Latest / Talkback / #TalkBack: Besides safe driving, should we also improve platform expectations for riders?
Transcript
- 0:01Speaker 1: Now, talk back with Daniel Martin on CNN 938. The
- 0:06Speaker 1: gig economy, my gosh, I remember doing shows on the
- 0:09Speaker 1: gig economy when it first started ballooning in Singapore. Some
- 0:15Speaker 1: people were saying, oh, it's just a temporary thing, you know,
- 0:17Speaker 1: like it won't last in the long term. Look at
- 0:20Speaker 1: how long it's lasted. There are, it's a significant employer,
- 0:22Speaker 1: so to speak, even though it's gig economy and uh
- 0:25Speaker 1: it's freelance in a way some, in some cases.
- 0:28Speaker 1: Quite a proportion of the population seeks their earnings that way.
- 0:34Speaker 1: Now when it comes to the safety issue on the
- 0:36Speaker 1: roads for our platform delivery riders, here is a worrying statistic.
- 0:42Speaker 1: 2 platform delivery riders died in road accidents for the
- 0:45Speaker 1: whole of last year.
- 0:48Speaker 1: This year, and it's so far mid-August.
- 0:51Speaker 1: It's 4 who've died in the first half of this year,
- 0:55Speaker 1: platform delivery riders.
- 0:57Speaker 1: We've seen therefore, the emergence of a new rider safe program.
- 1:02Speaker 1: Here's CNN's Nicholas Ung with more.
- 1:05Speaker 1: The goal of this new safety course is to help
- 1:08Speaker 1: riders improve their riding skills. It's set up by GRAB,
- 1:11Speaker 1: the Riding Safety Association of Singapore, and the union representing
- 1:15Speaker 1: platform workers. Experienced riders are part of this pilot session.
- 1:20Speaker 1: They'll help fine tune the course to meet the needs
- 1:22Speaker 1: of less experienced riders.
- 1:26Speaker 1: Hear that report on CNA. Asia and the CNA YouTube
- 1:30Speaker 1: page in full. Right now, I want to talk about, yes,
- 1:34Speaker 1: it is important. I mean, some of these packages that
- 1:37Speaker 1: they're getting from various different restaurants and uh,
- 1:40Speaker 1: Um, F&B or supermarkets, there's no standardized packaging. It's coming
- 1:44Speaker 1: in all shapes and sizes. How is it fitting into the, the,
- 1:48Speaker 1: the existing compartments that they have? Is it leading to
- 1:52Speaker 1: some unsafe practices? Is it affecting their ability to make
- 1:54Speaker 1: sharp turns and corners? Those are many of the things
- 1:57Speaker 1: that are covered in this rider safe program, or rather
- 2:02Speaker 1: ride safe program.
- 2:04Speaker 1: But I also want to look at another piece of
- 2:06Speaker 1: the puzzle.
- 2:08Speaker 1: Besides, you know, the element of safe driving, should we
- 2:11Speaker 1: also be looking at platform expectations for riders? These are
- 2:15Speaker 1: just questions that I have. Um, I'm not saying that
- 2:18Speaker 1: it is or isn't the case, but for example, are there, uh,
- 2:23Speaker 1: incentives to make sure that you complete deliveries very fast?
- 2:28Speaker 1: Are there cash incentives, other bonuses to make sure that
- 2:31Speaker 1: it rewards people to at least a minimum number of
- 2:34Speaker 1: delivery rides per day?
- 2:36Speaker 1: I wanna know more about the platform side, besides, outside
- 2:40Speaker 1: of the driving element.
- 2:42Speaker 1: Do you remember there was this period of time there
- 2:44Speaker 1: was a lot of heavy vehicles rushing on the highways
- 2:48Speaker 1: and it was discovered that one of the key factors
- 2:51Speaker 1: was they had to clock the certain number of point
- 2:54Speaker 1: A to point B's. If not, they wouldn't get paid
- 2:56Speaker 1: for that ride and stuff like that. So that was
- 2:58Speaker 1: addressed and I think that's improved the situation. So, let's
- 3:01Speaker 1: take a look at the other side of the puzzle
- 3:03Speaker 1: when it comes to the safety of our platform riders.
- 3:08Speaker 1: Besides safe driving, should we also improve platform expectations?
- 3:13Speaker 1: Have your say. It's Talkback with Daniel Martin. Call in
- 3:18Speaker 1: 669-11938 or WhatsApp 963-11938. 0 no, the phone is charging
- 3:27Speaker 1: far away from me and I cannot reach it. I,
- 3:29Speaker 1: I'll get it soon. OK, and I'll open up the
- 3:32Speaker 1: WhatsApp hotline. I can't wait to hear from you.
- 3:35Speaker 1: And your thoughts. Many of you might actually be performing
- 3:37Speaker 1: this job for uh at least a little while or
- 3:39Speaker 1: maybe for years now, and maybe you'll be able to
- 3:41Speaker 1: give me some insight, and many of you might also
- 3:44Speaker 1: have family members or relations who are in this gig economy.
- 3:48Speaker 1: Please talk to me about some of the expectations um
- 3:51Speaker 1: that the platform may place on the riders and how
- 3:53Speaker 1: this could actually be potentially contributing to unsafe practices on
- 4:00Speaker 1: the road.
- 4:01Speaker 1: Welcome back to the show, Professor Sumiarawal who's Professor of finance,
- 4:04Speaker 1: real estate and Economics at NUS Business School. He's written
- 4:08Speaker 1: on the platform economy as well as the idea of
- 4:10Speaker 1: protections for platform workers.
- 4:15Speaker 1: Professor Agarwal, welcome back. hello.
- 4:17Speaker 1: How are you? I'm well, thank you. I didn't know
- 4:19Speaker 1: this was a key area of interest for you. Um,
- 4:21Speaker 1: was this something that was interesting at the, particularly the
- 4:24Speaker 1: birth of the gig economy for you?
- 4:26Speaker 2: Yes, that's how I started getting into it, thinking about it.
- 4:29Speaker 2: I do work on transportation and understanding this three-sided market
- 4:34Speaker 2: where we are thinking about consumers, the riders, and the
- 4:39Speaker 2: stores or the the restaurants. It's very interesting how this
- 4:43Speaker 2: market works and how an app actually brings these three
- 4:47Speaker 2: segments together and the trade-offs that everybody has to make.
- 4:52Speaker 1: I wonder whether at the core of the issue is
- 4:55Speaker 1: the fact that this is because it was the gig economy. Um,
- 5:00Speaker 1: it was very much, you know, IT based companies launching these.
- 5:03Speaker 1: They weren't fleet managers, they were not human resource people,
- 5:06Speaker 1: they were not people, people, um, people employers. It was,
- 5:09Speaker 1: it was primarily tech companies that came up with, you know,
- 5:11Speaker 1: the tech and the apps and the systems and then
- 5:13Speaker 1: later found themselves managing large groups of people.
- 5:16Speaker 1: Is it at the root of it, the issue that
- 5:19Speaker 1: delivery work relies on very much a piecemeal algorithmic pay model?
- 5:27Speaker 2: Mm
- 5:29Speaker 2: Yes, I mean, this is exactly what it is, that
- 5:32Speaker 2: this algorithmic pay model sometimes creates incentives that can have
- 5:40Speaker 2: unintended consequences, essentially, for drivers to take excessive risk.
- 5:46Speaker 1: OK, um, operationalize that for us, um, pace it out
- 5:49Speaker 1: for us. If I, it's not the same as like,
- 5:51Speaker 1: you don't get paid the same for doing 10 food
- 5:53Speaker 1: deliveries per day.
- 5:55Speaker 2: It is, but what happens is, let's say you are
- 5:58Speaker 2: trying to do 10 deliveries, and then you are kind of, uh,
- 6:02Speaker 2: you've patched them up together, the last delivery, you're really
- 6:06Speaker 2: trying to make it as quickly as possible, so you
- 6:09Speaker 2: need to hurry up.
- 6:11Speaker 2: The way I would try to think about it, we'll
- 6:14Speaker 2: have to distinguish not just what the rider makes the decision,
- 6:18Speaker 2: but in what environment is the rider making these decisions.
- 6:23Speaker 2: So if the income depends on how many deliveries I
- 6:26Speaker 2: have to complete.
- 6:27Speaker 2: If the app continuously shows me my estimated time, if
- 6:32Speaker 2: the customers are waiting, and if the peak period is
- 6:36Speaker 2: when you earn the most, then all of this putting
- 6:39Speaker 2: together creates perverse incentives for the rider to take some
- 6:44Speaker 2: additional
- 6:44Speaker 2: risk.
- 6:46Speaker 1: So just like how we've seen surge pricing or midnight
- 6:49Speaker 1: pricing impact taxi availability or, or cost of um private
- 6:55Speaker 1: hire transportation, likewise, because that model is very much at
- 6:59Speaker 1: the core of our platform delivery riders model, then it could,
- 7:03Speaker 1: it's probably, it is coming up with situations where riders
- 7:07Speaker 1: are gonna try and fulfill as many orders as possible
- 7:10Speaker 1: in their push to earn more income.
- 7:13Speaker 2: Yeah, so let, let me be more explicit. I think
- 7:15Speaker 2: the way we want to, I want to think about
- 7:17Speaker 2: it is, one is the proximate cause of the accident
- 7:22Speaker 2: that because, you know, you are there in that situation,
- 7:25Speaker 2: you are taking slightly excessive risk, and that's what is
- 7:28Speaker 2: causing the accident. But there is also an underlying cause,
- 7:32Speaker 2: and the underlying cause is this incentives and the constraints
- 7:36Speaker 2: that the riders have to operate under that look, you
- 7:39Speaker 2: have to do X deliveries.
- 7:40Speaker 2: This is the time the customer is watching. Once you
- 7:43Speaker 2: tell the customer the delivery will arrive in 18 minutes,
- 7:46Speaker 2: and that the driver is late by 2 by 2 minutes,
- 7:50Speaker 2: the customers, doesn't give you all the stars, so the
- 7:53Speaker 2: driver essentially takes some risk. That is the underlying cause.
- 7:58Speaker 2: These all these things pushes the driver to do this.
- 8:02Speaker 1: I, the rider, didn't decide on how long that ride
- 8:05Speaker 1: was going to take. The app came up with that, no. Yes, exactly.
- 8:10Speaker 1: Uh, there's also this aspect to weave in, and it's
- 8:14Speaker 1: worth talking about that according to published reports, there are
- 8:19Speaker 1: tier-based bonus structures for platform delivery riders. Uh, for example,
- 8:26Speaker 1: this is from a 2026 article. Uh, if they were
- 8:29Speaker 1: able to complete 30 deliveries in 3 days, they would
- 8:32Speaker 1: get a $50 bonus.
- 8:34Speaker 2: Yes, and that's the point. This is this underlying cause
- 8:37Speaker 2: that I'm saying that creates incentives and constraints under which
- 8:42Speaker 2: the riders have to operate and make decisions. So the
- 8:47Speaker 2: proximate cause is, look, they, they made the wrong turn
- 8:50Speaker 2: or they were just being careless, but the fact is
- 8:53Speaker 2: that this all happens because of the underlying incentives and punishments.
- 8:58Speaker 1: You know, big multinational companies or, or mega construction companies
- 9:04Speaker 1: do do cost-benefit analysis in terms of
- 9:08Speaker 1: Numbers of lives that may be lost during construction and
- 9:11Speaker 1: things like that. That is uh uh something that they
- 9:13Speaker 1: do have to factor in. It's worrying to see that
- 9:15Speaker 1: an individual person might be thinking about how much damage
- 9:18Speaker 1: my vehicle might get, like, um how much a traffic
- 9:22Speaker 1: summons will cost me versus the bonus that I get.
- 9:26Speaker 1: This is, you think, running through the minds of our
- 9:29Speaker 1: platform riders?
- 9:30Speaker 2: Yeah, because I could be the top bonus earner. I mean,
- 9:35Speaker 2: this is true for any, I mean, even if you
- 9:37Speaker 2: look at Grab drivers who are taking Grab taxi rides,
- 9:41Speaker 2: they all want to earn the maximum bonus, or if
- 9:43Speaker 2: they're doing working on peak times, they can make so
- 9:45Speaker 2: much extra than the other. So that could say, OK,
- 9:49Speaker 2: I'm willing to take this job while it's raining heavily.
- 9:51Speaker 2: I'm willing to take the customer or being willing to
- 9:54Speaker 2: make this delivery. That's the excessive risk you end up
- 9:57Speaker 2: taking because at peak time you get paid more.
- 10:00Speaker 1: Going to our WhatsApp hotline 963-11938. 1 of my listeners
- 10:04Speaker 1: WhatsApping across, simply just saying, I've had two occasions where
- 10:08Speaker 1: the rider has messaged me and sent me photos that
- 10:11Speaker 1: my food has been on the floor because they were
- 10:14Speaker 1: involved in an accident. Both times they were trying to
- 10:16Speaker 1: make sure that I wasn't upset, but really I was
- 10:18Speaker 1: just more concerned for them and that they were OK,
- 10:21Speaker 1: but the fact that this happened twice to me.
- 10:24Speaker 1: Uh, does make me wonder how challenging it is for
- 10:27Speaker 1: them on the roads, particularly when they're using, oh, OK,
- 10:31Speaker 1: particularly when they're using vehicles that may be modified or
- 10:33Speaker 1: not road suited as well. Maybe that's another part of
- 10:37Speaker 1: the discussion. Thank you to our listener for that WhatsApp,
- 10:38Speaker 1: by the way. I, I myself have received on occasion, um,
- 10:42Speaker 1: a couple of times, messages from delivery riders who say,
- 10:45Speaker 1: I'm sorry, I can't complete the order. I've been involved
- 10:47Speaker 1: in an accident, and the fact that they have to
- 10:50Speaker 1: take the onus of responsibility to message me when they've
- 10:53Speaker 1: just been in an accident.
- 10:54Speaker 1: is bewildering to me, um, because I'm like, dude, deal
- 10:57Speaker 1: with the accident, don't worry about it, just send a
- 10:59Speaker 1: cancel signal or something like that. Please take care of yourself.
- 11:00Speaker 1: I hope you're OK. Uh, that's one thing, but also
- 11:03Speaker 1: the other thing that my listener brings up, in some
- 11:05Speaker 1: cases they might be using vehicles that are not really
- 11:07Speaker 1: suited for the job.
- 11:11Speaker 2: Yes, I mean, so this comes back to the game,
- 11:13Speaker 2: the design, the underlying cause, because by messaging you, they're
- 11:19Speaker 2: also signaling.
- 11:21Speaker 2: To the to the company that look, they're responsive, they are,
- 11:25Speaker 2: they care for the customer, and that earns them at
- 11:28Speaker 2: least some bonus point for the company that look, you
- 11:31Speaker 2: are getting, you are helping us with customer satisfaction and
- 11:34Speaker 2: customer care, even though you're right, at that point, they
- 11:37Speaker 2: should not care about letting you know that they're in
- 11:40Speaker 2: an accident, they should take care of their life.
- 11:42Speaker 1: Yes, it's like how I don't expect somebody on MC
- 11:45Speaker 1: to tell me that like they can't take a call
- 11:47Speaker 1: or something like that. They're an MC. You're not supposed
- 11:49Speaker 1: to be doing the call, um.
- 11:51Speaker 1: So I wonder that's still
- 11:52Speaker 2: OK, but if you are in a real time, real situation, accident,
- 12:00Speaker 2: I agree you should not be sending messages.
- 12:02Speaker 1: Doctor Agawal, you're you're managing your people differently, huh? Uh, no,
- 12:05Speaker 1: I'm kidding. I'm, I'm just joking. I'm I'm curious though, like.
- 12:09Speaker 1: You see, on the one hand, I can see how
- 12:12Speaker 1: such incentives and structures.
- 12:16Speaker 1: Can be viewed as dynamic and encouraging and productive, but
- 12:22Speaker 1: I can also see how it has the drawback that
- 12:25Speaker 1: we're discussing here.
- 12:28Speaker 2: Yes, I mean this is what I'm pointing out that look, they,
- 12:32Speaker 2: they are good because look, some structure helps because otherwise
- 12:37Speaker 2: you are managing a three-sided economy. You're managing restaurants, you're
- 12:42Speaker 2: managing customer expectations. So if you do not have some
- 12:46Speaker 2: kind of structure, the customers will be unhappy and they
- 12:49Speaker 2: will walk away from the platform, and that will be
- 12:51Speaker 2: bad for the driver or the rider.
- 12:53Speaker 2: And it will be bad for restaurants. So I think
- 12:56Speaker 2: once we are managing this very complicated structure, we have
- 13:00Speaker 2: to make sure that each party is safe and happy,
- 13:05Speaker 2: and that's not easy. So you have to think about training, incentives.
- 13:09Speaker 2: And design, we need to train the riders. We have
- 13:12Speaker 2: to create incentives for behavior that pays riders and platforms
- 13:18Speaker 2: and everybody, and we have to design this thing in
- 13:20Speaker 2: a way that all people, restaurants, customers, and riders are
- 13:25Speaker 1: happy. You know, the, the restaurants component, you're so right,
- 13:29Speaker 1: because very often.
- 13:30Speaker 1: In their attempt to be
- 13:34Speaker 1: Green or eco or friendly, they might have different kinds
- 13:36Speaker 1: of packaging, different shapes, different sizes, different formats, different temperatures.
- 13:41Speaker 1: There there's no structure, there's no consideration in terms of
- 13:44Speaker 1: for the delivery medium. Is this suited for the delivery
- 13:47Speaker 1: rider's vehicle? Can there be a system where we at
- 13:52Speaker 1: least structure that where it's standardized packaging for everyone so
- 13:57Speaker 1: that it can fit safely into vehicles and on-platform riders'
- 14:00Speaker 1: bikes and cars?
- 14:03Speaker 2: I would think that's difficult because just by the nature
- 14:06Speaker 2: of the business that it is, that it's just, uh,
- 14:10Speaker 2: you know, you're ordering a pizza versus you're ordering something,
- 14:13Speaker 2: you know, small, they will have different packaging requirements. Something hot,
- 14:18Speaker 2: you have to keep it in a different setting. So I,
- 14:21Speaker 2: I just don't know. I mean, maybe I'm not a
- 14:24Speaker 2: packaging expert, but it seems like, uh, difficult to achieve.
- 14:30Speaker 1: We do know that attempts have been made many times
- 14:33Speaker 1: to try and address the safety of the gig worker.
- 14:36Speaker 1: The Platform Workers Act context was introduced January 1st, 2025.
- 14:42Speaker 1: It mandated work injury compensation for our platform workers and
- 14:46Speaker 1: it does provide some sentiment of financial protection after an accident.
- 14:50Speaker 1: It does not cap algorithmic speed pressures or target any
- 14:55Speaker 1: kind of um tier-based bonus structure though.
- 15:00Speaker 1: Do, do you think this is, I mean, it's, it's,
- 15:03Speaker 1: I'm glad it's being recognized on the scale. It has
- 15:05Speaker 1: to be recognized on the scale because so many people
- 15:07Speaker 1: are involved in the gig economy, but protections like that.
- 15:12Speaker 1: Is more of an after the fact kind of situation,
- 15:15Speaker 1: after the injury has occurred.
- 15:18Speaker 2: Yes, I mean, I think that those are all very
- 15:20Speaker 2: good things. I mean, we should constantly be re-evaluating, thinking
- 15:26Speaker 2: about safety education, uh, or actually safety by design under
- 15:32Speaker 2: what situation, what kind of safety, uh, uh, uh, strategy
- 15:38Speaker 2: or planning or training should happen, uh, but I think
- 15:42Speaker 2: these are such.
- 15:44Speaker 2: Different situations because at peak time the driver is trying
- 15:48Speaker 2: to kind of push or overwork and make all the
- 15:53Speaker 2: deliveries and keep the customers happy and keep the restaurants
- 15:56Speaker 2: because at times what happens is the restaurant is delayed
- 15:58Speaker 2: by 7 minutes instead of the rider taking the blame
- 16:02Speaker 2: for those 7 minutes, the app should just say that
- 16:04Speaker 2: the the the.
- 16:06Speaker 2: Food got delayed by 7 minutes. It's not the driver's fault,
- 16:10Speaker 2: and so we will come as soon as possible, or
- 16:13Speaker 2: it started raining and so we'll be delayed and so
- 16:16Speaker 2: the driver should feel less pressure
- 16:18Speaker 2: because of that.
- 16:18Speaker 1: I think some of the apps do that already. So I,
- 16:20Speaker 1: I've had cases before, for example, where it tells me
- 16:22Speaker 1: your rider has reached the restaurant, but the restaurant is
- 16:24Speaker 1: still preparing the order.
- 16:25Speaker 1: Or uh
- 16:26Speaker 2: yeah, no, no, but, but I'm going one step further. Look, the,
- 16:30Speaker 2: the apps have real-time information second by second, OK? Instead
- 16:35Speaker 2: of once a rider is taking the order and coming,
- 16:40Speaker 2: it's important for them to convey this message to the
- 16:42Speaker 2: customer because customer satisfaction is what drives the rider behavior
- 16:47Speaker 2: and the platform behavior and even the restaurant.
- 16:50Speaker 2: To push the rider, go do it fast because I
- 16:52Speaker 2: was delayed in making the order, completing the order, so
- 16:56Speaker 2: you should deliver it fast. The rider should say, no,
- 16:58Speaker 2: this is not my job.
- 17:00Speaker 1: Oh, I have so many issues. That's, that's a whole
- 17:03Speaker 1: different talkback that I'll do on, on, on, on the
- 17:06Speaker 1: things that you're locked into sometimes when you make an
- 17:09Speaker 1: order through one of these apps and trying to get
- 17:11Speaker 1: some kind of customer service is often uh full of
- 17:15Speaker 1: a discussion with a very annoying chatbot.
- 17:19Speaker 1: Um, final question, yeah, you can't get out of it.
- 17:23Speaker 1: You can place the order but you can't cancel it. Um,
- 17:27Speaker 1: final point, do you think platforms need to
- 17:31Speaker 1: Audit their incentive algorithms for safety hazards.
- 17:35Speaker 2: Yes, I mean, look, this is the key to me. They,
- 17:39Speaker 2: this platform economy.
- 17:41Speaker 2: Has some of the best data and they cannot make
- 17:46Speaker 2: these algorithms and then just say, OK, this is done,
- 17:49Speaker 2: we'll be revisited in 10 number of years. They should
- 17:52Speaker 2: be with AI. They should be updating and auditing this
- 17:56Speaker 2: and figuring out how we can continuously improve that based
- 18:00Speaker 2: on the situation, based on where the order is coming from,
- 18:03Speaker 2: why don't this kind of.
- 18:05Speaker 2: Happen based on this delivery time from this restaurant or
- 18:09Speaker 2: for this area of customers. Clearly there is something else
- 18:13Speaker 2: going on and how we can change our algorithm
- 18:15Speaker 1: there. We live in a dynamic age of such granular data.
- 18:20Speaker 1: You're absolutely right, it can be utilized. Professor Sumida Gawal,
- 18:23Speaker 1: thank you so much for your time. He's professor of finance,
- 18:25Speaker 1: real estate and Economics at NUS Business School. I'm going
- 18:27Speaker 1: to squeeze in a quick WhatsApp from one of my
- 18:29Speaker 1: listeners who said,
- 18:30Speaker 1: Um, one of my relatives does this job. When they started,
- 18:34Speaker 1: they thought it was just a way to fill free time,
- 18:36Speaker 1: but then they found themselves doing the job for many
- 18:39Speaker 1: years now. That reminds me, thank you for the WhatsApp,
- 18:41Speaker 1: that reminds me of an article that I read in
- 18:42Speaker 1: preparation for this, which kind of talked about how some
- 18:45Speaker 1: people felt trapped in the job role. I wish you
- 18:49Speaker 1: had a chance to bring it up on today's discussion,
- 18:50Speaker 1: but thank you for that message and being a part
- 18:52Speaker 1: of Topa.