Popular in Professional

Culture and Society

The City as a Growth Machine: Toward a Political Economy of Place

New York University Harvey Moltoch

Education

The Role of Education in Reducing Racial Inequality

University of Southern California Pedro A. Noguera

Law

Diversity and Social Mobility in the UK legal profession

Law Society I. Stephanie Boyce

Internet Law

The Law of AI for Good

University of San Diego Orly Lobel

Pharmacy

Cytotoxic Activity and Anti-Migratory Effect of Berberine-Phytantriol Liquid Crystalline on Non-Small-Cell Lung Cancer

University of Technology Sydney Kamal Dua

Human Rights Law

Transforming Restorative Justice for Transitional Settings

University of Nottingham Kerry Clamp

Education

The Impact of Ride-Hailing Services on Congestion

View more insights in Professional - Business - Operational Research

View more figures from Carnegie Mellon University

View the page of Rahul Telang

Read the Study

Early research shows ride-hailing services’ global development and benefits. Rahul Telang highlights that policy scrutiny is expanding due to evidence of their harmful externalities.

Image courtesy of interviewee. January 16, 2024

Generating Workspace

Export as PDF

This video video discusses the growing popularity and impact of ride-hailing services like Uber and Ola globally. Over the past decade, these services have become the standard for taxis, prompting questions about their economic effects, employment opportunities, and potential negative consequences such as pollution and traffic congestion.

A key focus is on the congestion caused by ride-hailing services. The analysis examines the impact of a driver strike in India, where Uber and Ola services were temporarily halted, leading to a 10-14% reduction in congestion. This video video finding indicates that ride-hailing services contribute to increased traffic congestion, especially during peak hours.

This video video outlines mechanisms by which ride-hailing services exacerbate congestion, including "deadheading," where drivers drive without passengers while seeking fares, and the effect on route choices. When ride-hailing services are unavailable, lesser-used, smaller routes become more attractive due to reduced congestion, suggesting that these services could, paradoxically, lead to less efficient travel routes.

Furthermore, when ride-hailing services are available, they may deter citizens from using public transport options, shifting them back to these alternatives when ride-hailing is not an option. The discussion emphasizes the complexity of assessing their impact on urban transportation and offers insights into possible policy considerations to mitigate congestion caused by ride-hailing services.

Language:

Export as PDF

X-TIMESTAMP-MAP:LOCAL=00:00:00.000,MPEGTS=900000

00:00.015 --> 00:02.885
Think most people are familiar that the right hailing,

00:03.135 --> 00:05.645
right hailing services have become really popular

00:06.305 --> 00:09.085
and popular, not just in the us the world over.

00:09.585 --> 00:11.485
So if you look up any newspaper

00:11.705 --> 00:13.885
or in, in your experience, you know

00:13.885 --> 00:16.845
that the right hailing services have become the de factor

00:17.325 --> 00:20.485
standard for basically taking taxis.

00:21.185 --> 00:23.325
So then the question comes in that, you know,

00:23.345 --> 00:25.365
the ride healing services have become

00:25.425 --> 00:27.525
so popular over the last 10 years.

00:28.515 --> 00:30.965
What has been the impact of these services?

00:31.705 --> 00:34.365
And when I say impact, you know, the impact is very broad.

00:34.585 --> 00:37.005
You know, what is the economic benefit?

00:37.435 --> 00:40.805
What is the employment opportunities?

00:41.115 --> 00:43.445
What has been the impact on the innovation,

00:43.785 --> 00:45.845
had the new technology being developed?

00:46.145 --> 00:48.645
But then the question also comes in that, you know,

00:48.645 --> 00:50.005
does it lead to pollution?

00:50.435 --> 00:51.765
Does it lead to congestion?

00:51.835 --> 00:53.765
What are potentially the social cost

00:53.785 --> 00:55.205
of these six things happening?

00:55.425 --> 00:58.445
So, you know, one of those thing, a new service,

00:58.925 --> 01:01.485
a new technology has become very prevalent.

01:02.235 --> 01:05.165
Then the question obviously people want to ask is that

01:05.165 --> 01:08.685
what does the technology technology does to

01:09.565 --> 01:11.485
whatever context people want to look at?

01:11.625 --> 01:14.445
So congestion is, you know, one very important domain.

01:15.065 --> 01:18.205
So that is really the, the context of how this,

01:18.665 --> 01:20.405
how the whole thing basically evolved.

01:20.675 --> 01:23.405
This whole idea came in because Uber

01:23.625 --> 01:26.525
and Ola, they were the, one of the more popular

01:27.635 --> 01:29.325
ride hailing services in India.

01:30.565 --> 01:35.125
I think in 2017 or 18 or 19, you know, the OLA

01:35.265 --> 01:38.325
and the Uber driver said, okay, we are gonna have a strike.

01:38.985 --> 01:42.925
So they said, okay, I will take my, you know, services off

01:43.685 --> 01:47.325
Delhi Roads or Mumbai Roads or Bangalore roads.

01:47.865 --> 01:49.045
So the question came in that

01:49.215 --> 01:52.005
after the services basically stopped stop,

01:53.005 --> 01:54.175
what impact did it have?

01:54.235 --> 01:56.935
So at that time we said that what, you know, try

01:56.935 --> 02:00.215
to look at congestion that do ride hailing services

02:00.915 --> 02:03.215
or removal of ride hailing services.

02:03.405 --> 02:05.335
Does it lead to any change in congestion?

02:05.475 --> 02:06.535
So that was the idea.

02:07.315 --> 02:09.855
So the focus was that, okay, if you want

02:09.855 --> 02:13.775
to answer this question, what kind of data do we need?

02:14.035 --> 02:15.055
Can we get the data?

02:15.395 --> 02:18.495
Can we write down the good empirical specification

02:18.595 --> 02:20.695
to be able to estimate the, the question?

02:21.115 --> 02:24.695
So that was the context in which we basically said, okay,

02:24.695 --> 02:26.095
this is something we wanna look at.

02:26.445 --> 02:30.255
Then we went about getting the data, analyzing the data,

02:30.365 --> 02:32.495
writing the paper, and so on and so forth.

02:33.035 --> 02:34.535
So the main result, we found

02:34.565 --> 02:38.295
that when the ride healing services was basically went off

02:38.295 --> 02:42.345
the road, the congestion actually went down

02:42.725 --> 02:44.265
by about 10 to 11%.

02:44.775 --> 02:47.385
That basically said that the ride healing

02:47.945 --> 02:49.665
services are causing congestion.

02:50.135 --> 02:51.745
That is when the Uber

02:51.965 --> 02:54.825
and O are applying in the city,

02:55.685 --> 02:58.865
it actually raise at the congestion. And again, you

02:58.865 --> 03:00.165
Know, congestion means what?

03:00.225 --> 03:01.445
So we looked at the,

03:01.445 --> 03:05.165
because we were able to measure the traffic at hourly time,

03:05.665 --> 03:08.765
so we could classify that, is it the peak hour?

03:09.065 --> 03:10.285
Is the non peak hour?

03:11.265 --> 03:15.045
Is it like a very heavily traffic road versus not a very

03:15.045 --> 03:16.205
heavily traffic road?

03:16.625 --> 03:18.485
So we were able to classify that

03:18.825 --> 03:22.165
and basically we found that in the peak hour, you know,

03:22.405 --> 03:24.325
whatever the time when the traffic is the highest,

03:25.185 --> 03:27.325
that's when the congestion actually,

03:27.415 --> 03:29.605
after these services went off the road,

03:30.025 --> 03:33.365
the congestion went down anywhere between 10 to 14%.

03:34.485 --> 03:37.065
The issue that basically become very popular,

03:37.975 --> 03:41.065
that does the right hearing services cost congestion,

03:41.405 --> 03:43.705
you know, that question is also very unclear.

03:43.805 --> 03:45.745
You know, sometime it is a political question

03:46.515 --> 03:49.185
where these companies don't want to be classified

03:49.205 --> 03:50.305
as causing congestion.

03:50.895 --> 03:53.305
Then what do you do if the cost congestion?

03:53.535 --> 03:55.105
What can the policymaker do?

03:55.165 --> 03:58.585
You know, you cannot ban Uber, you cannot ban Ola.

03:58.775 --> 04:01.425
They're generating so much value to the customer.

04:01.925 --> 04:04.345
So then the first question basically become,

04:04.805 --> 04:08.785
so our finding our aim at some level was to actually show

04:08.895 --> 04:11.625
that, hey, these services is causing

04:11.765 --> 04:13.385
so much congestion in the city.

04:14.255 --> 04:16.345
Then the question is that, okay, what do you do with that?

04:16.775 --> 04:18.185
Then, you know, then you can think

04:18.185 --> 04:20.545
of all the potential mechanism that one can do

04:20.605 --> 04:23.545
and remember what we can only suggest,

04:23.685 --> 04:24.985
we can't actually do it.

04:25.405 --> 04:27.625
So we kind of, you know, our study at some level.

04:27.765 --> 04:30.665
So one of the things that we try to highlight

04:30.695 --> 04:32.745
that why are they causing congestion?

04:34.135 --> 04:37.635
So, you know, one can think of this as a very academic world

04:37.645 --> 04:39.795
where basically you can say that, you know what,

04:40.215 --> 04:43.515
if the ride hailing service become available to you,

04:44.775 --> 04:46.875
you'll not use your personal car

04:46.935 --> 04:48.515
but actually use ride hailing service.

04:49.415 --> 04:51.035
So then the thought will be that, hey,

04:51.215 --> 04:52.755
why should the congestion go up?

04:52.755 --> 04:55.755
Because if there was no ride healing services,

04:56.135 --> 04:57.515
you would've used your own car.

04:57.895 --> 04:59.355
So congestion would not have changed.

04:59.415 --> 05:01.195
So then, you know, we show that no,

05:01.255 --> 05:02.955
the congestion is actually going up.

05:03.295 --> 05:05.355
So then the question is that why is it going up?

05:05.695 --> 05:07.635
So one of the idea that we discuss

05:07.735 --> 05:10.915
and we kind of show what the, what the,

05:10.915 --> 05:13.995
what the ride healing services do is what is called the,

05:14.495 --> 05:15.515
you know, dead headache.

05:15.515 --> 05:17.835
That is dead headache that the driver

05:19.015 --> 05:22.515
is driving on the road without the passenger

05:22.825 --> 05:25.715
because they're actually trying to find somebody else

05:25.815 --> 05:27.235
or trying to find a customer.

05:27.695 --> 05:29.515
So the, the ride on the road

05:29.735 --> 05:33.275
and the cost congestion, so that's a me one

05:33.275 --> 05:34.435
of the mechanisms

05:34.435 --> 05:37.035
through which the congestion basically goes up

05:37.535 --> 05:40.435
and we try to actually show in our data that hey,

05:40.905 --> 05:43.635
ride hailing services actually causing congestion.

05:44.255 --> 05:47.235
The other element that we actually show that, you know,

05:47.235 --> 05:50.075
when the ride healing services basically go away,

05:52.045 --> 05:56.645
some rules, they become actually less congested.

05:57.105 --> 05:58.125
And what that means is

05:58.125 --> 06:01.725
that when they become less congested, people are willing

06:01.825 --> 06:04.525
to take this route because they're actually smaller.

06:04.825 --> 06:06.845
So you are actually traveling less,

06:07.465 --> 06:09.685
but the congestion is less so you would take it.

06:09.945 --> 06:12.645
But when the ride healing services are in operation,

06:13.275 --> 06:16.445
even though those routes are smaller, smaller,

06:16.445 --> 06:18.085
because the congestion is so high,

06:18.415 --> 06:20.485
you'd rather take a longer route.

06:20.985 --> 06:22.045
So we show that, you know,

06:22.045 --> 06:24.045
ride Healing service has caused this as well.

06:24.065 --> 06:26.885
So, you know, there's various mechanism we were able

06:26.885 --> 06:29.125
to highlight that what happened.

06:29.345 --> 06:31.805
You know, the other element is that, you know, the public

06:32.445 --> 06:34.565
transport, you know, when there's a ride hailing service

06:34.595 --> 06:36.485
available, people say,

06:36.645 --> 06:39.365
I will take ride hailing services versus taking

06:39.525 --> 06:40.605
a train or a metro.

06:41.435 --> 06:43.965
When the ride hailing services actually go away,

06:44.475 --> 06:46.605
then people say, okay, I'm gonna take a Subway

06:46.945 --> 06:48.125
or I'm gonna take a Metro.

06:48.545 --> 06:52.925
So we were able to identify all these potential causes,

06:52.995 --> 06:54.925
poor potential congestion,

06:55.345 --> 06:57.245
and then the question is that, what can we do?

06:57.385 --> 07:00.365
And then one can only speculate what can we do?

07:00.385 --> 07:02.725
But really the goal was to be able to say

07:02.725 --> 07:03.765
that this is happening

07:03.785 --> 07:05.685
and at least identify the mechanism

07:05.785 --> 07:07.125
that's causing the congestion.

00:00 - Think most people are familiar that the right hailing, right hailing services have become really popular and popular, not just in the us the world over. So if you look up any newspaper or in, in your experience, you know that the right hailing services have become the de factor

00:17 - standard for basically taking taxis. So then the question comes in that, you know, the ride healing services have become so popular over the last 10 years. What has been the impact of these services? And when I say impact, you know, the impact is very broad. You know, what is the economic benefit? What is the employment opportunities? What has been the impact on the innovation, had the new technology being developed?

00:46 - But then the question also comes in that, you know, does it lead to pollution? Does it lead to congestion? What are potentially the social cost of these six things happening? So, you know, one of those thing, a new service, a new technology has become very prevalent. Then the question obviously people want to ask is that what does the technology technology does to whatever context people want to look at?

01:11 - So congestion is, you know, one very important domain. So that is really the, the context of how this, how the whole thing basically evolved. This whole idea came in because Uber and Ola, they were the, one of the more popular ride hailing services in India. I think in 2017 or 18 or 19, you know, the OLA and the Uber driver said, okay, we are gonna have a strike. So they said, okay, I will take my, you know, services off Delhi Roads or Mumbai Roads or Bangalore roads.

01:47 - So the question came in that after the services basically stopped stop, what impact did it have? So at that time we said that what, you know, try to look at congestion that do ride hailing services or removal of ride hailing services. Does it lead to any change in congestion? So that was the idea. So the focus was that, okay, if you want to answer this question, what kind of data do we need?

02:14 - Can we get the data? Can we write down the good empirical specification to be able to estimate the, the question? So that was the context in which we basically said, okay, this is something we wanna look at. Then we went about getting the data, analyzing the data, writing the paper, and so on and so forth. So the main result, we found that when the ride healing services was basically went off the road, the congestion actually went down

02:42 - by about 10 to 11%. That basically said that the ride healing services are causing congestion. That is when the Uber and O are applying in the city, it actually raise at the congestion. And again, you Know, congestion means what? So we looked at the, because we were able to measure the traffic at hourly time, so we could classify that, is it the peak hour?

03:09 - Is the non peak hour? Is it like a very heavily traffic road versus not a very heavily traffic road? So we were able to classify that and basically we found that in the peak hour, you know, whatever the time when the traffic is the highest, that's when the congestion actually, after these services went off the road, the congestion went down anywhere between 10 to 14%. The issue that basically become very popular,

03:37 - that does the right hearing services cost congestion, you know, that question is also very unclear. You know, sometime it is a political question where these companies don't want to be classified as causing congestion. Then what do you do if the cost congestion? What can the policymaker do? You know, you cannot ban Uber, you cannot ban Ola. They're generating so much value to the customer. So then the first question basically become,

04:04 - so our finding our aim at some level was to actually show that, hey, these services is causing so much congestion in the city. Then the question is that, okay, what do you do with that? Then, you know, then you can think of all the potential mechanism that one can do and remember what we can only suggest, we can't actually do it. So we kind of, you know, our study at some level. So one of the things that we try to highlight

04:30 - that why are they causing congestion? So, you know, one can think of this as a very academic world where basically you can say that, you know what, if the ride hailing service become available to you, you'll not use your personal car but actually use ride hailing service. So then the thought will be that, hey, why should the congestion go up? Because if there was no ride healing services, you would've used your own car.

04:57 - So congestion would not have changed. So then, you know, we show that no, the congestion is actually going up. So then the question is that why is it going up? So one of the idea that we discuss and we kind of show what the, what the, what the ride healing services do is what is called the, you know, dead headache. That is dead headache that the driver is driving on the road without the passenger

05:22 - because they're actually trying to find somebody else or trying to find a customer. So the, the ride on the road and the cost congestion, so that's a me one of the mechanisms through which the congestion basically goes up and we try to actually show in our data that hey, ride hailing services actually causing congestion. The other element that we actually show that, you know, when the ride healing services basically go away,

05:52 - some rules, they become actually less congested. And what that means is that when they become less congested, people are willing to take this route because they're actually smaller. So you are actually traveling less, but the congestion is less so you would take it. But when the ride healing services are in operation, even though those routes are smaller, smaller, because the congestion is so high, you'd rather take a longer route.

06:20 - So we show that, you know, ride Healing service has caused this as well. So, you know, there's various mechanism we were able to highlight that what happened. You know, the other element is that, you know, the public transport, you know, when there's a ride hailing service available, people say, I will take ride hailing services versus taking a train or a metro. When the ride hailing services actually go away,

06:44 - then people say, okay, I'm gonna take a Subway or I'm gonna take a Metro. So we were able to identify all these potential causes, poor potential congestion, and then the question is that, what can we do? And then one can only speculate what can we do? But really the goal was to be able to say that this is happening and at least identify the mechanism that's causing the congestion.

Faculti AI can make mistakes