsible - but the actual implementation and oversight is handled by someone else down the line, and the top is reorganized so often it’s enough to make our brains melt. This obviously creates issues - a lack of coherency, sometimes a lack of direct responsibility, and the poor organization and implementation shows itself in transport poverty across the country.

There’s been a worrying trend over the past decade or so: lots of money into highways, very little into public transport.

Now, budget data was difficult to come by - successive administrations tend to divvy up ministerial portfolios in ingenious ways. For example, at some point we had a Ministry of Highways and Higher Education, a combination which makes absolutely no sense except to fudge the numbers. However, here’s some food for thought.

For example, we pointed out that in 2022, Highways and Transport were listed as separate Ministries. The capital expenditure estimate for the Transport Ministry was 16.5 billion rupees. Among the “scientifically” allocated portfolios was also a State Ministry of Vehicle Regulation, Bus Transport, Train Compartments, and Motor Car Industry. The capital expenditure allocation for this Ministry was 3 billion rupees which, going by the 2023 allocations, is just about enough to procure 600 buses. The capital expenditure estimate for the then Ministry of Highways was 270 billion rupees. That’s a mind-boggling sum of money and absurd division of labor.

In his budget speech on 14 November 2022, President and Finance Minister, Ranil Wickremesinghe, announced a program for logistics development.

Performance of the logistics infrastructure is another barrier for exporters, importers, and investors,” Wickremesinghe said, announcing that a three-year plan would be introduced to improve Sri Lanka’s ranking to be among the top 70 countries in logistics performance. We assume he was referring to the World Bank’s Logistics Performance Index, where, as we mentioned earlier, Sri Lanka is sitting uncomfortably at 94th.

Take this with a grain of salt. Over the years, governments have presented grand plans for public transportation but very little has come of these plans. We’ll be looking at these master plans and strategies in detail, but for now, the best way to look at actual commitment is look at the annual budget and the funds allocated.

Let’s look at the allocations for 2023 from the budget estimates:

On the whole, it looks like things are going the right way. We pointed out earlier that our roads are rookie numbers, and money is being yeeted forthwith. In the case of the Kandy multimodal transport terminal, the World Bank (WB) is providing a part of the funding. When it comes to the procurement of buses, the plan appears to be to use the balance of a 318 million dollar credit line from India.

But it’s not all about the roads when it comes to good and efficient transportation. The only major capital allocation for public transportation is the construction of the multi-modal hub in Kandy (4 billion rupees), and the procurement of 600 buses for SLTB (3.2 billion rupees).

These numbers indicate that public transport in the form of good bus and railway services are way down the list of priorities for the government. We go back to what we pointed out earlier: a government that optimizes for cars is optimizing mostly for rich men.

Now, you may think that the Expressway is a necessity. You might be on the side of the political spectrum that thinks 10 billion rupees on season ticket subsidies is wasteful expenditure.

But, let’s put this in context.

We’ve spent about 113 million US dollars - that’s over 40 billion rupees at current exchange rates - on the construction of the Lotus Tower alone. Colombo’s shining appendage reaching for the skies cost over 75% of the 2023 allocation for Railways. That’s over four times the subsidy for season tickets. It’s the equivalent of 10 multimodal transport terminals.

The future that could have been

You might be wondering why our current transport system isn’t keeping up with those of our peers and contemporaries across the world. The long and short of it is that we’ve had plans to develop the system beyond what it is right now.

It’s not like we haven’t had any plans to fix these issues. In fact, there have been two main plans in recent times:

  1. Urban Transport System Development Project for Colombo Metropolitan Region and Suburbs — which we’ll refer to as the CoMTrans Masterplan
  2. Western Region Megapolis Transport Master Plan — which we’ll refer to as the Megapolis Master Plan

You can read these plans using the links above.

Before we summarise the main points of these plans, we’d like you to keep a couple of things in mind. The complex nature of large scale infrastructure projects makes them very unpredictable. As a result, a lot of changes can happen during the lifespan of a project. These changes can include the scope, costs, timelines, technologies, and geography.

And also, the nature of planning and projects are so bizarre and convoluted at times that this saga looks a lot like this. It could have been a nice meal, but it’s messy, it’s on the floor, and there are probably flies crawling all over it.

With that in mind, let’s dig in and delve into the madness. Consider yourself warned.

CoMTrans Masterplan

The CoMTrans Masterplan was released in 2014 and was presented by Dhammika Perera. Yes, that Dhammika Perera. It was put together in partnership with the Japanese International Cooperation Agency (JICA).

There are three main objectives you should care about.

  • Reducing traffic by introducing Monorail, Bus Rapid Transit, and Bus Priority lines.
  • Making it easier for you to jump between different modes of transport by building Multi Modal Centres and hubs. That way you can drive your car to a hub and hop on a train or a bus from there to work.
  • Improving reliability and comfort of existing railway lines by electrifying and modernising existing rail infrastructure.

When it’s all put together, our transport network would have looked like this.

Public transport network as presented by the CoMTrans Study Team

Let’s unpack this. Let’s start by looking at Rapid Transit infrastructure projects first.

Monorails are the futuristic looking trains that hover over cities like in the picture below. They operate on a single rail as opposed to regular train which operates on two or more rails. Because the train wraps around the rail, these lines have to be elevated and can’t run on the ground.

Image Source:

We were planning to build six monorail lines spanning 58 km connecting areas like Malabe, Homagama, Keleniya, and Kesbawa to Colombo. The first two routes connecting Malabe-Fort-Kotahena and Kollupitiya-Town Hall were expected to be completed by 2020. The remaining lines were to be completed by 2035.

Bus Rapid Transit puts buses on their own dedicated lanes. BRT is different from Bus Priority Lanes because no other vehicle would be able to use these lanes. In the end, this would mean that  buses wouldn’t get stuck in traffic. As a result, travel times become more predictable.

These lines can sometimes be elevated or run on roads that are completely separate from the main roads. Here’s a visual that should clear things up a bit.

Image Source: buses are typically more comfortable than your standard Leyland bus. In some countries, these buses can have extra carriages.

As for Bus Priority Lines, these run parallel to the main road, but they get priority at intersections. Think of them as bus-train hybrids. Take a look at this example from Guatemala City.

30 meter long Transmetro in Guatemala City for 300 passengers

We had plans to build four BRT routes by 2025.

  • Route 1: Fort - Moratuwa
  • Route 2: Kelaniya - Fort and Kelaniya - Kadawatha
  • Route 3: Kelaniya - Moratuwa
  • Route 4: Wattala - Battaramulla - Moratuwa

This network would have spanned 135.8KM and was projected to cost us a total US$ 165 million (2013).

Now that we have multiple transit lines that people can use, we have to make it convenient for people to get in and out of these vehicles.

To encourage more people to use public transport, we also need to make sure people can drive to their nearest transit stop, park their vehicles, and jump on the train or a bus to get to work. This is where the development of MultiModal Centre and Hubs comes into the picture. The largest of these was planned to be built in Fort. It was referred to as the Multi-modal Transport Hub (MmTH).

In addition to the hub in Fort, three other Multi-Modal Centers (MMC) were planned for Malabe, Kelaniya, and Moratuwa. These hubs would have been similar to the Kottawa Multimodal Centre that’s in operation now.

Electrifying and modernizing the existing railways was a major recommendation of the report. If you’ve ever used the railway system, you’d know that train delays and derailments are frequent.

The Railway Efficiency Improvement project aimed to support immediate improvements in the SLR’s capacity to run its operations, maintenance, safety, skills development, and implementation.

The train modernization project was divided into four chunks:

LineEstimated CompletionInitial Cost (2013 USD)Operating and Recurring Costs (2013 USD)
Coast Line (Colombo Fort - Kalutara South)2035$596,600,000.00$11,900,000.00
Main Line (Colombo Fort - Veyangoda)2025$730,600,000.00$14,600,000.00
Puttalam Line (Ragama - Negombo)2025$375,100,000.00$7,500,000.00
Main Line (Colombo Fort - Maradana)2020$90,300,000.00$500,000.00

There were two additional new rail projects outlined in the plan.

  • The first was the Airport Connection. The idea was to construct a railway that would connect Katunayake South to the Airport Terminal. The project was supposed to be completed by 2025.
  • The second was the Dompe Line, a Freight Rail Line meant to connect Kelaniya to Dompe for the specific purpose of transporting bulk products like oil. It was designed to stymie traffic caused by cargo carrying trucks in northern parts of Colombo. Its completion was also expected to arrive by 2025.

Out of the project recommendations in the CoMTrans plan, the rail modernization project made the most progress. The Megapolis Project, which we’ll explore in the next section, continued this section of the plan.

The Western Region Megapolis Plan

The concept of the Western Megapolis has been around for a while now.

The Megapolis Plan, as we’ll refer to it from here on, divides the Western Province into 12 planning areas and industrial zones. It drew a lot of inspiration from industrial cities such as those of Dubai, Singapore, Tokyo, and Seoul. It was first introduced in 1991, by our now president Ranil Wickremesinghe, when he was the Minister of Sri Lankan Industries’ Science and Technology. It’s one of the pet projects: he revisits every time he’s in power.

Here’s a hypothetical projection of what this project would have looked like, if it ever saw the light of day.

In September 2015, Ranil Wickremesinghe appointed Champika Ranawaka as the Minister of Megapolis and Western Development, who in turn presented the Megapolis Master Plan in November 2016.

The Megapolis report uses data from the CoMTrans report quite extensively. In fact, they unequivocally agree on the problems and some solutions. For example, the Railway Modernization and Electrification project is very similar across these two plans. The biggest difference between the two is when it comes to Rapid Transit plans.

However, while they agree in places, the Megapolis plan replaced all rapid transit projects outlined in the CoMTrans plan. That is to say: the Monorail, Bus Rapid Transit, and Bus Priority lines were replaced by a Light Rail Transit System (LRT).

Why, though? Well, the CoMTrans project estimated that the population of the Western Province would grow to 7.5 million people by 2035. This assumes an Annual Compounding Growth rate of 1.4%.

The Megapolis report disagreed with this estimate. They projected that the population of the Western Province would grow to 9.1 million people by 2035.

Based on these new projections, the Megapolis plan argued that the BRT projects wouldn’t be able to keep up with demand. They also argued that in order for BRT systems to be efficient, large sections of the route would need to be elevated. This would make the cost-effectiveness argument for BRT void. Instead, they recommended an alternative transit network made out of rail.

For this new rail network, the Megapolis team selected Light Rail Transit over Monorail. The Megapolis team released a document titled “Why Light Rail Transit (LRT) was selected over Monorail” explaining the reasons behind this decision. Their main arguments were that it’s a more widely used technology, that there were lots more companies manufacturing rolling stock for light rail, and that competition would mean better pricing for the buyer (i.e.: Sri Lanka). They also projected that we’d save some US$500 million by ditching the newfangled monorail and going LRT.

This is what the proposed LRT network would have looked like.

Where are we now (as of 2022 November)?

In 2019, Gotabaya Rajapakse won the presidential election and formed a new SLPP government. Like clockwork, the cutting of the previous government’s plans started.

In September 2020, Secretary to the President, P. B. Jayasundara, sent out a memo terminating the LRT project outlined by the Megapolis master plan. The memo claimed that the LRT project wasn’t a cost-effective solution for the Colombo urban area.

This decision was met with a lot of national outrage. For one, the loan extended by the Japanese International Cooperation Agency (JICA) for the project was seen as very forgiving. Second, it was considered a diplomatic fuck-up given the long-standing relationship between Sri Lanka and Japan.

As of today? We have nothing. The Government also seems to have pissed off another big donor - the Asian Development Bank - so we’ve effectively alienated one of our longest-standing, richest partners and shot ourselves in the foot multiple times. Udaya Gammanpila has pushed some stuff into Parliament that essentially looks like pieces of these ambitious plans, rebranded: four elevated rail tracks above Colombo. Gamini Lokuge also fielded another attempt to bring back the monorail. But the reality is that until something manifests in reality, these plans are ephemeral at best. After all, we still owe money on that cancelled JICA project, and our economy isn’t going anywhere in a hurry.

How can we do better?

Remember our axiom? Good transport helps. Bad transport hurts.

Diagram to illustrate the relationship between transport disadvantage, social disadvantage and social exclusion. The graph was taken from the paper Transport and social exclusion: Where are we now?, by Karen Lucas, and redrawn for ease of reading.

In the process of doing this research, we came across a few ‘obvious’ suggestions over and over again:

  • Circular transport networks for local travel. A circular transport network would allow people within a community to access their neighborhoods. Instead of a linear “home, office and back home” set of routes, people should be able to access everything from grocery stores, to restaurants, local playgrounds, salons, hardware stores, public places and so on.
  • Bike lanes and improved pavements are a given. This one comes up quite often.
  • Buses and trains that stop paying so much attention to the “last mile” and instead focus on providing better transportation between population hubs. In Sri Lankan terms, for example, this would mean simplifying the multiple types of 138 buses, or reducing the number of out-of-Colombo buses that go all the way to Fort, and instead pass off passengers to better-provisioned lines.
  • Human usability. The height of bus steps could be lowered significantly. The same could be said of handrails on buses. Bus services as they are don’t make it easy for people with disabilities.
  • Taking women and children into account in transport decision making. Bus services could implement measures to become safer for women and children, which is . . . really, a majority of people using them. This includes both infrastructure and enforcement.
  • Better terminals are something we could have more of. Ones with better lighting, cleaner toilets, and reliable security at terminals to help resolve conflict and harassment whenever they may arise.
  • Ticket systems that incentivise off-peak travel (think of it as surge pricing in reverse: trying to load-balance a system).

On the minus side: these are hard problems. On the plus side: none of them are impossible, and there are plenty of lessons in this world about how we can do better and what we should avoid. In fact, we wrote a whole other article about how other countries - Singapore, South Korea, Thailand and Malaysia - managed.

Of course, we need actual coordination between the things that move and the things that decide (aka the politicians). We need a baseline level of forward-planning. We need robust statistics and the ability to make decisions based on sound theory, and not just ad-hoc cronyism.

That might be the hardest task of all.

Thank you for reading.

This report was created by Watchdog Sri Lanka. Yudhanjaya Wijeratne, Nisal Periyapperuma, Umesh Moramudali, Tineeka de Silva, Joshua Perera, Nadim Majeed, Nishadi Gunatilake, Ishan Marikar, Sathyajit Loganathan, and Sachintha Adikari contributed to this work. We’d like to thank Professor Amal Kumarage for sharing his research and Sriganesh Lokanathan for his input.

Data for this piece


[1] A word here which refers to damn near everything - from roads to rail to the tollbooth offices to security cameras to train engines to traffic cones to speedbumps to the beams that hold up highways . . . you get the picture.

[2] A lot of the data will be from 2020. This is partly because a lot of the most recent government figures are for 2020, but also because 2021 and 2022 aren’t good reference points - one year we had COVID lockdowns, the next year we had economic collapse. These are outlier events that temporarily nuke the general trend.

[3] No actual people were harmed in the making of these fractions (at least, we got rid of the bodies).

[4] In fact, it’s practically cryptic. The explanations are atrocious, and we have serious questions about the variable choices - income and education, for example, tend to be correlated; and in some of their charts, they interpret ‘occupation’ as a separable variable from ‘income’, when these are again two highly correlated things. There are also variables like ‘vehicle ownership’ as being highly correlated with all choices- of course you can’t use a vehicle you don’t have.

[5] Disclaimer: no puppies were harmed in the making of this paper. A fair collection of numbers were scrunched, but we’re fairly sure you’re thanking us for that.

[6] Also, there are interesting parallels. For example, the Georgian railway has some pretty famous corruption being reported all the time, especially around tenders, construction deals, and bribery.

[7] We’ve written a lot about this subject. See: and

[8] A disclaimer on the numbers here - while we strive for perfect accuracy, this is not always possible. Mostly you have to trust that the experts at the institutions gathering this data know what they’re doing.

This data is from the Department of Census and Statistics, which is the entity that estimates Sri Lanka’s GDP based on the standards defined by international institutions such as the IMF and World Bank. Be mindful of the word “estimate” here. This means that the Department of Census and Statistics calculates the amount of goods and services produced in Sri Lanka within a given timeframe based on the primary and secondary data available to them. Transport is considered a service activity which includes four main economic activities, namely, ‘land transport’, ‘Water transport’, ‘Air transport’ and ‘Warehousing and support services to transportation’.

GDP estimates for transport are made using revenue reports of the Department of Railways, SLTB, bus stock, daily operated number of buses and few other indicators. When it comes to private vehicles used for transport services (cars, three wheelers, lorries etc.), the GDP estimate is calculated by multiplying the average revenue generated per vehicle by the total number of vehicles, with necessary adjustments. Storage facility estimates are derived based on statistical data provided by the Sri Lanka Ports Authority and Airport and Aviation Services (Sri Lanka) Limited.

Public Transport Sytems Costs

How do other countries handle transport? Some brief case studies

Inter Provincial Bus Routes – Normal Buses

Inter Provincial Bus Routes - Semiluxury Buses

Inter Provincial Bus Routes - Luxury Buses

Inter Provincial Bus Routes - Super Luxury

Western Province Bus Routes

A Brief History of Sri Lankan Transport

[9] We’re using data from the 2019 HIES (the Household Income and Expenditure Survey) by the Department of Census and Statistics, as well as pricing data scraped from classifieds websites. Results are intended to give you a back-of-the-napkin idea, not be accurate down to the last T.

[10] Including, but not limited to:

இலங்கையில் போக்குவரத்து: வேகமற்றது, வேதனை தருவது