Odds, Ends, and Evens

Will the MMDA’s new number coding scheme help traffic or make it worse?


The MMDA is set to implement yet another iteration of number coding, this time restricting odd- or even-ending plate numbers on EDSA throughout the weekday. The proposed scheme will not replace the existing Unified Vehicular Volume Reduction Program, but instead be additional to the current scheme.

MMDA Proposed Odd-Even Scheme for Mondays to Fridays

Cars are banned from EDSA during the following times:

Odd Numbers

Even numbers

7-9 AM

9-11 AM
1-3 PM

3-5 PM

5-7 PM

7-9 PM

11 AM – 1 PM : Grace Period for all cars

Let’s analyze what the new coding scheme means for transport in Manila.

What do we already know about coding?

There are few iron laws in transport economics, but just as evidence reveals, time and again, that building more roads does not get rid of traffic (Duranton and Turner 2011); number coding (in the form it currently takes, as of this piece’s writing) has been shown to worsen traffic by increasing driving and car ownership. Research in different country settings by Cantillo and Ortuzar (2013) shows that benefits of traditional number coding policies are short lived, while negative effects persist.

This coding scheme is quite a bit different, though, because there is an added restriction in the form of hours within a day, every weekday, in addition to days within the week. Let’s look at what this policy could mean.

How might travel patterns be affected?

There are roughly five ways that drivers may change their behavior in response to the new policy.

  • Drivers may shift their departure times.
  • Drivers may change their routes to avoid EDSA.
  • Drivers may shift to using cars with plates that fit their desired travel pattern. This may be by purchasing another car or by choosing to use a taxi service (including “ride-sharing” services such as Uber or Grab).
  • Drivers may shift to public transport or other modes, which will not be subject to coding on EDSA.
  • Drivers choose to travel less.

Note that this is not a flowchart-type decision tree, and individuals’ decisions will depend on a number of factors, including their sensitivity to travel time and schedule shifts, as well as the ability of the transport system to meet their travel needs.

Evening things out – “Peak Spreading”

The first two possible behavior shifts – changing departure times or travel routes – represent an outcome known as peak spreading. The theory is that instead of loading the network very heavily at a particular point in time or space, spread the loading out so that congestion at its worst is less severe. This may improve travel times during peak hours, but peak spreading has a major limitation: it can only ease traffic temporarily. Eventually, induced demand catches up with you (see Duranton and Turner, 2011) and what once were non-peak hours and routes also become congested.

The MMDA’s announcement of the new scheme came with a promise to clear a number of “Mabuhay Lanes” around the metro, to facilitate spreading the vehicle load to different physical locations on the network. While this certainly means new, less loaded routes for cars, it’s worth asking at what cost this comes. Some of these “mabuhay lanes” may be currently obstructed because they have become extensions of peoples’ homes, businesses, and communities – places of economic and social importance in their lives. How many of these places have to become rivers of high-speed traffic to make life easier for Manila’s drivers?

Taken by (different) cars – simple license plate substitution

The third possible reaction is that commuters may choose to use cars whose plates correspond to their desired travel patterns. This may be as simple as substituting to a second car that has the right plate number, or choosing to take a taxi (including Ubers, Grabs, and similar services). In these cases, there is no peak spreading or congestion reduction and private vehicle travel stays the same. If most drivers react this way, the policy may be rendered ineffective. Even then, people may still purchase new cars to fit their travel patterns, eventually leading to all the induced driving effects of old coding.

Decoded – Will lifting coding on public transport do the trick?

The MMDA’s statement also mentioned that alongside the policy, coding will be lifted on public utility vehicles to encourage drivers to take public transport. Every person that shifts from a private car to a public vehicle is an incremental improvement in how we use space. These gains, while small, are precious in an urban environment where space, according to transit expert Jarrett Walker, is the “ultimate currency.”

Lifting coding on public transport is a welcome move. Fleet sizes are artificially high because of coding, and that pushes fares up for commuters without adding value. But the government has to guarantee service levels and quality if it wants to attract existing car riders and drivers to public transport. Some passengers who travel by car do so because public transport currently fails to meet their needs for reliability, comfort, safety, or coverage. While industrial reform in the PUV sector would help meet these needs, this is a longer-term goal and will not bring immediate improvements.

Furthermore, part of the MMDA’s plan appears to be to “remove the illegal terminals in Pasay and in Cubao, Quezon City, within the month [of April].” Care must be taken to ensure that this move does not make public transport worse. Illegal terminals, like colorum services, exist because of underlying demand that is not met by the formal system. Passengers who ride to and from those terminals also need to get to work, school, and other activities. Removing the terminals without providing useful substitutes will burden these riders further.

Odd Men (and Women) Out

Finally, travelers may react to the new policy by choosing to travel less on EDSA. This may not be a short-term shift, but if the policy is in place long enough, it may impact where people choose to locate their homes, jobs, and other activities. While travel reduction reduces congestion, this may also represent reductions in access to economic opportunity, so we have to be careful about calling this a victory.

The policy also places a disproportionate burden on odd-ending plates. Assuming that most workers adhere to an 8:00 or 9:00 AM clock in up to a 6 PM clock out, odd-ending plates are at a bit of a disadvantage. Because their cars are banned during the traditional morning (7-9 AM) and evening (5-7 AM) rush hours, workers with odd-ending plates are under more pressure to change their travel behavior in response to the policy.

The current coding system has a semblance of fairness in that everyone is inconvenienced to about the same degree. In the new system, drivers with odd-ending plates will bear more of the burden for decongesting EDSA. Flex-time hours may mitigate this hardship somewhat, but keeping to the traditional workday schedule also means things like resting and rising with the day/night cycle, seeing your kids after they come home from school, and more. And if drivers with odd-ending plates react by buying cars with even-ending  plates, we could end up even worse off than we are under the existing coding scheme.

Why dont’t we just go for congestion charging?

On its face, the new scheme is more difficult to understand. Rather than just two ending numbers banned during most hours of the day, now all plate numbers are banned in one way or another during multiple short windows during the day. Enforcement, however, should be rather simple through the MMDA’s existing no-contact apprehension system, which would be presumably expanded to implement this policy.

But if we concede automated enforcement is the way to go, why not make the small marginal investment for congestion charging on EDSA? Installing automated license plate recognition technology on top of the existing CCTV network should be simple. The balance could be billed upon registration and paid to the LTO or MMDA offices. In fact, you could probably repurpose the May Huli Ba site to let people check their congestion charge balance. If you want people to shift to more space-efficient modes while traveling on EDSA, it’s a more elegant, proven solution that cannot be evaded as simply as coding. Achieving a peak spreading objective can be done more transparently and precisely through congestion charging, by adjusting the toll higher during peak hours.

How do we get the change we want?

Since the new scheme is added on top of the existing UVVRP,  we can expect more of the same transport outcomes – higher car ownership, more driving, more pollution, and more traffic. As I have argued previously, one of the key ways to make progress is to scrap UVVRP and put in a demand management scheme that is proven to work in the long term, such as congestion charging or parking levies. While congestion charging seems inequitable to some, it’s nowhere near as oppressive to the poor as auto-dependent development is – and that’s where we are now. Congestion charging AKA road pricing may seem daunting, but we already have some infrastructure for it, and it’s high time for a progressive solution instead of more of what doesn’t work.

Assuming that this new everyday odd-even coding scheme will proceed, we have to keep the end goal in mind as we plan complementary transport policy. We’ll need to ensure that the near-term peak spreading effects are followed up by long-term modal shifts to public transport, walking, or biking. Substitution to private cars with different plate numbers should be minimized. Access to economic opportunity should be maintained as much as possible.

To achieve these goals, government will need the full policy toolbox it holds between different government agencies. In addition to the MMDA’s power to enforce a coding or demand management scheme along EDSA, the government can use the LTFRB’s powers to define bus stops on EDSA that will be strictly enforced. In return for the lifting of number coding, franchise operators should agree to a stricter enforcement regime that defines and makes them accountable for service quality. The currently “illegal” terminals should be formalized and improved, and existing bus stops along EDSA should be upgraded to humane standards for safety, at the very least – including working lights and well-maintained shelters.

Making a dent in Manila’s hellish traffic won’t be easy, but clarity about how we attack it will help us avoid making it worse. While the new form of number coding seems imposing, we have to be forthcoming about what it can and can’t do, and where other blanks have to be filled.

EDIT: An earlier version of this story erroneously referred to the new scheme as a replacement of the current UVVRP. This piece has been corrected to reflect that it will complement the existing UVVRP.

1 thought on “Odds, Ends, and Evens”

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s