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

A social network transmission model of investor relations

View more insights in Professional - Accounting

View more figures from University of California Los Angeles

View the page of Judson Caskey

Read the Study

Judson Caskey discusses how an investor relations strategy can maximize information flow through a firm’s social network.

Image courtesy of interviewee. March 27, 2024

Generating Workspace

Export as PDF

This video video discusses the role of investor relations (IR) in facilitating the flow of information to the investment community. The focus is on how IR can enhance understanding among investors through effective communication strategies. It highlights the importance of not only sharing information but ensuring that investors process it, which can occur through social transmission—whereby information is communicated among individuals rather than solely from official sources.

The study examines how targeting specific individuals or groups, especially those with many connections (hubs), can speed up the dissemination of information through networks. If IR efforts can initially reach only a small fraction of the investor population, targeting these hubs proves beneficial. Conversely, if the initial communication can reach a large number of people directly, a more general broadcast approach may be more effective.

Furthermore, the text explains how the structure of networks, characterized by heterogeneous individuals, influences the flow of information. It models this information spread using concepts from disease propagation, as both processes involve person-to-person transmission.

The implications of information flow on bid-ask spreads in trading are also explored. Information spreads create a hump pattern in bid-ask spreads: initially low due to a few informed traders, then rising as more become informed, followed by a decline as information is fully absorbed. Rapid information flow leads to sharp initial increases in spreads followed by quick decreases, while slower information flow results in more gradual changes.

In conclusion, effective IR strategies differ depending on the ability to reach specific groups versus a broader audience, with implications for how quickly information spreads and the resulting market effects.

Language:

Export as PDF

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

00:00.525 --> 00:02.965
Investor relations groups have, several objectives.

00:02.965 --> 00:05.965
The one that we focus on is helping the investment community

00:06.095 --> 00:08.085
understand the company.

00:08.705 --> 00:12.045
And, that understanding depends not only on providing

00:12.045 --> 00:13.645
information,

00:13.645 --> 00:16.365
but also making sure that investors actually observe

00:16.465 --> 00:18.365
and process the information.

00:19.065 --> 00:22.405
Social transmission refers to person to person communication

00:23.105 --> 00:24.405
and we examine the role

00:24.405 --> 00:27.005
that it plays in disseminating information, you know,

00:27.005 --> 00:28.165
from investor relations

00:28.165 --> 00:30.005
or from companies to the group of investors.

00:30.345 --> 00:34.145
Um, social transmission plays a key role.

00:34.625 --> 00:37.225
'cause just because I investor relations makes information

00:37.225 --> 00:39.785
available, that doesn't mean investors process it.

00:39.805 --> 00:42.665
For example, many investors may not be aware

00:42.665 --> 00:45.065
of a company press release because they're busy working

00:45.205 --> 00:46.665
or they're having lunch or something,

00:47.565 --> 00:50.225
and such people may learn about the existence

00:50.225 --> 00:52.185
of the press release from their contacts,

00:52.765 --> 00:54.985
and then they in turn might make others aware of it.

00:55.645 --> 00:58.745
Uh, furthermore, social networks can spread the

00:58.745 --> 01:00.625
interpretation of information just

01:00.625 --> 01:02.505
as they can spread the awareness of it.

01:03.085 --> 01:05.425
So they might, people might rely on their contacts not only

01:05.425 --> 01:07.705
to be aware that a company's released information,

01:07.845 --> 01:09.625
but to know what to make of it.

01:10.235 --> 01:11.505
First, to give a backdrop.

01:11.685 --> 01:15.145
Um, so you understand like what's relevant for our study.

01:15.205 --> 01:18.145
So we're looking at, how investor relations,

01:18.505 --> 01:21.025
communication strategy can impact the flow of information.

01:21.085 --> 01:24.905
And specifically we look at the role play by two aspects

01:24.905 --> 01:25.905
of the communication.

01:25.965 --> 01:29.665
So the first is targeting specific individuals

01:29.665 --> 01:33.105
or groups by, for example, directly contacting them

01:33.165 --> 01:35.825
to make sure that they're aware of a company communication.

01:36.325 --> 01:39.265
And then the second aspect is how large of a group

01:39.865 --> 01:42.345
investor relations can initially reach by, for example,

01:42.345 --> 01:45.785
coordinating with, media outlets versus,

01:45.785 --> 01:47.425
simply making information available

01:47.485 --> 01:48.585
on the corporate website.

01:49.485 --> 01:53.895
So given the focus of our study, it's important

01:54.475 --> 01:57.495
to think about, settings

01:57.495 --> 01:59.695
where individuals have some heterogeneity because

01:59.725 --> 02:01.575
otherwise, you know, there would be no reason

02:01.595 --> 02:04.575
to target any particular group of individuals.

02:04.575 --> 02:07.095
So with that in mind, so the prior literature and networks

02:07.555 --> 02:10.975
and trade has not really examined information flow in

02:11.015 --> 02:13.455
networks with heterogeneous individuals.

02:13.955 --> 02:18.185
Uh, so a couple of papers have looked at, settings

02:18.185 --> 02:19.825
where investors are in networks,

02:20.395 --> 02:23.145
where there are heterogene heterogeneous individuals,

02:23.525 --> 02:25.385
but they use that to characterize

02:25.765 --> 02:27.345
how correlated is information,

02:27.345 --> 02:30.745
but not the flow of information over time.

02:31.745 --> 02:34.645
Uh, so that aspect, is new for our paper.

02:34.645 --> 02:37.285
And then a couple of other studies have looked at the flow

02:37.285 --> 02:41.365
of information over time, but not in a network setting.

02:41.625 --> 02:46.085
So in these studies, they, imagine investors, you know,

02:46.085 --> 02:47.925
think about like being particles in a sense,

02:47.925 --> 02:49.885
maybe like particles that might bounce around

02:49.985 --> 02:51.765
and, you know, trade information,

02:51.825 --> 02:54.725
but there's no sense in which one is different from another.

02:54.835 --> 02:57.245
Some just kind of get lucky and bump into more people

02:57.245 --> 02:58.285
and get more information.

02:59.265 --> 03:01.845
So that's kind of, so the prior literature look

03:02.425 --> 03:06.205
hasn't really touched on this idea of how

03:07.155 --> 03:09.525
does the structure of the network, how people are linked

03:09.585 --> 03:12.405
to each other, impact how information flows.

03:12.705 --> 03:14.965
And our study looks at how does the structure

03:15.025 --> 03:17.725
of initial communications impact

03:17.865 --> 03:20.165
how information, flows.

03:20.265 --> 03:23.965
And then the goal is to inform a population of investors

03:24.035 --> 03:25.445
that we represent by a network.

03:25.625 --> 03:26.805
And then we also examine

03:26.945 --> 03:29.725
how the information flows affect bid a spreads.

03:30.825 --> 03:33.645
So to model the information flow we borrow from the physics

03:34.005 --> 03:35.965
literature and networks, and specifically on

03:36.125 --> 03:37.205
disease propagation.

03:38.065 --> 03:39.885
Um, it sounds may sound kinda like odd

03:39.885 --> 03:41.645
that we're looking at disease propagation,

03:41.705 --> 03:44.325
but it propagates from person to person just

03:44.385 --> 03:45.805
as you know, information might.

03:46.345 --> 03:50.625
Um, so in our setting, some people in the network serve

03:50.645 --> 03:53.185
as hubs in the sense that they're connected to many people.

03:54.335 --> 03:59.155
And when it's beneficial to target communication, it's going

03:59.155 --> 04:01.475
to be beneficial to target these hubs

04:01.505 --> 04:03.515
that will speed the information flow.

04:04.715 --> 04:08.175
And our main finding is that targeting specific groups

04:08.235 --> 04:11.655
of highly connected individuals speeds the information flow,

04:12.315 --> 04:15.215
but only when the,

04:15.725 --> 04:18.415
initial communication can reach only a

04:18.415 --> 04:19.535
small fraction of people.

04:20.315 --> 04:23.095
If investor relations can initially reach a large group,

04:23.675 --> 04:26.895
the overall group of investors processes the information

04:26.895 --> 04:30.135
more quickly if investor relations makes no attempt

04:30.135 --> 04:33.455
to target, but instead has a general broadcast such as

04:33.455 --> 04:34.535
through media outlets.

04:36.355 --> 04:38.575
So for the intuition behind this,

04:38.995 --> 04:41.695
so I think it's quite intuitive why targeting highly

04:41.695 --> 04:44.175
connected individuals would help information flow.

04:44.375 --> 04:47.175
A person with a thousand connections can quickly reach a

04:47.335 --> 04:51.295
thousand people, then those a thousand people then reach out

04:51.415 --> 04:53.335
to their own contacts and so on.

04:53.605 --> 04:55.015
Whereas reaching a person

04:55.015 --> 04:57.535
with only a few connections can only quickly

04:57.545 --> 04:58.695
reach a few people.

05:00.125 --> 05:03.305
So this logic breaks down when the initial communication

05:03.415 --> 05:05.905
reaches a large fraction of the population.

05:06.805 --> 05:09.025
And the reason why is highly connected individuals

05:09.315 --> 05:12.865
inherently tend to reside in a network's interior.

05:13.765 --> 05:15.345
And so that targeting will quickly

05:15.545 --> 05:17.105
saturate the network's interior.

05:17.645 --> 05:20.225
And then information slowly flows outward

05:20.645 --> 05:21.985
to the periphery of the network.

05:22.965 --> 05:25.865
And so when the initial communication can reach,

05:26.095 --> 05:28.505
many people, a random broadcast

05:29.375 --> 05:31.865
will inherently reach some people on the periphery

05:31.865 --> 05:35.025
of the network and some people on the interior

05:35.195 --> 05:37.665
where it'll eventually hit these highly connected people.

05:38.875 --> 05:41.895
The information that reaches the interior, you know,

05:41.895 --> 05:44.255
will hit these highly connected individuals

05:44.435 --> 05:45.655
and saturate the interior

05:46.115 --> 05:48.415
and flow outward, whereas the information

05:48.415 --> 05:52.415
that initially hits the periphery will start to inform them

05:52.715 --> 05:53.775
and then flow inward.

05:53.835 --> 05:56.335
So overall, when you can reach a large fraction

05:56.335 --> 05:59.205
of investors, these effects work together

05:59.665 --> 06:03.205
to more quickly inform the overall population than it would

06:03.305 --> 06:06.085
if you had just targeted the highly connected people.

06:06.865 --> 06:11.115
Uh, and so that's the, you know, main insights in terms

06:11.115 --> 06:12.275
of information flow.

06:12.275 --> 06:13.435
And then we also study

06:13.535 --> 06:16.555
how information flow can affect bid as spreads.

06:17.615 --> 06:20.515
So when invest, when information flows across a pool

06:20.515 --> 06:21.835
of investors, it tends

06:21.835 --> 06:24.915
to create a hump shaped pattern in bid as spreads.

06:24.915 --> 06:27.875
So at first, when initi when information is first released,

06:28.585 --> 06:30.275
only a few people have information.

06:30.275 --> 06:33.675
So the overall order flow is not very informative.

06:34.375 --> 06:37.035
Uh, it's most likely driven by uninformed investors

06:37.175 --> 06:38.195
and spreads are low.

06:39.295 --> 06:40.515
As more

06:40.515 --> 06:44.755
and more investors process the information flow, the

06:45.015 --> 06:46.795
and the order flow becomes more

06:46.795 --> 06:48.635
and more likely to be from informed traders

06:48.635 --> 06:52.615
and spreads rise, and then they later on decline

06:52.635 --> 06:55.335
as the prices increasingly a pound impound.

06:55.335 --> 06:57.735
The information, the speed

06:57.755 --> 07:01.215
of this rise in decline process corresponds to

07:01.215 --> 07:03.055
how quickly information flows

07:03.055 --> 07:04.535
through the network of investors.

07:05.515 --> 07:07.255
So when information flows very rapidly,

07:08.035 --> 07:09.695
the spreads will quickly jump

07:10.445 --> 07:12.175
when the information is released,

07:12.175 --> 07:14.895
because the order, the investors quickly become informed,

07:14.895 --> 07:18.135
the order flow as highly informative spreads jump,

07:18.155 --> 07:19.815
but then they precipitously drop.

07:20.925 --> 07:22.855
When the information flows more slowly,

07:23.635 --> 07:27.775
the spreads will slowly rise and then slowly fall.

07:28.195 --> 07:31.055
And the upshot is that fast information flow

07:31.715 --> 07:34.495
causes a temporarily very high bid as spread,

07:34.555 --> 07:36.695
but spreads tend to be lower on average

07:36.965 --> 07:39.575
because that spike is very short-lived.

07:40.355 --> 07:41.615
The main conclusion is

07:41.615 --> 07:44.295
that investor relations communication strategy can

07:44.295 --> 07:46.375
significantly impact the flow of information

07:46.435 --> 07:47.695
and should take into account

07:47.915 --> 07:49.775
how investors interact with each other.

07:50.975 --> 07:52.625
When investor relations wishes

07:52.725 --> 07:55.985
to communicate complex information that is really difficult

07:56.005 --> 07:58.065
to convey to a broad audience,

07:58.125 --> 08:00.105
it should target highly connected people

08:00.725 --> 08:03.545
and then rely on person to meet person communication

08:03.925 --> 08:05.625
to spread that to the broader community.

08:06.645 --> 08:09.665
Uh, when the information can be efficiently broadcast,

08:10.085 --> 08:11.425
the investor relations need

08:11.425 --> 08:13.585
and bother targeting specific groups of people,

08:14.215 --> 08:16.665
they can just broadcast to a large audience

08:17.245 --> 08:20.505
and the information will flow quickly to the entire pool

08:20.505 --> 08:22.545
of investors regardless of whether

08:22.605 --> 08:24.745
or not, highly connected individuals

08:24.745 --> 08:26.825
or amongst the first to become aware of it.

08:28.285 --> 08:29.305
And then last,

08:29.305 --> 08:32.465
when information flows quickly across the pool of investors,

08:32.685 --> 08:35.025
we predict that spreads will sharply increase

08:35.565 --> 08:38.385
and then decrease after the release of information.

08:39.125 --> 08:41.265
So you can imagine a company has a press release,

08:41.885 --> 08:43.105
big jump in spreads,

08:43.105 --> 08:46.105
but then a few minutes later, maybe it drops back to normal.

08:46.415 --> 08:48.505
Whereas when information flows slowly,

08:49.125 --> 08:52.345
the spreads may gradually rise and then decline.

08:52.405 --> 08:54.825
And so while the highest spread might not be as

08:54.825 --> 08:56.605
High as with a quick information flow,

08:57.145 --> 08:59.245
the spreads stay higher for longer.

08:59.865 --> 09:01.845
And so the average, bid

09:01.845 --> 09:04.925
as spreads can actually be lower when you have this fast

09:04.925 --> 09:06.645
information flow precisely

09:06.645 --> 09:09.925
because that spike in spreads is, very short-lived.

00:00 - Investor relations groups have, several objectives. The one that we focus on is helping the investment community understand the company. And, that understanding depends not only on providing information, but also making sure that investors actually observe

00:16 - and process the information. Social transmission refers to person to person communication and we examine the role that it plays in disseminating information, you know, from investor relations or from companies to the group of investors. social transmission plays a key role. 'cause just because I investor relations makes information available, that doesn't mean investors process it. For example, many investors may not be aware

00:42 - of a company press release because they're busy working or they're having lunch or something, and such people may learn about the existence of the press release from their contacts, and then they in turn might make others aware of it. furthermore, social networks can spread the interpretation of information just as they can spread the awareness of it. So they might, people might rely on their contacts not only to be aware that a company's released information,

01:07 - but to know what to make of it. First, to give a backdrop. so you understand like what's relevant for our study. So we're looking at, how investor relations, communication strategy can impact the flow of information. And specifically we look at the role play by two aspects of the communication. So the first is targeting specific individuals or groups by, for example, directly contacting them to make sure that they're aware of a company communication.

01:36 - And then the second aspect is how large of a group investor relations can initially reach by, for example, coordinating with, media outlets versus, simply making information available on the corporate website. So given the focus of our study, it's important to think about, settings where individuals have some heterogeneity because otherwise, you know, there would be no reason to target any particular group of individuals.

02:04 - So with that in mind, so the prior literature and networks and trade has not really examined information flow in networks with heterogeneous individuals. so a couple of papers have looked at, settings where investors are in networks, where there are heterogene heterogeneous individuals, but they use that to characterize how correlated is information, but not the flow of information over time. so that aspect, is new for our paper.

02:34 - And then a couple of other studies have looked at the flow of information over time, but not in a network setting. So in these studies, they, imagine investors, you know, think about like being particles in a sense, maybe like particles that might bounce around and, you know, trade information, but there's no sense in which one is different from another. Some just kind of get lucky and bump into more people and get more information. So that's kind of, so the prior literature look

03:02 - hasn't really touched on this idea of how does the structure of the network, how people are linked to each other, impact how information flows. And our study looks at how does the structure of initial communications impact how information, flows. And then the goal is to inform a population of investors that we represent by a network. And then we also examine how the information flows affect bid a spreads.

03:30 - So to model the information flow we borrow from the physics literature and networks, and specifically on disease propagation. it sounds may sound kinda like odd that we're looking at disease propagation, but it propagates from person to person just as you know, information might. so in our setting, some people in the network serve as hubs in the sense that they're connected to many people. And when it's beneficial to target communication, it's going

03:59 - to be beneficial to target these hubs that will speed the information flow. And our main finding is that targeting specific groups of highly connected individuals speeds the information flow, but only when the, initial communication can reach only a small fraction of people. If investor relations can initially reach a large group, the overall group of investors processes the information more quickly if investor relations makes no attempt

04:30 - to target, but instead has a general broadcast such as through media outlets. So for the intuition behind this, so I think it's quite intuitive why targeting highly connected individuals would help information flow. A person with a thousand connections can quickly reach a thousand people, then those a thousand people then reach out to their own contacts and so on. Whereas reaching a person with only a few connections can only quickly

04:57 - reach a few people. So this logic breaks down when the initial communication reaches a large fraction of the population. And the reason why is highly connected individuals inherently tend to reside in a network's interior. And so that targeting will quickly saturate the network's interior. And then information slowly flows outward to the periphery of the network. And so when the initial communication can reach,

05:26 - many people, a random broadcast will inherently reach some people on the periphery of the network and some people on the interior where it'll eventually hit these highly connected people. The information that reaches the interior, you know, will hit these highly connected individuals and saturate the interior and flow outward, whereas the information that initially hits the periphery will start to inform them and then flow inward.

05:53 - So overall, when you can reach a large fraction of investors, these effects work together to more quickly inform the overall population than it would if you had just targeted the highly connected people. and so that's the, you know, main insights in terms of information flow. And then we also study how information flow can affect bid as spreads. So when invest, when information flows across a pool of investors, it tends

06:21 - to create a hump shaped pattern in bid as spreads. So at first, when initi when information is first released, only a few people have information. So the overall order flow is not very informative. it's most likely driven by uninformed investors and spreads are low. As more and more investors process the information flow, the and the order flow becomes more and more likely to be from informed traders

06:48 - and spreads rise, and then they later on decline as the prices increasingly a pound impound. The information, the speed of this rise in decline process corresponds to how quickly information flows through the network of investors. So when information flows very rapidly, the spreads will quickly jump when the information is released, because the order, the investors quickly become informed,

07:14 - the order flow as highly informative spreads jump, but then they precipitously drop. When the information flows more slowly, the spreads will slowly rise and then slowly fall. And the upshot is that fast information flow causes a temporarily very high bid as spread, but spreads tend to be lower on average because that spike is very short-lived. The main conclusion is that investor relations communication strategy can

07:44 - significantly impact the flow of information and should take into account how investors interact with each other. When investor relations wishes to communicate complex information that is really difficult to convey to a broad audience, it should target highly connected people and then rely on person to meet person communication to spread that to the broader community. when the information can be efficiently broadcast,

08:10 - the investor relations need and bother targeting specific groups of people, they can just broadcast to a large audience and the information will flow quickly to the entire pool of investors regardless of whether or not, highly connected individuals or amongst the first to become aware of it. And then last, when information flows quickly across the pool of investors, we predict that spreads will sharply increase

08:35 - and then decrease after the release of information. So you can imagine a company has a press release, big jump in spreads, but then a few minutes later, maybe it drops back to normal. Whereas when information flows slowly, the spreads may gradually rise and then decline. And so while the highest spread might not be as High as with a quick information flow, the spreads stay higher for longer. And so the average, bid

09:01 - as spreads can actually be lower when you have this fast information flow precisely because that spike in spreads is, very short-lived.

Faculti AI can make mistakes