Commentary: It’s worth rethinking the role of taxis in our transport network, as industry shrinks

SINGAPORE: Are taxis in a death spiral as an essential transport service in Singapore?
Recently, my wife was at the One Holland Village mall looking for a taxi. After waiting at the mall for some time, she finally got a taxi on the Comfort app. No point waiting at the taxi stand, the driver told her laughingly, there would be no taxis.
The demand for and supply of taxi rides depend on each other in a self-fulfilling way. Taxi drivers go to taxi stands expecting customers, and customers go to taxi stands expecting taxis.
However, as my wife’s experience illustrates, the cycle can work in the opposite direction. If taxi drivers do not expect passengers at a taxi stand, they will not waste time queuing there. If customers do not expect taxis at a stand, they will not go there.
The same goes for the whole industry: The more it contracts, the more it would tend to continue contracting – hence my speculation of a death spiral.
But would it really be conceivable for Singapore to have no taxis in the future?
If the only way to get a taxi is through an app, taxi services would be just one tap away from ride-hail services. From the customer viewpoint, taxis would lose the competitive advantage of immediate, convenient availability on the street without any booking.
Side-by-side on the phone (even on the same app), the competition boils down to how quickly we can get a ride and how much it costs. It may be challenging for taxis on both counts.
Availability is a challenge as the pool of taxis dwindles. Over the last 10 years, between 2014 and 2023, the number of taxis fell by more than half from over 28,700 to 13,300. It is no coincidence that, during that same period, the number of private hire cars exploded from 1,600 to 53,400 as Uber, then Grab followed by Gojek entered the Singapore market.
Moreover, taxi fares seem to keep going up. In December 2023, taxi operators introduced a new weekend peak-hour surcharge and extended the daily evening peak surcharge by one hour. Comfort DelGro, the largest taxi operator, also increased its flagdown fare and distance-based rates for metered rides.
The two factors – availability and prices – are subtly related. Recently, my research team surveyed over 800 ride-hail drivers and found that 45 per cent drove or had previously driven taxis. Taxi operators are quite possibly raising prices to stem the defection of drivers to ride-hail platforms.
Besides competition from ride-hail services, demographic trends in the demand and supply of taxi services have also forced the industry to shrink.
On the demand side, younger people are better able to book transport through mobile apps. Older people are less tech-savvy and tend to rely on conventional methods – street hail and taxis stands. But the older residents who depend on taxis are a declining market segment.
On the supply side, taxi drivers tend to be old. In 2022, the average taxi driver was 60 years old. That is 15 years older than the median among ride-hail drivers surveyed by my team.
Moreover, the average age of the ride-hail drivers who currently or previously operated taxis is 47. Unsurprisingly, it is the younger taxi drivers who are switching to ride-hail platforms or serving them concurrently.
By statute, taxi drivers must retire at the age of 75 (that age limit was raised from 70 to 73 in 2006 and then again to 75 in 2012). On the other hand, LTA statistics show no sign of any obvious increase in new drivers. Absent fresh blood, the supply of taxi services will surely decline with the passage of time.
In the longer term, it is conceivable that taxis will shrink to become a high-end niche service, catering to tourists at the airport, zoo and casinos.
How might it be in Singapore’s interest to maintain some form of professional taxi industry for the masses?
Importantly, the law requires taxis on the street and taxi stands to serve any customer regardless of distance. Yes, some drivers can sometimes be choosy about passengers going in a convenient direction, but this tends to be confined around shift changes.
The Mass Rapid Transit (MRT) is the backbone of our public transport network, the optimal mode of transport in a land-scarce nation. But it is not cost effective to provide train service to everyone’s doorstep. To serve the so-called “last mile”, we rely on buses, bicycles and walking.
Presently, taxis and private ride-hails are competing in the space of point-to-point direct commutes, for those willing to pay more for faster journeys than the MRT but do not want to drive a car themselves. This is currently under-priced because our current system charges for car ownership, not road usage, as I have previously argued.
If the market has determined that ride-hails are dominating this space, there is another that taxis could conceivably fill: As a point-to-point last-mile service. Ideally, we could even step out of an MRT station to the taxi stand, and then conveniently complete the last mile.
But not if there are no taxis at the stand and we must book through an app. Then the old bugbear of taxi drivers avoiding short trips re-emerges.
My wife was lucky that day at Holland Village. She did get a taxi. Some weeks later, we had a different experience at the same mall. We spent minutes fruitlessly tapping on multiple apps. Meanwhile, several taxis dropped passengers and drove away presumably to serve bookings. The ride to our home was just too short for any driver to want it.
We ended up taking the bus. But that is an impractical solution for older people, those with disability, families with toddlers, or people doing their weekly grocery shopping. It is worth rethinking how taxis and ride-hail can better complement trains to form an effective transport system.
Ivan Png is Distinguished Professor at the National University of Singapore (NUS) and Principal Investigator of BeWork, a research project funded by the Social Sciences Research Council.